Model Based Design and Performance Analysis of Solar Absorption Cooling and Heating System

Transcription

1 Model Based Design and Performance Analysis of Solar Absorption ooling and Heating System Ming Qu arnegie Mellon University School of Architecture Ph.D. ommittee Prof. Volker Hartkopf, Ph.D. (hair) Prof. David Archer, Ph.D. Prof. Khee Poh Lam, Ph.D.

2

3

4 opyright Declaration I hereby declare that I am the sole author of this thesis. I authorize arnegie Mellon University, Pittsburgh, Pennsylvania to lend this thesis to other institutions or individuals for the purpose of scholarly research. I authorize arnegie Mellon University, Pittsburgh, Pennsylvania to reproduce this thesis by photo copying or by other means, in total or in part, at the request of other institutions or individuals for the purpose of scholarly research. opyright 2008 by Ming Qu i

5 Acknowledgement I wish to express my gratitude to my advisor, Dr. Volker Hartkopf, for his invaluable vision, support, and encouragement. His enthusiasm and inspiration were essential to the success of this research. Surely, I would like to extend my sincere appreciation and profound gratitude to Dr. David Archer who has played a pivotal role in this thesis. He has far exceeded his duty as an advisor. He gave me a deep understanding of mechanical engineering; he taught me how to develop critical thinking and work effectively. He has been an ever-present source of guidance and encouragement throughout my doctoral program. It gives me great pleasure to thank Dr. Khee Poh Lam for providing valuable suggestions and carefully reviewing and constructively critiquing of my work. I owe many thanks to my dear colleague and husband, Hongxi Yin, who gave me continuous support and took care of our babies, Ryan and David, who fill us with joy every day. This thesis is dedicated to my parents in their confidence, their high expectations, and their hearty blessing. ii

6 Model Based Design and Performance Analysis of Solar Absorption ooling and Heating System iii

7 Table of ontents 1 Introduction Background and motivation Solar receivers Absorption ycle Solar collector coupled with absorption chillers urrent studies on solar absorption cooling and heating systems Research objective Research approach The planning of the test system The development of solar collector model The development of annual system performance simulation The installation of the test system The test program and experimental data gathering Data analyses, model validation and simulation evaluation hapter overview Solar absorption cooling and heating test system and program Parabolic trough solar collector Device description Major components and characteristic Absorption chiller Device description Major components and characteristics Solar absorption cooling and heating test system System description The solar collection loop The load loop Instrumentation, control and data acquisition system The test program The results of PTS test at the transient states The results of PTS test at a steady state iv

8 2.4.3 The results of solar absorption cooling / heating daily test The results of solar heating daily test by using heat exchanger The interpretation of PTS performance data The energy balance of the PTS The selection of experimental data PTS performance predicted by statistic tool Discussion of test program Solar collector performance model PTS model assumption Energy balance analysis Heat transfer analysis alculation procedure Solar irradiation absorption Direct normal solar radiation Incident angle and incident modifier End-loss Shadow-loss onclusion Model-based experimental data analysis of PTS Analytical method Model validation Model-based PTS performance analysis Temperature distribution in the receiver pipe Thermal losses PTS efficiency and solar radiation PTS efficiency and incident angle of solar beam PTS efficiency and wind speed PTS efficiency and fluid type PTS efficiency and flow rate PTS efficiency and air in the annular space PTS efficiency and glass envelope Recommendations on the PTS s design Bellow design v

9 4.3.2 Glass cover Diameter of the glass envelope Diameter of the absorber pipe Solar absorption cooling and heating system simulation Model approach Model assumptions Weather Assumptions in the model of solar energy supply system System components and operation controls omponents in the solar heating base-case Operational controls in the solar heating base-case omponents and operation controls in the solar cooling base-case Simulation evaluation Base-case result of solar cooling and heating simulation Building simulation results Solar energy system simulation results Simulation-based design and performance analysis on solar cooling and heating Orientation of PTS Orientation of the PTS for increased, effective solar energy recovery Orientation, tracking limitation, and solar beam irradiation on the PTS Orientation and overall system performance System operation and control onstant-flow or constant-outlet temperature control of the PTS Storage tank requirements The volume of the storage tank Storage used for shifting energy for later use in solar heating Storage used for shifting energy for later use in solar cooling Storage used for preheating Auxiliary heater for preheating in the solar collection loop The length and diameter of collection loop pipe and solar system performance The area of solar collector and storage tank Guidelines for design and operation of solar cooling and heating system vi

10 7 ontributions and areas of future research ontributions Areas of future research Improving the tracking system of the PTS Extending the operational controls of the PTS, the absorption chiller, and the heat recovery exchanger Integrate thermal storage in the cooling/heating system ost model References vii

11 List of Figures Figure 1-1 Simplified system arrangement of solar absorption cooling and heating system... 2 Figure 1-2 Two types of solar collector... 4 Figure 1-3 Electric chiller and absorption chiller... 4 Figure 1-4 Solar collector efficiency and operating temperature required by absorption chiller... 6 Figure 1-5 Research approach schematic chart Figure 1-6 Process and instrumentation diagram of the test solar absorption cooling and heating system Figure 2-1 The PTS s installed on the IW Figure 2-2 Broad BJ16A parabolic trough solar collectors and the receiver tube Figure 2-3 Absorption chiller Figure 2-4 Absorption chiller in cooling cycle Figure 2-5 Absorption chiller in heating cycle Figure 2-6 Broad pump and control package Figure 2-7 Overall solar absorption cooling and heating test system Figure 2-8 Structure of control system...24 Figure 2-9 Interface of the WebTRL Figure 2-10 PTS test diagram at transient state Figure 2-11 Solar absorption cooling /heating system daily test Figure 2-12 Solar heating daily test by using heat exchanger Figure 2-13 Operating temperatures of the PTS test at transient state on 29 March Figure 2-14 Energy flows of the PTS test at the transient state on 29 March Figure 2-15 Operating temperatures of the PTS test at steady state on 22 April Figure 2-16 Energy flows of the PTS test at steady state on 22 April Figure 2-17 Operating temperatures of solar cooling test on 31 July Figure 2-18 Operating temperatures of solar cooling test on 16 July Figure 2-19 ooling capacity of solar cooling system on 31 July Figure 2-20 ooling capacity of solar cooling system on 16 July Figure 2-21 Operating temperatures of solar absorption heating test on 9 March Figure 2-22 Heating capacity of solar absorption heating system on 9 March Figure 2-23 Operating temperatures of HX based solar heating system on 2 March Figure 2-24 Heating apacity of HX based solar heating system on 2 March Figure 2-25 Scatter plot of I*Aa*cos(theta) and m*(po*to-pi*ti) viii

12 Figure 2-26 Scatter plot of average operation temperature and m*(po*to-pi*ti) Figure 3-1 Energy flow in the PTS Figure 3-2 The thermal network Figure 3-3 The connection between the PTSs Figure 3-4 Incident angle of the PTS Figure 3-5 Incident angle modifier and incident angle Figure 3-6 End-loss of the PTS Figure 3-7 The length of the end-loss in the solar field Figure 3-8 Shadow loss from the adjacent solar collector array Figure 4-1 Measured temperature distribution of the glass envelope Figure 4-2 omparison between the measured data and calculation solutions Figure 4-3 Temperature distribution in the receiver pipe Figure 4-4 Thermal losses through the receiver pipe Figure 4-5 PTS s efficiency and direct normal solar radiation at 0 incident angle Figure 4-6 PTS s efficiency and incident angle Figure 4-7 PTS s efficiency and direct normal solar radiation at 15 incident angle Figure 4-8 PTS s efficiency and wind speed Figure 4-9 PTS s efficiency and fluid type Figure 4-10 PTS s efficiency & flow rate Figure 4-11 PTS s efficiency and air in the annual space Figure 4-12 Thermal losses with Sun or No-sun Figure 4-13 PTS s efficiency and glass cover Figure 4-14 New bellow design from SOLEL Figure 5-1 Information flow of TRNSYS simulation Figure 5-2 Monthly average dry bulb temperature of Pittsburgh Figure 5-3 Direct normal solar radiation in Pittsburgh Figure 5-4 Daily average solar radiation throughout a year in Pittsburgh Figure 5-5 TRNSYS information flow diagram of solar cooling base case Figure 5-6 TRNSYS information flow diagram of solar heating base case Figure 5-7 Boiling temperature and pressure of aqueous propylene glycol solutions Figure 5-8 BROAD PTS tracking range Figure 5-9 Operation temperature comparison between solar heating evaluation simulation and experiment ix

13 Figure 5-10 Energy flow comparison between solar heating evaluation simulation and experiment Figure 5-11 Operating temperature comparison between cooling evaluation simulation and experiement Figure 5-12 Energy flow comparison between solar cooling evaluation simulation and experiment Figure 5-13 IW building heating and cooling load estimated by building simulation Figure 5-14 Useful solar energy and IW sensible heating load in January Figure 5-15 Useful solar energy and IW sensible cooling on 30 December Figure 5-16 Useful solar energy, cooling load and energy provided by chiller in August Figure 5-17 Useful solar energy, cooling load and energy provided by chiller on 09 August Figure 6-1 Twelve orientations in the simulation Figure 6-2 Orientation and solar beam irradiation on a PTS on 10 June in Pittsburgh Figure 6-3 Orientation and solar beam irradiation on a PTS on 2 December in Pittsburgh Figure 6-4 Orientation and solar beam irradiation on a PTS in summer of Pittsburgh Figure 6-5 Orientation and solar beam irradiation on a PTS in winter of Pittsburgh Figure 6-6 Annual solar beam irradiation on PTS with different orientations in Pittsburgh Figure 6-7 Tracking angle and orientation of the PTS on 21Jun Figure 6-8 Tracking angle and orientation of the PTS on 21 Dec Figure 6-9 Solar beam irradiation and orientation of the PTS in Pittsburgh Figure 6-10 System performance comparison of alternate controls on 9 August Figure 6-11 System operating temperature comparison of alternate controls on 9 August Figure 6-12 Trnsys information flow diagram of solar heating system with storage Figure 6-13 Solar energy collected, heating load, and energy provided on 14, 15 November Figure 6-14 Trnsys information flow diagram of solar cooling system with storage for shifting energy Figure 6-15 Operating temperature of solar cooling system with and without storage on 9 August Figure 6-16 Trnsys information flow diagram for solar cooling with storage for preheating Figure 6-17 Effect of a heater on energy flow for solar cooling on 09 August Figure 6-18 Effect of a heater on operating temperature for solar cooling on 9 August Figure 6-19 Solar fraction and pipe size under two control strategies Figure 6-20 System energy performance and pipe size on 9 August x

14 Figure 6-21 Operating temperature and pipe size on 9 August Figure 6-22 Effect of PTS area and storage volume on the solar fraction in IW cooling and heating Figure 6-23 Idealized IW solar cooling/heating system performance and system sensitivity analysis xi

15 List of Tables Table 2-1 Specifications of the parabolic trough solar collector installed Table 2-2 Specifications of the absorption chiller installed Table 2-3 Instrumentation of IW solar cooling and heating system Table 2-4 Four types of tests conducted Table 2-5 Heat capacity of the solar collection loop Table 2-6 Operating condition in the PTS performance tests Table 2-7 Heating system performance comparison between HX based and absorption chiller based Table 3-1 Heat transfer correlations used in the PTS model Table 3-2 Parameters and values used in the PTS model Table 4-1 Glass temperature measurements in the test Table 4-2 omparison between measured values and model calculations Table 5-1 ontrol mode in the base-case simulation of solar heating system Table 5-2 ontrol mode in the base-case simulation of solar cooling system Table 5-3 System performance estimated by IW solar heating system base-case simulation Table 5-4 System performance estimated by IW solar cooling system base-case simulation Table 6-1 Effect of PTS s orientation on overall system performance Table 6-2 Effect of flow controls on overall system performance Table 6-3 Effect of storage volume on solar heating system performance Table 6-4 Effect of storage volume on solar cooling system performance Table 6-5 Effect of Preheat storage tank volume on solar cooling performance Table 6-6 Effect of collection loop volume on solar heating system performance Table 6-7 Effect of collection loop volume on solar cooling system performance xii

16 Abstract The work presented in this thesis deals with the question of how solar energy might most effectively and efficiently be used in supplying energy for the operation of a building. The approach to dealing with this question has involved a specific building space, arnegie Mellon s Intelligent Workplace; a specific solar system, parabolic trough solar thermal receivers, Parabolic Trough Solar ollector s; and a specific building energy use, space cooling and heating. The work has involved the design, installation, and test of a system incorporating PTS s, an absorption chiller, a heat recovery exchanger, auxiliary equipment, instrumentation and controls. Mathematical models based on fundamental scientific and engineering principles have been developed and programmed for both the PTS s and the overall IW cooling and heating system, These models have been improved and validated through comparisons of predicted and measured PTS and IW cooling and heating system performance. The work reported in this thesis has developed suggestions and methods for the effective design and evaluation of PTS s and also for the optimized design and operation of solar absorption cooling and heating systems, so that the system is able to reduce building energy consumption, and achieve environmental benefits in the operation of buildings by the use of renewable, solar energy. xiii

17 1 Introduction HVA systems are the major users of electricity in commercial buildings. In the United States, commercial air conditioning makes up 40% of the summer time peak electrical demand (Kulkarni 1994). In recent years, the increasing power demand of building HVA system, and the increasing costs of energy have caused people to seek alternative cheaper, renewable energy sources for building cooling and heating. In addition, the environmental issues such as global warming, ozone depletion, and energy conservation, are other important factors, impelling people to look for space cooling and heating without involving gas or electricity. The use of solar energy for building cooling and heating can potentially provide the solution to these economic and environmental problems. An important motivation for research and development in solar cooling is the coincidence of comfort cooling demand and the availability of solar radiation. Building cooling systems which use solar thermal energy can make use of absorption cycles, desiccant cycles, on solar-mechanical processes. ompared to solar desiccant cycles and solar-mechanical processes, solar absorption cycle technology is more developed; solar thermally driven systems can provide reliable and quiet cooling. In addition, combining solar heating and domestic hot water production with cooling can improve economic performance of the system, compared to solar heating or solar cooling alone. Solar absorption cooling was a subject of significant research interest from 1970 to 1980, when a number of demonstration projects were conducted in the United States. However, these systems failed to establish a significant global market for cooling systems due to their high initial cost, lack of commercial hot water driven absorption chillers, and scarcity of demonstrations and impartial assessments by reputable institutions (Kulkarni 1994). This thesis investigates the technical and energy efficient aspects of using high temperature solar thermal receivers with a two stage absorption chiller to cool and heat a building space; it reassesses the feasibility of commercializing solar absorption cooling technology by considering the recent improvements in solar collection technologies, severe electric shortages, and the environmental problems. This research contributes in depth knowledge and methods for the design and operation of solar absorption cooling systems that reduce energy consumption, decrease operational costs, and minimize air quality problems. 1

18 The work reported in this thesis has developed, validated and applied a numerical performance model, a parabolic trough solar collector (PTS) model; this model provides a tool for the analysis of performance data and for the design and operation of a solar absorption cooling and heating system for a building. Additionally, the work developed a comprehensive simulation and applied it to system optimization and sensitivity analyses. This thesis provides generic guidelines on the design and operation of a solar absorption cooling and heating system in order to reduce energy consumption and operation costs, as well as benefit the environment. 1.1 Background and motivation Solar heating systems have been studied for almost 70 years since the Massachusetts Institute of Technology began their studies in 1938 (Beckman 1980). urrently solar heating is relatively mature. However, despite substantial research and development efforts on solar cooling, most have focused on simulation and isolated experiments. Few involved the current state of hot water driven, two stage absorption chillers and well designed solar chillers with effective control systems. The requirements for effective operation and maintenance practices have not been explored. A typical solar absorption cooling and heating system is comprised primarily of the solar receivers and the absorption chiller. The solar receivers convert solar radiation to thermal energy in a heated fluid; the absorption chiller then uses this energy in summer to generate chilled water. In the cooling cycle, the system acts as a heat pump; it gets the heat from chilled water and from the sun and rejects heat by cooling water. In winter, instead of using cooling water, the system directly transfer heat to the loop providing space heating. A simplified schematic of a solar absorption cooling and heating system is illustrated in Figure 1-1. solar collectors absorption chiller Hot water building cooling water hilled water heating cycle cooling cycle Figure 1-1 Simplified system arrangement of solar absorption cooling and heating system 2

19 1.1.1 Solar receivers Solar receivers are normally are classified in two groups: flat plate collectors and concentrating collectors as shown in Figure 1-2. A flat plate collector can use both direct and diffuse solar irradiation on a fixed receiving plate, while a concentrating device can only use the direct solar irradiation since the diffuse solar irradiation from various directions focused by the reflector away its focal line where the receiver pipe located. oncentrating collectors, with a relatively small absorption area, can heat the heat transfer fluid to temperatures, far above those attainable by flat plate collectors. Flat plate collectors are normally used in applications that require only low temperatures, less than 100 ; and concentrating collectors are utilized in medium or high temperature applications, up to Absorption ycle Absorption chillers can be thermally driven by using heat from the sun, from engine exhaust gases, or from other variety of sources, to provide reliable and quiet cooling. They do not use atmosphere harming halogenated refrigerants, and they can be used to reduce summer electric peak demand. To illustrate how an absorption chiller works, a comparison between an absorption chiller and an electrically driven vapor compression chiller is shown in Figure 1-3. In the electrically driven vapor compression chiller, on the left of Figure 1-3, a refrigerant vapor is compressed to a higher pressure by a compressor and condensed by rejecting heat to the ambient in the condenser. The refrigerant liquid then flows through the expansion valve to an evaporator maintained at a low pressure mixture of liquid and vapor. This liquid refrigerant initial flushes into a mixture of vapor and liquid. The liquid then vaporizes in the evaporator as it absorbs heat from the water to be cooled thermally producing the cooling effect. Instead of using a compressor, an absorption chiller, on the right of Figure 1-3, produces the same compression effect, raising the pressure of water vapor by absorbing vapor at low pressure in the absorber and consequently desorbing this vapor at a high pressure in the regenerator. The water vapor is absorbed at a low pressure by the concentrated sorbent solution in the absorber. A solution pump then pumps up the diluted solution to a higher pressure and temperature in the 3

20 Flat-plate collector Parabolic trough solar collector (PTS) (one axis tracking) Evacuated-tube collector Integrated compound parabolic collector (fixed in a flat assembly) Flat-plate collector oncentrating collector Figure 1-2 Two types of solar collector Rejected heat ondenser Rejected heat ondenser Heat input Regenerator Refrigerant expansion valve P ompressor Work P Refrigerant expansion valve Solution pump Solution expansion valve Evaporator Evaporator Absorber Heat absorbed from chilled water T Heat absorbed from chilled water T Rejected heat Electric chiller Absorption chiller Figure 1-3 Electric chiller and absorption chiller 4

21 regenerator. The water vapor is boiled off from the diluted sorbent solution in the regenerator by adding thermal energy, and the water vapor is condensed by rejecting heat to cooling water in the condenser. The other processes in the absorption chiller, the vaporization of the water refrigerant and the removal of heat from the chilled water in the evaporator, operate similarly to the ones in the electric chiller. If the temperature of the refrigerant water vapor (steam) produced in the regenerator, is high enough, then it can be used to produce more refrigerant vapor from the weak solution. In return, this vapor is then used in a chiller, which then serves as a double effect absorption chiller. From the thermodynamic point of view, the absorption chiller is a combination of a heat engine and heat pump. The heat engine absorbs heat at a high temperature, rejects heat at a lower temperature, and produces work. The work drives a heat pump that absorbs heat at a low temperature and rejects it to the ambient at a higher temperature Solar collector coupled with absorption chillers All the varieties of solar collectors can be used for cooling, given the broad operating temperature ranges of absorption chillers. Figure 1-4 plots the efficiency of the four types of solar collectors - - flat plate, evacuated tube without (ET) and with concentration (P), and linear parabolic trough (PTS) depending on their operating temperature. Typically, a single effect LiBr/H 2 O absorption chiller requires a heat source; and a double effect one requires a heat source to generate chilled water. Normally flat plat solar collectors are coupled with single effect chillers due to their relatively high efficiency at low operating temperature; and evacuated tube and parabolic trough collectors are used to provide a high temperature heat source for double effect absorption chillers. This thesis is focused on the solar absorption cooling system, comprised of the PTS s and the double effect absorption chiller as indicated by the rectangular outline in Figure urrent studies on solar absorption cooling and heating systems Substantial research and development efforts on solar absorption cooling are in progress; and the technology is evolving. Most of the research focuses on simulation analyses of solar driven single 5

22 effect absorption cooling systems. In general, it has been concluded that this approach is the most economic configuration with the highest system performance. Although theoretically ET or P collectors can heat fluid to 150~160, the required temperature by the double effect absorption chiller, there are few successful studies showing that these two types of solar receivers can be successfully incorporated with double effect absorption chillers. Duff (Duff 2004) has reported that m 2 of integrated compound parabolic concentrating collectors (IP) have been operated at to serve a 70kW (20 ton) hot water driven double effect McQuay/Sanyo chiller to serve a commercial building in Sacramento, alifornia. At the beginning of the operation in 1998, daily collection efficiencies were nearly 50% in the operating range of While later, the highest operating temperature was at 110 with 55% daily collection efficiencies, due to the lower operating temperature with less heat loss. Figure 1-4 Solar collector efficiency and operating temperature required by absorption chiller 6

23 At U.S. Army s Yuma Proving Ground in Arizona a 1245 m 2 of Hexel PTS s solar system supplying heated water to a 160 ton LiBr/H2O double effect absorption cooling system has been successfully operated for nearly 14 years since its installation in 1979 (Hewett 1995). Hewett discussed the technical and economic performances of this project and drew the conclusion that the economics of solar absorption cooling systems are unattractive compared to conventional alternatives (in 1995). The primary rationale included the high capital and operating costs of solar collectors with absorption chillers, compared to those of electrically driven vapor compression chillers. Supplementing experimental research, many simulation based studies have been performed on solar driven double effect absorption cooling systems. Lokurlu carried out system simulation analyses in TRNSYS (Lokurlu 2002). He concluded that the combination of PTS and a double effect absorption chiller is promising for cooling systems with a load of at least 100 kw; Since the PTSs were not available for the study, this work remained in the simulation and preliminary design stage. Wardono developed a computer model of the double effect LiBr/H2O absorption cooling system coupled with tilted flat plate solar collectors for an application in Albuquerque, NM (Wardono 1996). He calculated that the total solar energy input depends upon the ambient temperature, sky clearness index, and system design; and he indicated that the solar contribution for supplying energy to the double effect LiBr/H2O cooling system did not significantly vary for various latitudes. 1.3 Research objective The primary purpose of this research program is the development of systems which reduce the energy requirements for the operation of buildings by a factor of two or greater, and the provision of techniques and tools for the design and evaluation of such systems. This thesis re-assesses the technical aspect of a solar driven double effect absorption cooling system by means of experimental equipment, modeling and system simulation. The aim of this research is to develop methods and tools for the effective design and optimization of a solar absorption cooling and heating system that reduces energy consumption, decreases 7

24 capital, operation, and maintenance costs, and also benefits the environment. The special objectives of this thesis are: the establishment of a unique experimental set up and procedures for investing a solar absorption cooling and heating system the carrying out a test program on both a parabolic trough solar collector and an absorption cooling and heating system the construction of a computerized performance model of a parabolic trough solar collector based on energy balances and heat transfer correlations and validation of the model with data obtained from the test program the development of annual overall solar cooling and heating system simulation, the use of this simulation to design the system and to optimize its performance; and validation of the simulation by data obtained from the test program the analysis of the experimental data, refinement of the model, and improvement of the design and operation on the basis of the simulations The solar collector model and overall solar system simulations developed are now being used as tools to adapt parabolic trough solar collectors and double effect absorption chillers to various climate zones, building applications, and system configurations in order to provide optimized design and operation guidelines. In both its practical and theoretical aspects, this study contributes important knowledge for the application of the solar absorption cooling and heating systems. The practical observation and operation of the particular solar cooling and heating system has laid the groundwork for improvement of the control and integration of solar collectors and double effect absorption chillers; thus the analytical methods provide a platform to analyze and improve systems. 1.4 Research approach To achieve the thesis objectives, this research is focused on solar collector model development, system performance simulation, system installation and test of a high temperature based solar absorption cooling and heating system as illustrated in Figure

25 In this research, mathematical modeling, system simulation, equipment testing, system testing, and data analyses were combined to provide a deeper understanding of the system, to discover the possible improvements in the solar collector design, to optimize the system design and operation, and to provide a framework to design and evaluate solar absorption cooling and heating systems. The research has been carried out by the following steps The planning of the test system Prior to the design and installation of the solar cooling and heating system, studies were performed regarding the solar field location, the structure for supporting the solar collectors, system energy balances, the piping, and other engineering issues. In the mean time, a comprehensive parabolic trough solar collector model and annual system performance simulations were developed to assist the system design and operation The development of solar collector model The performance model of the solar collector is focused on a coated absorber pipe enclosed in an evacuated glass envelope: the receiver of the parabolic trough solar collector (PTS). This steady state, single dimensional model is based on fundamental material and energy balances together with heat transfer correlations programmed in the Engineering Equation Solver (EES). Incident solar energy on the solar collector is distributed among useful energy gain, optical losses, and thermal losses. To represent different optical losses, coefficients have been introduced in the model. This model deals with the thermal losses resulting from conduction, convection and radiation heat transfer to the surroundings, from the receiver. This PTS model is based on energy balance relations for the absorber pipe and the glass envelope together with heat transfer correlations for the various the energy streams among them and the surroundings. It has 192 variables and 160 equations. This model predicts how the efficiency of the PTS is influenced by direct normal solar radiation, the incidence angle, collector dimensions, material properties, the operating temperature, the presence of air in the annular space, the wind speed, the type of the fluid, and the operating flow rate. This model was used to select the proper operating conditions, to detect the possible problems in the operation of the collectors such as incipient boiling at the tube surface, and to select a circulation pump. Additionally the estimated performance of PTS has been used to optimize design parameters and operating conditions of the solar absorption cooling and heating system. 9

26 1.4.3 The development of annual system performance simulation The system performance simulations were modeled in the TRNSYS transient simulation program by two sections; the building simulation and the solar energy system simulation. The sophisticated building simulation calculates the building heating and cooling loads based on the inputs of the configuration of the building; weather conditions; the schedules for occupancy, lighting, equipment; and set points for temperature and humidity in the test building space. The thermal system simulations estimated the required energy input to meet the calculated building loads, either as available solar radiation or as natural gas auxiliary fuel. The thermal system simulation included all of major system components and operation strategies. Most of the major components in the test solar energy supply system are available in the simulation library. Only two new components, the solar receiver and the system controls, were written to integrate controls for PTS, chiller, pumps and fan coils. Figure 1-5 Research approach schematic chart 10

27 The system simulation has been used as a generic system model to optimize system design and operation, and to assess the impact of these parameters on the system performance: the orientation of the PTS; the location and volume of the storage tank; the piping diameter, length, thickness, and the insulation; as well as the operating strategy The installation of the test system The solar thermal system was designed to cool or heat the south zone of the Robert L. Preger Intelligent Workplace (IW), an office space for multiple uses class rooms, laboratories, meeting spaces, offices for faculty and students, at arnegie Mellon University. To meet the cooling and heating loads of this space, a 16 kw double effect absorption chiller was selected and installed. Figure 1-6 Process and instrumentation diagram of the test solar absorption cooling and heating system This chiller is driven either by hot water or by natural gas to provide cooling in the summer and heating in the winter. The chiller switches between the cooling mode and heating mode by adjusting a two-position valve. This chiller incorporates a cooling tower to reject heat from its operation as required in the cooling cycle. To satisfy the requirement of the double effect absorption chiller, 52 m2 of linear parabolic trough solar thermal receivers, (PTS) were installed, including a circulating propylene glycol water mixture, instrumentation for flow, temperature, pressure and direct normal solar radiation; circulation pumps, an expansion/pressure tank; and a drain/ filling apparatus. A web-based automation system was also installed to operate 11

28 the solar collector, heat exchanger and the absorption chiller with their auxiliary system, monitor the overall system status, and collect experimental data. In addition, there is a heat exchanger installed parallel to the absorption chiller to compare the system performance between the solar heating systems using an absorption chiller and the heat exchanger. The process instrumentation diagram of the test solar absorption cooling and heating system is shown in Figure The test program and experimental data gathering The experiments provided significant knowledge and understanding of the PTS s and the solar absorption cooling and heating system. The test program characterized the equipment and the systems, validated the mathematic model of the solar collector, and the system simulation for evaluating the annual system performance. The experiments in the test program are classified into three groups: Solar collector performance testing the solar collectors operated at a steady state, the inlet temperature of heat transfer fluid (HTF) entering the solar collector receiver arrays and the direct normal solar radiation are constant; testing the solar collection at a transient state, when the operation temperature of the solar collectors increases with time until an elevated temperature is researched. Solar heating in the morning, circulating the heat transfer fluid through the bypass until the desired temperature is reached, and then diverting the HTF to the absorption chiller/ heat exchanger to produce hot water for space heating. The building load, heat exchanger HX-1 maintains the hot water from the chiller / HX-2 in the reasonable range by the available flow of chilled water. Ultimately, when the solar energy is no longer adequate to operate chiller / HX-2 due to the heat loss from the system and the reduction of the direct normal solar radiation, the HTF is then diverted through the bypass. Solar cooling in the morning, circulating the HTF through bypass until the temperature required by the absorption chiller is reached, then diverting the HTF flow through the absorption chiller to produce chilled water for space cooling. Ultimately, when the amount of solar energy supply is no longer adequate to operate the absorption chiller, the HTF is then diverted through the bypass. During these tests, as indicated in Figure 1.6, data on temperature, pressure, flow, and direct normal solar radiation were gathered by the data acquisition system for further data analyses. 12

29 1.4.6 Data analyses, model validation and simulation evaluation After the data were collected, they were analyzed using the basic steady state heat balance of the receiver and the balance of the system to determine fundamental characteristics of the system such as the solar collector efficiency, the collector and system heat capacities, and the heat losses from the system making use of statistical and other mathematical tools. The data then were used to validate the PTS model and to evaluate system simulation by comparisons with the solutions of the solar collector model or system simulations. There were two observations from the comparison. First, the initial collector model predicted higher collector efficiency than the measurements. This discrepancy proved to be due to the absorptivity of the glass envelope, a significant parameter impacting on the collector efficiency. This absorptivity was apparently much higher than the value given by Broad, the equipment manufacturer. After the properties of the glass envelope in the initial collector model were adjusted, the deviation between the experimental data and the model solution was minimized. Second, initial system performance simulation estimated much shorter preheat time than those observed at the beginning of each day. This discrepancy proved to be due to the pipe component in the TRNSYS modeled as an empty pipe without the heat capacity. A small storage tank, inserted in the system simulation, added appropriate heat capacity to the piping. As a consequence, the experimental data and the system simulation solution came into agreement. 1.5 hapter overview This thesis contains seven chapters followed by references, appendixes, and nomenclature. hapter 1, Introduction introduces the background and motivation of this dissertation and summarizes the research objectives and approach. hapter 2, Solar absorption cooling and heating test system and program introduces the overall experimental system setups of solar absorption cooling and heating test system. It presents detailed information on the system devices, instrumentation and control. The test program and data acquisition are introduced. The experimental data are analyzed. hapter 3, Solar collector performance model addresses the model assumptions, energy balance and heat transfer and calculation procedure. It also introduces the facts of the absorption of solar irradiation impact on the solar collector efficiency. 13

30 hapter 4, Experimental data analysis on PTS model assesses and validates the solar model by using experimental data. The model is used to analyze the PTS performance under various weather and operational condition. hapter 5, Solar absorption cooling and heating system simulation introduces the simulation assumptions and in depth description of operation and control in the base case of the solar cooling and heating simulation. This chapter also includes the assessment and evaluation of the system performance model by using the experimental daily data. hapter 6, Simulation-based design and performance analysis on absorption cooling and heating system presents the system optimization and system sensitivity analysis by serial system comparison simulations. The guidelines of design and operation for solar cooling and heating system are provided. hapter 7, onclusions and recommendations summarizes the finding and contributions of this thesis and suggests future areas for research and the issues involved, including: the thermal storage equipment, an advanced control system, and the integration of cooling and heating devices. 14

31 2 Solar absorption cooling and heating test system and program A solar absorption cooling and heating test system has been designed, installed, and tested in the IW at arnegie Mellon University of Pittsburgh. As shown in Figure 1-6, the system consists of 52 m 2 of parabolic trough solar collectors, PTS s; a 16 kw double effect absorption chiller; a heat recovery exchanger; and a variable, simulated building load exchanger to measure the performance of the solar collector and the overall solar cooling and heating system. A web-based data acquisition and control system was developed and installed to operate the solar system while storing and displaying the test measurement data. The PTS was tested at various operating conditions: direct solar irradiation, wind load, heat transfer fluid, flow rate, and temperature. Daily tests on the solar cooling and heating were conducted at various weather conditions: clear day, mostly sunny day, mostly cloudy and overcast in winter and summer, respectively. In the future, this solar absorption cooling and heating test system will be integrated with the cooling and heating units of the IW and incorporated in the campus cooling and heating grids. 2.1 Parabolic trough solar collector Device description The solar collectors installed in the IW are single axis tracking solar concentrators: parabolic trough solar collectors (PTSs). They track the altitude of the sun as it travels from east to west during the day to ensure that the radiation from the sun is continuously focused on the linear receiver. These PTS s, provided by Broad Air onditioning o., have a 52m 2 aperture surface, which is the total open cross sectional area of four modules of parabolic reflectors. Figure 2-1 shows two arrays of PTS s installed in series in the two valleys of the IW saw-tooth roof. They are connected by the supply and return lines of the heat transfer fluid (HTF), an aqueous solution containing 50% propylene glycol, with the absorption chiller installed on the southeast platform adjacent to the IW. The tracking axes of the PTS s are oriented 15 east of true north because this orientation minimizes the collector height, wind loading, and structural requirements of the installation. 15

32 The PTSs The IW Pipelines Figure 2-1 The PTS s installed on the IW Parabolic trough reflector (Stine 1987) Support structure Receiver tube Tracking mechanism Figure 2-2 Broad BJ16A parabolic trough solar collectors and the receiver tube 16

33 2.1.2 Major components and characteristic The installed PTS, shown in Figure 2-2, comprises a parabolic trough reflector mirror; a receiver tube, a surface treated absorber pipe at its focal line surrounded by an evacuated transparent tube; supporting structure; and a tracking mechanism. A module of the Broad PTS receiver weighs 200 kg. It is designed to withstand a 31 m/sec wind load. The m 2 aperture area and 0.68 m 2 receiver area corresponds to a 19.6 concentration ratio. The parabolic reflector trough, according to Broad, has a reflectance of 0.8. Most of solar radiation on the reflector is focused on the receiver tube, after it impinges on this mirror. The receiver tube consists of an absorber pipe at its center surrounded by a glass envelope as shown in the right bottom of Figure 2-2. The absorber pipe is coated with selective blackened nickel, which has a high absorptivity of 0.96 for short wave length solar radiation and a low emissivity of 0.14 at 100 for long wave length heat radiation. The glass envelope plays an important role in reducing convective and the radiative losses from the receiver tube to the atmosphere. Its surface temperature is much lower than that of the absorber pipe, and it is opaque to thermal radiation from that tube. In addition, the annular space between the absorber pipe and the glass envelope is evacuated in order to decrease the conduction and convection between them. The Broad PTS tracking drive is a large semi circular gear engaged with a small gear powered by a 24V servo motor. PTS s typically have a higher efficiency than plate solar collectors such as a flat solar collector or an evacuated-tube collector when the operation temperature is high. The characteristics of the PTS installed in the IW are addressed in Table Absorption chiller Device description The absorption chiller installed in IW is a dual fired two-stage, water-libr chiller with a cooling tower. This chiller, also fabricated by Broad Air onditioning o., has a 16 kw rated cooling / heating capacity driven by either hot water or natural gas. Figure 2-3 is the absorption chiller installed on the southeast platform of the IW. It is connected with solar collection loop, and its chilled / hot water supply and return lines. Its working schematic flow diagrams 1 are illustrated in Figure 2-4 and Figure From the absorption chiller brochure of Broad Air onditioning o. 17

34 Table 2-1 Specifications of the parabolic trough solar collector installed Parabolic trough solar collector features (BJ16A) Manufacturer: Operating temperature: Module size: Module operating weight: Drive group size: Delta-T loop size: Rim angle, 2 : Reflectors: Focal length: Receiver: Sun tracking Tracking drive System Wind loads Broad Air onditioning Broad Town, hang Sha, Hu Nan, HINA * 5.75 m; m kg 2 modules; m 2 2 drive groups; 52.9 m 2 73 Typical reflectance cm Absorber OD: 3.8 cm Base material: Stainless steel 304L oating: Black nickel Typical absorptivity: 0.96 Typical emittance: 0.14@100 Pyrex glass cover OD: 10.2 cm Transmissivity: 0.91 Vacuum in the annular space Single-axis elevation tracking based on the calculated sun position 24 V powered Servo motor Small gear, big gear. 16 m/sec (tracking) 31 m/sec (stowed) Major components and characteristics The absorption chiller consists of five major and minor heat transfer components, three pumps, a cooling tower, and other associated valves and pipe fittings in Figure 2-4. The five major components include: an evaporator (marked 4), an absorber (marked 5), a hightemperature regenerator (marked 2), a low-temperature regenerator (marked 1), and a condenser (marked 3). The three minor heat transfer components are: a high temperature heat interchanger 2 Rim angle is the angle between the line from vertex to focus and the line from focus to the parabola ridge point. 18

35 (marked 6), a low-temperature heat interchanger (marked 7), and a refrigerant by-pass heat exchanger (marked 7a). The three pumps are: a solution pump (marked 8), a chilled / hot water pump (marked 12), and a refrigerant pump (marked 9).Table 2-2 lists the chiller specifications from the manufacturer; these were the commissioning data before the chiller sent out. The chiller specification data are useful in evaluating the experiment data of the chiller. Figure 2-3 Absorption chiller Both of these two flow diagrams intend to address the general working principle; they do not include the heating coils inside the regenerator. In the cooling cycle, the water vapor is absorbed into concentrated LiBr solution in the Absorber shown in Figure 2-4. A solution pump then pumps up the dilute LiBr solution Figure 2-4 Absorption chiller in cooling cycle Figure 2-5 Absorption chiller in heating cycle 19

36 to the Regenerator that operates at a higher pressure and a higher temperature to vaporize water from the solution making use of thermal energy from the solar collectors or from the natural gas burner. The water vapor is condensed by rejecting heat to cooling water in the ondenser. Next, condensate water is passed through an expansion nozzle into the Evaporator. The water is vaporized there at a low pressure, absorbing heat transferred from chilled water flow. The installed double effect absorption chiller has a coefficient of performance of 1.2 when operated in cooling mode, driven by either fluid heated in solar receivers or by natural gas. A single valve, marked 24 on both Figures 2-4 and Figure 2-5, can be opened to switch the chiller from the cooling to heating mode. In the heating mode, the water vaporized from the LiBr solution in the Regenerator, directly flows into the Evaporator, as shown in Figure 2-5. The Evaporator now acts as a ondenser and heats a second water stream that is used for heating the IW. At the design operating conditions in the cooling mode, the chiller cools 9 gpm of 14 return chilled water to 7 and rejects the heat to cooling water at 30 from its integral cooling tower. In the heating mode, the chiller heats 9 gpm of 50 return hot water to 57 by condensing vapor from the Regenerator in its Evaporator. 2.3 Solar absorption cooling and heating test system System description A solar absorption cooling and heating test system in the IW was set up to test the parabolic trough solar collector and system performance. This test system, shown in Figure 1-6, consists of two loops: the solar collection loop and the variable load loop. The absorption chiller / the heat exchanger are in the middle. It is connected with the HTF supply system on the left and the cooling / heating load system on the right. The necessary instrumentation and control for the system have been installed to operate the system and to process the experimental data The solar collection loop The solar collection loop comprises two main circulation pumps, an expansion tank, and threeway valve in addition to the PTS s, the absorption chiller, and heat exchanger HX-2. One of two pumps, marked S1 in Figure 1-6, was provided by Broad to circulate the HTF in the solar loop including the absorption chiller. It is packed with three-way valve, Broad instrumentation, supply and return lines, and the PTS control panel together in a metal cabinet, shown in Figure 2-6. To 20

37 reduce the difficulties of the system integration and control, another pump, marked as S4, and a heat exchanger, HX-2, were selected and installed in the solar loop. The three-way valve was installed in Broad pump & control package for adjusting the amount of the HTF flowing though by-pass and the absorption chiller. Its control will be discussed in the control section of the test system. The volume of the HTF varies with the temperature changes in a closed system due to thermal expansion and contraction. An expansion / compression tank is required to accommodate the varying quantity of liquid and operate the HTF within a selected pressure range. Since the operation temperature of the absorption solar cooling and heating test system is between -20 to 180, an expansion tank, provided by Zilmet in Italy, was installed in the mechanical room to maintain a controlled pressure and to prevent vaporization in the closed solar collection loop. It is located at near the point of lowest pressure in the system and shared by the two solar collection loops (solar absorption chiller loop and solar heat exchanger loop). Its maximum working pressure is 10 bars and its pre-charge is 36 to 44 psi (2.5 to 3 bar). Table 2-2 Specifications of the absorption chiller installed Solution Power Heating water ooling water Name Quantity Unit ooling capacity kw hilled water return temperature hilled water supply temperature 7.63 hilled water flow rate 2 m 3 /h Heating capacity 4.5 kw Heating water return temperature 50 Heating water supply temperature 52 Heating water flow rate 2 m 3 /h kw Hot source consumptions m 3 /h Power voltage 220 V Power frequency 60 HZ Maximum power consumption kw Water-LiBr sorbent solution mass 70 kg Water-LiBr sorbent concentration 56 % 21

38 The heat exchanger, HX-2, was installed in the solar collection loop parallel to the absorption chiller to compare the solar heating system performance of two configures by the absorption chiller and by a heat exchanger. This heat exchanger from Bell and Gossett is a brazed plate heat exchanger, which offers the highest level of thermal efficiency and durability in a compact, low cost unit. Broad pump S-1 ontrol panel back view Temperature and pressure sensors Threeway valve Supply and return lines Figure 2-6 Broad pump and control package According to the specification sheet of the provider, the overall heat transfer coefficient is 359 Btu/hr-ft 2 -F (2.05 kw/m 2 -) and the total heat transfer surface area is ft 2 (1.05 m 2 ). The Maximum temperature is around 232, and the maximum pressure is 435 psi (30 bar). At the design conditions, the expansion tank sets the initial pressure in the system. The HTF circulates through the three way valve into the by-pass line back to the solar collection loop. When the temperature of the HTF meets the requirements of the absorption chiller, the HTF flows through the absorption chiller and solar energy is then used to cool / heat the space The load loop Systematic testing, of the PTS and solar cooling and heating system, requires a load that can readily be adjusted. This load is provided by a shell-and-tube heat exchanger HX-1 fed with chilled water or hot water from the campus grids for heating or cooling tests, respectively. The 22

39 flow of the chilled / heated water is controlled by a valve to achieve a desired set point temperature and cooling / heating load. Figure 2-7 shows the pictures of the overall system. The solar collection loop The load loop Figure 2-7 Overall solar absorption cooling and heating test system Instrumentation, control and data acquisition system For operation of IW solar cooling and heating system, an instrumentation, control and data acquisition systems has been provided by the Automated Logical o. (AL). The AL control system collects the measurement data from the operation of the solar collector, the absorption chiller, heat exchangers and variable load heat exchanger to evaluate the device and the system performance. The AL control system is a web-based, BAnet as protocol, control and data display system. The system can be operated and observed in anywhere of the world through a standard web browser, without the need for special software on the workstation. This system works together with two individual control systems from Broad, which are used to operate the PTS s and the absorption chiller, respectively as indicated in Figure 2-8. The PTS control system and the chiller control system, provided by Broad, are used to operate the devices separately by different portable control panels and also to communicate and corporate with each 23

40 other in the solar driven cooling or heating operation. The data collected in both of the Broad control systems are in a local domain; it is necessary to copy them to a working computer in order to analyze the data. In addition to operating devices other than the PTS and the absorption chiller in the system, the AL control system also serves as the major data acquisition and data analysis system Data acquisition and display The AL control software, called the Web ontrol Server (WebTRL), is used to program the system operation control logic, set or change control parameters, and display the system operation. Figure 2-9 is the graphic interface of WebTRL designed for the solar thermal system. The WebTRL plots and presents historical data in various forms, such as graphics, trends, and reports. The measurement data can be sampled in any time step from a second to a year Instrumentation in AL system Figure 1-6 shows the process and instrumentation diagram for the solar absorption cooling and heating system. This figure includes all of components and instrumentation installed in the system. There were a total of 21 sensors installed in AL control system including flow rate sensors, RTD temperature sensors, pressure sensors, and pyrheliometer for direct normal solar radiation, as listed in Table 2-3. Figure 2-8 Structure of control system 24

41 ontrols for the solar cooling and heating test system There are three control systems in the test system as addressed before. To integrate two Broad control systems into the AL system, third-party integration has to be used so that the PTS control and the chiller control can directly communicate with AL control system through a standard communication part. However, since this integration requires that Broad open their private communication protocol to AL, the control system integration was not implemented. Therefore an educated operator is required to integrate the operation of the overall system. None the less, in all of these individual control systems, the control is electronic based, not like traditional pneumatic or electric controls, with advanced control algorithms making the individual system operation efficient and reliable. The control elements for the PTS system include Tracking the sun Startup and shutdown the circulation pump Startup and shutdown automatically tracking the sun Defocusing to prevent any hazards The details of the PTS control principles are discussed in Appendix 1. Figure 2-9 Interface of the WebTRL 25

42 Table 2-3 Instrumentation of IW solar cooling and heating system Label Sensor location Mediu Range m T1 PTS s inlet HTF (-45.5 ) to 260 Manufacturer accuracy T1b PTS s inlet HTF (-45.5 ) to 260 The solar collection loop Temperature T2 PTSs outlet HTF (-45.5 ) to 260 T2b PTSs outlet HTF (-45.5 ) to 260 T17 middle PTSs HTF (-45.5 ) to 260 T3 hiller inlet HTF (-45.5 ) to 260 T4 hiller outlet HTF (-45.5 ) to 260 T15 HX2 inlet HTF (-45.5 ) to 260 T16 HX2 outlet HTF (-45.5 ) to 260 Sensor: ±0.3 at 0 ; ±0.8 at 100 ; ±1.3 at 200 ; Transmitter: ±0.1% of span flow F1 PTSs inlet HTF 0 to 10 gpm F2 hiller inlet HTF 0 to 10 gpm ±0.2% at 100% flow, ±0.1% at 40% flow Pressure P1 middle PTSs HTF 0 to 250 psi P2 PTSs outlet HTF 0 to 250 psi ±0.25% of full span The load loop flow Temperature T5 hiller HW/HW inlet water (-10 ) to 110 T6 hiller HW/HW outlet water (-10 ) to 110 T7 HX1 outlet to grids water (-45.5 ) to 260 T8 HX1 inlet from grids water (-45.5 ) to 260 F3 hiller HW/HW inlet water 0 to 10 gpm F4 HX1 outlet to grids water 0 to 10 gpm ±0.10% at 25% of span ±0.40% others Sensor: ±0.3 at 0 ; ±0.8 at 100 ; ±1.3 at 200 ; Transmitter: ±0.1% of span ±0.2% at 100% flow, ±0.1% at 40% flow Others pyrheliometer IW roof Irrad. 0 to 1400 W/m 2 ±0.5% of full span gas meter hiller gas inlet gas 1.4 cfh to 200 cfh ±0.8% of full span 26

43 The control elements for the chiller system are Startup and shutdown hilled water / Hot water temperature control ooling water control Natural gas volume control Hot source flow control Three-way valve control Vacuum maintenance rystallization judgment and de-crystallization Safety and alarm The details of the chiller control principles can be found in Appendix 2.A. of Hongxi Yin s thesis An Absorption hiller in a Micro BHP Application: Model based Design and Performance Analysis (Yin 2006). In addition to the controls by above control system, controls in AL are follows: Startup and shutdown the circulation pump S-4, S-5 Switch the operation operations: AL solar collection loop only, Broad solar collection loop only, Broad solar cooling and heating system, AL solar heating system Adjust variable load Safety and alarm With the control, instrumentation, and data acquisition systems, the solar absorption cooling and heating system can be operated to complete various test programs. The performance of the PTS and solar thermal system has been investigated, and the results and approaches are addressed in the following sections. 2.4 The test program The test program was planned, and then conducted in 37 days from February to September of The weather conditions for the tests included clear days, mostly sunny days, mostly cloudy and overcast days in both winter and summer. There were four varieties of tests used to evaluate the PTS and the system performance: PTS tests at transient states, PTS tests at a steady state, 27

44 solar absorption cooling/heating tests using the absorption chiller, and solar heating tests using heat recovery exchanger, HX-2. The operating temperature, operating system, and the number of days for these tests are listed in Table 2-4. Table 2-4 Four types of tests conducted Test name days PTS test at transient states 10 PTS test at a steady state 17 Solar absorption cooling / heating daily test Solar heating daily test by using heat exchanger Operated system Solar + HX Solar + hiller Solar + HX Solar + hiller HTF temperature operated ( ) ; Solar + hiller Solar + HX 0-93 PTS test at transient states In this type of the test, the HTF was heated in the PTS s and circulated through the by pass or the heat exchanger, HX-2, in the solar collection loop as shown in Figure The PTS s were operated in a transient state in which the operating temperature of the solar increased with time due to the solar heat gain. The operation ceased when a critical temperature, the maximum operating temperature specified by Broad, was reached. The data from this type of test was used in the determination of the optical efficiency, the heat capacity, and the heat and pressure losses of the PTS s and the balance of the system. Solar collectors Solar collectors HX2 - S5 S1 Figure 2-10 PTS test diagram at transient state PTS test at a steady state In these tests, as in the transient state tests, the HTF was heated in the PTS s and circulated through the by pass or HX-2 at the beginning of test, until the desired elevated temperature was 28

45 reached. Then the HTF was switched to flow through the absorption chiller or through the heat recovery exchanger HX-2 and the pump S5. A building load was simulated by the load exchanger HX-1 and was adjusted to maintain the solar loop at a near constant operating temperature, a quasi-steady state. The quasi-steady state refers to the condition of the collector when the flow rate and inlet fluid temperature are constant, but the exit temperature changes slightly due to the normal variations in solar irradiance that occur with time for clear sky conditions. If the steady state became difficult to maintain due to the heat loss of system and the reduction of solar radiation, the operation was halted. This type of the test, as illustrated in Figures 2-11 and 2-12, is used to determine the performance of solar collector. The experimental data collected are presented in Table 2:1 of Appendix 2. Solar collectors Absorp. hiller HX1 - S1 Figure 2-11 Solar absorption cooling /heating system daily test Solar collectors HX2 - HX1 - S5 S4 Figure 2-12 Solar heating daily test by using heat exchanger Solar absorption cooling / heating daily test In these tests, when the desired HTF temperature was reached, the HTF was diverted to absorption chiller to produce chilled water or heated water for space cooling or heating. The 29

46 simulated building load was modified to maintain the temperature of the chilled water or heated water from the chiller within a reasonable range by regulating the flow of heated water or chilled water from the grids of the building flowing through the load exchanger, HX-1, as indicated in Figure Finally, when amount of solar energy supply was no longer adequate to operate the chiller due to the heat loss of system and the reduction of the direct normal solar radiation, the HTF was switched back through the by pass. Solar heating daily test by using heat exchanger In these tests, the HTF was circulated and heated through HX-2 without running the load pump, S4 on the cold side of the exchanger until the desired operating temperature was reached. Then HX-2 was used to provide heat to the load exchanger, HX-1 by operating the load pump, S4. The simulated building load was modified to maintain the hot water from HX-2 within a reasonable range by regulating the flow of chilled water flowing on the cold side of the HX-1, as shown in Figure Finally, when amount of solar energy supply was no longer adequate to operate HX- 2 due to the heat loss of system and the reduction of the direct normal solar radiation, the system operation was halted. Overall, the purpose of the solar thermal building cooling/heating test program is to determine operating conditions such as temperature, flow rate and pressure of the HTF in the PTS s and the system in various situations. the characteristics of the PTS s: their optical and overall efficiencies, heat capacity, heat and pressure losses over a range of operating conditions. the performance of the absorption chiller: its capacity and OP for both cooling and heating, depending on the operating conditions. the time required for the PTS s and the system to reach a desired operating temperature. the characteristics of system, such as heat capacity and heat and pressure losses the potential of solar thermal energy for space cooling/heating in a building. the choice between the chiller and the heat recovery exchanger HX-2 as representing the most effective system for heating. possible design and operational measures for improving the performance of the PTS s, the chiller, and the overall solar thermal cooling/heating system. techniques for the design of such systems for optimal, economic performance. 30

47 The data acquisition system gathered the data of direct normal solar radiation, temperature, pressure, and flow rate throughout the system. Samples of the gathered data for use in analyzing the performance of the PTS s and the solar building cooling/heating system follow The results of PTS test at the transient states The PTS tests at transient conditions were conducted by focusing the PTS s, heating the HTF to a desired temperature without rejecting heat, and then defocusing the PTS allowing the HTF temperature to drop. This process has been repeated under various flow rates, starting times during a day, and ambient temperatures. Experimental data from a test on 29 March 07, typical T-t and Q-t charts, are shown in Figure 2-13 and These charts show that the HTF temperature was nearly uniform in the solar collection loop during the heating period. In addition, after the PTS was defocused, the HTF at the outlet of the PTS was lower than one at the inlet due to the heat losses from the receiver to the surroundings. In Figure 2-14, the fluctuated solar energy flow impinging on the solar collector indicated effects of clouds. The energy collected by solar receivers is about 10% of solar energy impinging on the solar collector because under transient state operation, significant of solar energy is transferred to HTF, piping and fittings and what delivered is small Temperature in T_sr_out T_sr_in 9:30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Mar.29, 2007 Figure 2-13 Operating temperatures of the PTS test at transient state on 29 March 07 31

48 Idn*Aa*cos(theta) Power rate in kw :30 9:50 10:10 10:30 Q_solar_delivery 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Time on Mar.29,2007 Figure 2-14 Energy flows of the PTS test at the transient state on 29 March System heat capacity test The PTS s heat capacity determines the amount of energy and thus the time required to raise the temperature of the HTF circulating through the system to a desired operating value during the system warm up. Since thermal losses can confound measurements of the PTS s thermal capacity, a special test was performed under conditions that minimize such losses. The test used chilled water and heated water from campus grids: the PTS s are not focused. First, the HTF was circulated through the heat recovery exchanger, HX-2, in the solar collection loop and cooled by rejecting heat to chilled water from grid through the cold side of the test load exchanger, HX-1, until its temperature was lowered to a value 5-15 below the ambient temperature. Next, a stream through of heated water from the building grid replaced the chilled water circulating through HX-1. The HTF was heated by this source until its temperature reached a value 5-15 higher than ambient temperature. The test was then halted. It was assumed that the net exchange of heat between the system and the ambient surroundings in these two process steps was nil. The heat capacity of the system was calculated by dividing the total heat transferred from and to the HTF in HX-2 in these two process cooling and heating steps by the difference between the maximum and minimum HTF temperatures. The test basically included two processes: cooling and heating. The system heat capacity was also calculated from the masses and specific heats of the PTS s, the exchanger, the pipe, and the HTF fluid they contain. 32

49 Table 2-5 shows good agreement between the measured and the calculated values of the total heat capacity of the system. Table 2-5 Heat capacity of the solar collection loop Test # T ambient ( ) Measured heat capacity (kj/ ) alculated heat capacity (kj/ ) 576~ The results of PTS test at a steady state The PTS tests were performed only on clear days when the direct normal solar radiation was greater than 630 W/m 2, and its variability was less than ± 4% throughout the tests. Since wind velocity greatly impacts the convective heat loss from the PTS s, all of performance tests were performed with wind speed less than 4.5 m/s. Turbulent flow was maintained within the absorber pipe to ensure good heat transfer between the fluid and the pipe. In the various tests, either the absorption chiller or the heat recovery exchanger, HX-2, were involved. The operating temperatures of HX-2; the pump, S4; and the temperature sensors T5 and T6 are limited to those lower than 107. To test higher operating temperature in the PTS s, the absorption chiller was used. The PTS tests have been conducted in 17 clear days in the period February to July The ranges of operation conditions are listed in Table 2-6. Figure 2-15 and 2-16 are the experimental temperature-time, and energy rate-time data from a typical PTS performance test at a steady state on 22 April 07. The plots show the temperatures at the inlet and outlet of the PTS s and the two heat exchangers, HX-2 and HX-1. The time step for the measurements was 1.0 minute. During the test period, three quasi steady states were established for which the PTS performance was calculated. Table 2-6 Operating condition in the PTS performance tests T ambient ( ) Wind (m/s) T inlet ( ) FR solarloop (gpm) Reynolds Number Incident angle Direct Normal solar irradiation (W/m 2 ) -2.5 ~ ~ ~ ~ ~ ~ ~

50 Temperature in Power rate in kw T_sr_out T_HX2_cs_ T_sr_in T_HX2_cs_ T_HX1_cs_ou T_HX1_cs_ 9:30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Apr.22, 2007 Figure 2-15 Operating temperatures of the PTS test at steady state on 22 April Q_HX2_hs 9:30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 Idn*Aa*cos(theta) Q_useful_solar Q_HX2_cs 14:10 14:30 14:50 15:10 15:30 15:50 16:10 Figure 2-16 Energy flows of the PTS test at steady state on 22 April The results of solar absorption cooling / heating daily test 16:30 16:50 17:10 17:30 Time on Apr.22,2007 Solar tests with the absorption chiller, cooling/heating, were conducted during clear or mostly clear days in Pittsburgh throughout a year. The operation proceeded normally from morning startup to shutdown in the evening or until terminated by the appearance of clouds. The experimental data from these cooling/heating tests were used to characterize the PTS s, the absorption chiller, and the overall system. This performance information is the key in 34

51 determining the overall effectiveness of the solar system in providing cooling and heating for the building throughout a year Solar absorption cooling daily test Solar cooling tests with the absorption chiller have been performed in thirteen days from June to August The weather conditions changed rapidly during this period, and it was difficult to achieve stable direct normal solar radiation throughout a day. The HTF, 50% propylene glycol water solution, was heated through the by-pass in the solar receives and circulated through the Regenerator of the absorption chiller to produce the chilled water that is, in turn, circulated to the load test heat exchanger. In the solar absorption cooling tests, following procedures were used to operate the system: 1. Start up the PTS by operating it in linkage operation mode; the PTS automatically tracks the sun 2. Start up the absorption chiller driven by natural gas 3. Heat the HTF and circulate it through by-pass in the solar collection loop 4. Transfer the chiller operation from gas to HTF, when the HTF reaches the operating temperature required by the chiller. Shut down the gas burner; open the three-valve, so the HTF flows through the Regenerator of the chiller to produce the chilled water 5. Defocused the PTSs by adjusting the initial angle to reduce the amount of the collected solar energy, if the solar energy is in excess. If the solar energy is not adequate for requirement to operate the chiller at full capacity, adjust the flow rate of cold side of heat exchanger HX-1, so that the hot source from solar can still be sufficient to support the partial capacity operation of the chiller. 6. If the solar energy is not enough for the chiller s requirement, turn on the gas burner and stop the HTF flow through the chiller and by pass it through the three way valve back to the solar collection loop. 7. Shut down the PTSs and the absorption chiller The experimental data show that it take about three hours or more to heat the system and its HTF from the ambient temperature to 160, the temperature at which the absorption chiller is programmed to switch from natural gas to solar energy. The absorption chiller operated about four hours a day on the available solar energy during a sunny day in Pittsburgh. After 16:30, the solar collector could not provide HTF at a temperature high enough to operate absorption chiller 35

52 efficiently. The HTF loop from the solar field then bypassed the absorption chiller, and the natural gas burner would be used to drive the chiller. Experiments in three of thirteen days provided reasonable data on system performance for solar cooling. Figure 2-17 through 2-18 show system temperatures and heat quantities for cooling operation throughout two days in July They indicate the operational process: the HTF was heated up from time to time; when the temperature desired by the chiller is reached, the HTF was diverted through chiller; adjust the heated water flow in the cold side of HX-1 to remove heat. Figure 2-17 and 2-18 show throughout a day, the measured temperatures of the HTF at the exit of the PTS s and at the inlet and outlet of the chiller and also of the chilled water at the inlet and outlet of the chiller. The rapid rise of the temperatures of the HTF at the chiller inlet and outlet was the result of the three way valve opening to admit the HTF to the chiller when its temperature exceeds 160. Prior to this time, the chiller was heated by the flow of natural gas to the regenerator. Figure 2-19 and 2-20 show the corresponding calculated heat quantities: the solar input; the product of direct normal solar irradiation from pyrheliometer measurements, actual aperture surface area, and the cosine of the incident angle the delivered thermal energy to the chiller: the product of the HTF flow and the temperature difference over the chiller. the cooling capacity provided by the chiller: the product of the chilled water flow and the temperature difference over the chiller. As indicated by these figures, energy delivered by solar receiver was larger than the energy used by chiller at the beginning of the chiller operation while the relation between them was reversed in the later afternoon. When the HTF was operated at 150 ~ 160, the overall solar efficiency of the PTS s was approximately 33% to 37%. The OP of the installed absorption chiller was in the range 1.0 to 1.2. The solar OP of the overall installed solar cooling system, the product of the OP of absorption chiller and the solar collector efficiency, was therefore about 0.33 to The maximum capacity of the absorption chiller was 12 kw. One of the reasons for this capacity, significantly lower than the chillers design capacity of 16 kw, was related to rate of heat transfer between HTF and LiBr solution. Rate of heat transfer depends on the temperature of HTF and heat transfer coefficient of area. hiller capacity was limited by the operating temperature of HTF, which was about 150 ~ 160. To get full chiller capacity, higher temperature of HTF is 36

53 required. Another reason is the weather conditions in Pittsburgh. Since it was very humid, the direct normal solar radiation was not high, typically about 600~900 W/m 2. Temperature in Temperature in :40 9:00 9:20 9:40 T_sr_out T_chiller_HTF_in T_chiller_HTF_out 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 T_chiller_HW_return T_chiller_HW_supply 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 17:00 Local time on Jul.31, 2007 Figure 2-17 Operating temperatures of solar cooling test on 31 July 07 9:20 9:40 10:00 T_sr_out T_chiller_HTF_in T_chiller_HTF_out T_chiller_HW_return T_chiller_HW_supply 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 17:00 17:20 Local time on Jul.16, 2007 Figure 2-18 Operating temperatures of solar cooling test on 16 July 07 37

54 Power rate in kw Power rate in kw I DN *Aa*cos(θ) Q_chiller_cooling Q_solar_delivered Q_chiller_solarinput 8:40 9:00 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 17:00 Figure 2-19 ooling capacity of solar cooling system on 31 July 07 Idn*Aa*cos(θ) Q_chiller_cooling Q_solar_delivered Q_chiller_solarinput 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 Local time on Jul.31, :40 16:00 16:20 16:40 17:00 17:20 Local time on Jul.16, 2007 Figure 2-20 ooling capacity of solar cooling system on 16 July Solar absorption heating daily test Solar heating tests using the absorption chiller have been carried out in three sunny cold days from February to April 2007, during which weather was still cold, about 0. Following the procedure outlined for the solar absorption cooling tests, the HTF, the propylene glycol solution, was heated in the PTS s, circulated through the Regenerator of absorption chiller and then returned to the PTS s. 38

55 Temperature in T_sr_out T_chiller_HTF_in T_chiller_HFT_out T_chiller_HW_supply T_chiller_HW_return 8:40 9:00 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 Local time on Mar.09, 2007 Figure 2-21 Operating temperatures of solar absorption heating test on 9 March 07 Power rate in kw I DN *Aa*cos(θ) Q_chiller_heating Q_solar_delivered Q_chiller_solarinput 8:40 9:00 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 Local time on Mar.09, 2007 Figure 2-22 Heating capacity of solar absorption heating system on 9 March 07 The heated water produced in the Evaporator of the chiller was circulated to the load test heat exchanger, HX-1. In order to ensure good heat transfer in the solar collectors, especially during the cold weather, calculations were carried out to determine the flow rate to achieve turbulent flow; the results are presented in Table 2:2 of Appendix 2. Similar to the curves presented above for cooling with the solar absorption chiller, Figure 2-21 and 2-22 show the measured temperature and heat flow quantities throughout the day for 9 March Figure 2-21 indicated that there is large temperature difference between HTF source and hot water generated by chiller. That is more loses during heat transfer from HTF to LiBr solution. This high temperature difference is due to the boiling point elevation of LiBr solution. The charts show that the heat delivered by the system 39

56 was about 12 kw, when the hot water was produced by the chiller at 37 o, some 20 lower than the rated supply temperature, 57. If hot water was produced at 52 o by increasing the entering temperature to 50 o, some kw energy input was required; the heating efficiency was significantly reduced, about Reasons for this reduced heat efficiency of the absorption chiller may be the increased thermal losses to the surroundings during the condensation process in the Evaporator and direct absorption of water vapor produced in the Regenerator by LiBr solution in the Evaporator The results of solar heating daily test by using heat exchanger Solar heating tests making use of the heat recovery exchanger HX-2 have been carried out in three sunny, cold days from February to April Experimental data have been obtained to compare the system performance in solar heating using either the absorption chiller with the HTF at 140 o, as reported above, or the heat recovery exchanger with the HTF at 68 o The operation process of solar heating daily tests by using HX was as follows: 1. Start up the PTS by operating it in the automated operation mode; the PTS automatically tracks the sun. 2. Heat and circulate the HTF through HX-2 in the solar collection loop without running the load loop pump. 3. When the HTF reaches the temperature required by heating demands, turn on the load loop pump, so that the heat exchanger starts to generate hot water for load. 4. Reject heat from HX-2 through hot water circulated to HX-1; the flow rate of chilled water at the cold side of HX-1 is adjusted to insure the hot water generated by HX-2 balance the solar energy captured. 5. When the solar energy is not adequate for heating device s requirement, defocus the PTS shut down the load loop pump S-4. Figure 2-23 and 2-24 are the experimental results of a solar heating daily test using HX-2 on 2 March As shown in Figure 2-23, HTF flow through the heat exchanger was initiated when the outlet temperature of the collectors reached 90 ; this temperature dropped to 68 and remained stable as heat was delivered to the water streams flowing through the heat exchangers; The temperature difference between HTF source and hot water generated by chiller is much less, so that less heat lost during heat transfer between the HTF and LiBr solution. 40

57 Temperature in Power rate in kw T_sr_out T_HTF_HX_i n T_HX_HW_out T_HTF_HX_out T_HX_HW_in 9:30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Mar.02, 2007 Figure 2-23 Operating temperatures of HX based solar heating system on 2 March :30 9:50 10:10 10:30 Idn*Aa*cos(θ) Q_solar_delivered Q_HX_solarinput Q_HX_heating 10:50 11:10 11:30 11:50 12:10 12:30 12:50 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Mar.02,2007 Figure 2-24 Heating apacity of HX based solar heating system on 2 March 07 A comparison of solar heating using an absorption chiller or a heat recovery exchanger, based on the tests, presented in Table 2-7, shows clearly that a heat recovery exchanger is more effective. Table 2-7 Heating system performance comparison between HX based and absorption chiller based Solar + HX heating Solar + absorption chiller heating Q_useful (kw) 16 ~29 15~19 Operation hours (hrs/day) in March 5 2 Storage is possible yes not quite 41

58 Use of the exchanger avoids the large temperature difference between the HTF and the heated water in the absorption chiller. And it allows the collectors to be operated at a lower temperature, thus reducing heat losses from the system. 2.5 The interpretation of PTS performance data The experimental data obtained from the Broad PTS s were analyzed by inserting them in the overall energy balance equation for these receivers. Statistical regressions were performed to identify their constant optical efficiency, their heat loss dependent on the operating temperature, and their heat capacity based on measurements obtained during periods of changing temperature The energy balance of the PTS Due to the losses in the transmission and reflection of radiation and in the conduction, convection, and radiation of heat, the PTSs are not able to convert all of the incident solar radiation into the thermal energy of the flowing medium. If the PTS is selected as a control volume, the energy balance of the PTS is indicated in Equation 2-1. The first term in the equation is the total solar energy absorbed by the receiver tube of the PTS; the second term is the useful energy delivered by the HTF to outside of the PTS; the third term is the thermal loss from the PTS to the surroundings; the right side of the equal sign is the energy used to heat the receiver pipe of the PTS and the HTF inside of the receiver pipe. I ( Tin + Tout) d Aa *cos( θ )* IAM * α + m& *( p intin p outtout) Qloss PTS = M cv 2 Equation p _ cv dt DN * 1 Where, I DN = direct normal solar radiation A a = aperture area of the solar trough θ = incident angle, degree IAM= incident angle modifier, addressed in hapter 3 α = optical efficiency, the fraction of the incident solar energy absorbed by the receiver tube of the PT; this fraction is related to the reflectance of the reflector mirror, the absorptivity 42

59 and transmittance of the glass envelope, the absorptivity of the coating of the absorber pipe, the tracking error, and other optical errors m& = HTF flow rate through the PTSs p _ out, p _ in = specific heat of HTF at the outlet, inlet of the PTSs T out, T in = HTF temperature at the outlet, inlet of the PTSs M _ cv1 = mass of the fluid, receiver pipe, and fitting in the PTSs p _ cv1 = specific heat of the fluid, receiver pipe, and fitting in the PTSs ( Tin + T d 2 dt out the PTSs ) = the average rate of temperature change of the HTF, receiver pipe, and fitting in Q loss_pts = thermal losses from the PTSs to the surroundings When the PTS is operated at a steady state, no heat is stored in the control volume; the right side of the equal sign of Equation 2-1 is equal to zero. Solar collector efficiency can be defined by the Equation 2-2. where, m * ( p _ outtout p _ intin ) η = & Equation 2-2 I DN * Aa * cos( θ ) * IAM T η = α β * Equation 2-3 I DN 2 T T η = α β * γ * Equation 2-4 I DN I DN α = optical efficiency β = coefficient of the first order 43

60 γ = coefficient of the second order T = average operating temperature above the ambient temperature Empirical relations are frequently provided for the η of PTS s based on their optical efficiency, α, and their heat loss dependent on their operating temperature. Such equations are shown in Equations 2-3, based on a linear relation of heat loss with temperature, on Equation 2-4 using a second order relation. Actually the heat losses by conduction and convection from the PTS s are expected to be linear with the operating temperature; but the heat loss by radiation to the surroundings, proportional to the fourth power of the operating temperature The selection of experimental data Based on the experimental data from the PTS tests at steady state conditions, the statistical method was used to correlate the performance of the PTS s. First, 55 steady state data sets were selected from the 1800 collected during the 15 day tests. Second, the mean values of the experimental data were calculated based on all of single data collected per minute in each of these steady state sets. Third, the system energy balances were checked by using the calculated 55 steady state data sets. The results showed that the measurements were in reasonable agreement with the energy balances, taking into account the sensor accuracy and the impacts on energy of other devices such as the pumps. Fourth, the mean values of experimental data were introduced into the statistical tool to generate the correlation of the calculated η values by using multiple regressions PTS performance predicted by statistic tool According to the rule of thumb for multiple regressions, the size of the data sample must be at least 20 times as many cases as independent variables. In the PTS efficiency multiple regression of Equation 2-3, there are two independent variables: I DN *cos( θ ) and (T in + Tout )/2 - Tam, so 55 data sets are sufficient to determine the performance of the PTS. The data were analyzed and performed the multiple-regression in MINITAB commercial statistics software. Figure 2-25 shows that the energy transferred to the HTF has a good linear relation with the total solar energy impinging on the reflector trough. In addition, Figure 2-26 shows that the solar energy transferred to the HTF also can be represented by a linear relation with the average operation temperature above the ambient. Two clouds in Figure 2-26 are due to the limitation of two system configurations: absorption chiller system and heat exchanger system. The heat exchanger system 44

61 only permits the HTF operated under 100 due to the requirement of instrumentation and equipment. The absorption chiller can insure a steady state operation of solar receiver only when the HTF is greater than 125. ombination of these two linear relationships, the PTS efficiency was determined by a multiple regression as indicated in Equation 2-5. The optical efficiency, α, is and the linear coefficient of thermal losses, β, is 1.4 W/ o m2. (Tin + Tout )/2 - Tam η = Equation 2-5 I *cos( θ ) DN 35 Scatterplot of mcp(tout-tin) vs I*Aa*cos(θ)*IAM (kw) 30 mcp(tout-tin) I*A a*cos(θ)*iam (kw) Figure 2-25 Scatter plot of I*Aa*cos(theta) and m*(po*to-pi*ti) 35 Scatterplot of mcp(tout-tin) vs Average Operated temp. [] 30 mcp(tout-tin) Average Operated temp. [] Figure 2-26 Scatter plot of average operation temperature and m*(po*to-pi*ti) 45

62 In addition, multiple-regression was also implemented to consider the second order solar collector efficiency Equation 2-4; however, the results showed considerable experimental scatter. The major reason for this scatter may be the fact that the actual thermal loss from radiation is proportional to the fourth power differences of the operation temperature and the sky temperature rather than the second power. 2.6 Discussion of test program The solar PTS - absorption chiller based cooling and heating system has been evaluated in four types of test as shown in Table 2-4. The system heat capacity was determined by tests at a trsiant state. The PTS performance was determined by tests at a steady state. The simplified solar collector efficiency equation was defined in the form of a linear relation with the operation temperature of the PTS. The optical efficiency, α, was estimated by data to be This optical efficiency is related to the reflectance and cleanliness of the reflector, optical error, tracking error, transmittance and absorptivity of glass, and the absorptivity of the absorber pipe coating. The coefficient, β, in the empirical equation for the PTS overall efficiency, Equation 2-5, was estimated by data to be 1.4 W/ o m 2. Experimental data shows that the solar thermal system normally takes about three to four hours to heat the system to 160 o, the nominal operating temperature of the absorption chiller. The overall system has a high heat capacity. This heat capacity consumes much useful solar energy and prolongs the warm up period before solar energy is available to be used by the absorption chiller or the heat recovery exchanger. Design approaches and operational methods for reducing the heat capacity of the system and the warm up period have been devised and are discussed in later chapters. The basic performance of the installed absorption chiller has been studied. In the cooling cycle, the full rated capacity of the absorption chiller is about 12 kw with a OP of Its heating capacity, however, is low, about 4 5 kw: and the heating efficiency is about Solar heating using a heat recovery exchanger, rather than an absorption chiller, is recommended because it has higher solar energy utilization, a shorter solar time for warm up, and a greater efficiency. 46

63 3 Solar collector performance model In the previous chapter, the solar collector performance tests, measurements, and calculation were described. This chapter deals with the overall modeling of the extent to which solar energy is captured and delivered as thermal energy to the flow of a HTF in a PTS. This overall comprehensive performance model will refine the understanding of the principles of the PTS, analyze the experimental data from the test program, assist in the PTS design, and evaluate the system performance of the solar cooling and heating system. The first element of the PTS model in front section of this chapter deals with the solar energy that reaches the PTS s reflector, that is directed to the absorber pipe and that is transferred to the HTF. This PTS model is based on the basic energy balance relations for the absorber pipe and the glass envelope together with heat transfer correlations among them and the surroundings. This PTS model considers the effects of direct normal solar radiation, incident angle, receiver configuration, fluid thermodynamic properties, ambient conditions and operating conditions on the performance of the collector. The model has already been used to size system devices, to choose proper operating conditions, and to detect possible operating problems. For the manufacturer of the solar collector, it can be used to improve design of the PTS; and it can be used to optimize system operation and control for the solar cooling and heating system. The second section of this chapter deals with the extent to which the available solar energy reaches the PTS reflector and is directed to the absorber pipe. It is useful in the design and installation of the PTS s. 3.1 PTS model assumption In modeling the PTS, a number of assumptions have been made to provide a foundation without obscuring the physical situation. These important assumptions are: the PTS is operated at steady-state condition (SS); there is no net heat stored in control volume. there is a negligible temperature increase along the length of the PTS: a one dimensional model. the sky is a blackbody for long wave length radiation at an equivalent sky temperature. the temperature gradients around the absorber pipe and the glass envelope are negligible. the shading effect of the absorber pipe on the reflector is negligible. 47

64 the dust and dirt on the PTS are negligible. 3.2 Energy balance analysis Incident solar energy on the solar collector is distributed among useful energy gain, optical losses, and thermal losses. The optical losses are caused by the shadows, the reflector reflectance, the reflectance and the absorption of the glass envelope, the reflectance and the emission of the absorber pipe, errors from tracking, errors from focusing, and errors from the alignment as well. oefficients are introduced in the model to represent each of these optical losses. The thermal losses are the result of conduction, convection and radiation heat transfer to the surroundings. The PTS model that has been developed is based on energy balance relations for the absorber pipe and the glass envelope together with heat transfer correlations for the various the energy streams among them and the surroundings, as shown in Figure 3-1. Heat conducted to supporting bracket ' q& a _ connection Heat transferred to the fluid q & conv_ f onduction, convection & radiation between absorber pipe and the glass envelope Heat radiated from the glass envelope to the sky q& ' rad _ g _ sky Heat convected from the glass envelope to the ambient air ' q& conv_ g _ air Absorbed by absorber pipe q &' Abs _ a Reflected by absorber pipe Reflected by glass envelope q &' ref _ g Focused direct normal solar radiation q' & Focs Transmitted q &' Tra _ g Absorbed by glass tube q &' Abs _ g Figure 3-1 Energy flow in the PTS As the PTS tracks the altitude of the sun, only direct normal solar radiation is focused to the PTS by the parabolic trough mirror. The diffuse solar radiation is reflected away from the focal line by the parabolic mirror and lost in the sky. The direct normal solar radiation reflected and focused by the parabolic trough reflector impinges on the outer glass envelope surface. This direct solar radiation is attenuated by the reflectance, r, 48

65 of the parabolic trough mirror and by the absorber pipe interception factor, R, due to optical errors in tracking, focusing, and alignment, so not all of solar energy reaches the absorber pipe. Only a small portion of the direct solar radiation is absorbed by the glass envelope, q &' abs _ g of Equation 3-2; a small portion is reflected by the glass envelope, q &' of Equation 3-3; the larger ref _ g portion is transmitted through the glass envelope, q &' in Equation 3-4 and Figure 3-1. After tra _ g the transmitted solar radiation impinges on the absorber pipe, most of it is absorbed, Equation 3-5. Little of it is reflected back to the glass envelope. This reflected radiation is considered negligible. As the solar radiation continues, the absorber pipe is heated at an elevated temperature. This temperature induces heat transfer to the fluid flowing in the pipe, to the glass envelope, and in turn to the surrounding environment. Figure 3-1 also indicates three major thermal losses of the PTS to the surrounding environment: conduction from the absorber pipe to the supporting structure, convection from the glass envelope to the ambient air, and radiation from the glass envelope to the sky. There are two independent control volumes in the PTS: the glass envelope and the absorber pipe. For the control volume of the glass envelope, the energy balance is presented in Equation 3-6, and for the control volume of the absorber pipe, the energy balance is presented in Equation 3-7. q &' abs _ a, of I DN * r * R* IAM * Aa q& ' focs = Equation 3-1 L q& ' _ = q&' * α Equation 3-2 abs g focs g q &' _ = q&' * R Equation 3-3 ref g focs g q& ' _ = q&' * τ Equation 3-4 tra g focs g q& ' = q& ' * α Equation 3-5 abs _ a tra _ g a q & + q& + q& = q& + q& Equation 3-6 ' abs _ g ' rad _ a _ g ' conv _ a _ g ' rad _ g _ sky ' conv _ g _ air q & = q& + q& + q& + q& Equation 3-7 ' abs _ a ' rad _ a _ g ' conv _ a _ g ' conv _ a _ fluid ' a _ connection η = q&' u * L* dt A a I dn dt Equation

66 q &' u = q& ' q& ' Equation 3-9 Abs thermoloss q &' = q& ' + q& ' Equation 3-10 Abs abs _ a abs _ g q & = q& + q& + q& Equation 3-11 ' Thermoloss ' rad _ g _ sky ' conv _ g _ air ' a _ connection Where, I dn = direct normal solar radiation, [W/m 2 ] r = reflector mirror s reflectance, dimensionless R = absorber interception factor, dimensionless IAM = incident angle modifier, addressed in late text, dimension less A a = aperture surface of the PTS, m 2 L = length of the PTS, m α g = glass envelope absorptivity, dimensionless τ g = glass envelope transmittance, dimensionless R = glass envelope reflectance, dimensionless g α a = absorber pipe absorptivity, dimensionless q' & focs = direct solar radiation hit on the outer glass envelope surface, [W/m] q &' abs _ g = solar radiation absorbed by glass envelope, [W/m] q &' = solar radiation reflected by glass envelope, [W/m] ref _ g q &' tra _ g = solar radiation transmitted through glass envelope, [W/m] 50

67 q &' abs _ a = solar radiation absorbed by absorber pipe, [W/m] q& = radiation heat transfer between the absorber pipe and the glass envelope, [W/m] ' rad _ a _ g q& = convection heat transfer between the absorber pipe and the glass envelope, [W/m] ' conv _ a _ g q& ' rad _ g _ sky = radiation heat transfer from the glass envelope to the sky, [W/m] q& ' conv _ g _ air = convection heat transfer from the glass envelope to the ambient air, [W/m] q& ' conv _ a _ fluid = convection heat transfer from the absorber pipe to the fluid, [W/m] q& ' a _ connection = heat transfer through the connection structure, [W/m] q' & Abs = direct solar radiation absorbed by the PTS, [W/m] q' & u = useful solar energy delivered by the PTS, [W/m] q' & Thermoloss = heat losses from the to the surrounding environment, [W/m] The efficiency of the PTS is the ratio between the useful solar gain and the direct normal incident solar energy on the aperture area over a given time period, as indicated in Equation 3-8. The useful solar gain equals the captured solar energy less the total thermal losses, as indicated in Equation 3-9. The captured solar energy q' & can be calculated on Equation by 3-10; and the total abs thermal loss, q' &, is calculated by Equation thermoloss 3.3 Heat transfer analysis Although a comprehensive heat transfer analysis on the PTS is a complicated problem, a relatively simple heat transfer analyses can yield very useful results. The simplified heat transfer network of the PTS employed in the model is illustrated in Figure

68 R a_connection R conv f R cond a R conv a g T air,sky R cond_g R conv_g_air T air T f Fluid T ai T ao Absorber pipe R rad a g T g_i T g_o Glass envelope R rad_g_sky T sky Figure 3-2 The thermal network Where, R conv_f =onvective resistance in the fluid,[. m 2 /W] R cond_a =onductive resistance of the absorber pipe, [. m 2 /W] R a_connection =Heat transfer resistance of from the connection to the surroundings, [. m 2 /W] R conv _ a_g =onvective resistance of the outer absorber pipe surface to the inner glass envelope surface, [. m 2 /W] R rad _ a_g =Radiation resistance from the outer absorber pipe surface to the inner glass envelope surface, [. m 2 /W] R cond_g =onductive resistance of the glass envelope, [. m 2 /W] R conv_g_air =onvective resistance of the outer glass envelope surface to the ambient air, [. m 2 /W] R rad_g_sky =Radiation resistance from the outer glass envelope surface to the sky, [. m 2 /W] From the left to the right are the heat transfer fluid, the absorber pipe, the glass envelope, and the surrounding environment (the sky and ambient air). Heat is transferred from the absorber pipe to the fluid by convection from internal flow of heat transfer fluid, HTF, in the absorber pipe. This heat conducted through the absorber pipe. There are two stream of heat transferred from the outer 52

69 surface of the absorber pipe to the inner surface of the glass envelope: the convection and radiation. Since the annual space between them is evacuated, the convection is very small. This heat delivered to the inner surface of the glass envelope plus the heat absorbed in the envelope is conducted to the outer surface of the envelope, and then delivered to the surroundings by convection to the ambient air and radiation to the sky. At the connection between two PTSs as shown in Figure 3-3, heat is directly lost to the surroundings by convection and radiation of heat conducted through the supporting structure. the connection between two PTSs in one module the connection between two modules Figure 3-3 The connection between the PTSs Table 3-1 summarizes heat transfer correlations for each of components of the PTS. In addition to calculation of heat transfer and energy balance in the PTS, the model also provides the pressure drop calculation in the solar field and the thermal loss through piping in the solar loop. The performance model of the PTS has been validated by comparisons between the model solutions and the test data. Experimental data has been used to adjust/ correct certain model parameters, principally the absorptivity of the glass envelope, to achieve agreement among predicted and measured collector operating data. 3.4 alculation procedure A performance model has been programmed for a solar thermal collector based on the PTS. This steady state, single dimensional model is a set of nonlinear algebraic equations programmed in the Engineering Equation Solver (EES). EES, developed by the University of Wisconsin Madison, is different from other existing numerical equation solving programs in two respects. 53

70 Table 3-1 Heat transfer correlations used in the PTS model Object Heat transfer Reference Equation comment convection Holman 1997; q& conv _ f = hcov_ f π * Dh ( Taverage T f _ b ) A high level of of internal flow Incorpera 1990 h = Nu * k f / D h accuracy is desired for smooth tubes. For turbulent flow 2 ( f / 8)(Re 1000) Pr where, f = (0.79*ln(Re) 1.64) f HTF Nu = 1/ 2 2 / ( f / 8) (Pr 1) For laminar flow *( d / L) * Re* Pr f Nu = / *[( d / L) Re Pr f ] conduction Holman πk ( Ti To ) Absorber pipe through the q& cond _ a = ln( ro / ri ) absorber pipe convection between absorber pipe and glass envelope Holman 1997; Dudley 1994; Incorpera 1990 q& conv _ a _ g = πdaoh( Tao T When 100<Rac<10 7 gi k where, 2 a 9γ 5 h = b = [ ] r ln( r / r ) + bλ( r / r + 1) a 2( γ + 1) i o i i o ) the temperature of absorber surface is uniform Absorber pipe & glass envelope radiation between absorber pipe and glass envelope Holman 1997; 20 2 λ = 2.331(10 )(( T 3 + T4 ) / 2) / Pδ If Rayleigh number Rac < 100, if Rac* =<100,keff = k, else heat transfer equations are follows. 2π keff q = ( Ti To ) ln( Do / Di ) where, keff Pr 1/ 4 1/ 4 = 0.386( ) ( Rac*) k Pr 4 Ra [ln( Do / Di )], 3 c* = Ra g * β * ( T 3 3 / 5 3 / 5 L L ( D + D ) 5 i To ) L RaL = v α q& rad _ a _ g i 4 4 σπdao ( Tao Tgi ) = 1 Dao 1 + ( 1) ε D ε ao gi o gi f f The glass is opaque to infrared radiation. ε gi = 0.86 Glass envelope Glass envelope & surroundings onnection & surroundings onduction through glass envelope onvection from the glass outer surface to the ambient air Radiation from the glass outer surface to the sky onductionconvection from the connection to the ambient air (fin) onductionconvection from Supporting bracket to the ambient air as fin Holman 1997; q& cond _ a 2πk ( Ti To ) = ln( r / r ) Holman 1997; q& conv _ g _ air = πdgoh( Tgo T ) For natural convection Holman 1997; Duffie 1980 Holman 1997; Holman 1997; o for 10-5 <Ra< / 2 Ra Nu = ( 9 / [ 1+ (0.559 / Pr) ] For forced convection 1/ 3 Nu = (Re) n Pr q& = σ * A * ε *( T rad _ g _ sky T sky = T go i go 4 go T 4 sky ) / 9 ) 1 / q& max = 2π ( r2 c r1 ) * h * ( To T ) / L Where, h can be calculated as same equations in convection from glass to ambient air. q& connection = η f * q& max q& cond _ breaket = hpak ( Tbase T ) / L Uniform temperature distributed around the glass envelope 54

71 First, it can automatically identify and group equations that must be solved simultaneously. Second, it can provide many built in mathematical and thermo physical property functions useful for engineering calculations. The EES equations of the PTS, including mass and energy balance, heat transfer equations, thermal property functions and the assumptions are annotated in Appendix 3.1. The interface of the developed model is also shown in Figure 3:1 of Appendix 3. The procedure for the EES calculation is straightforward: first the algebraic equations are entered into EES. The total number of the parameters is equal to the sum of the number of equation and the number of the known parameters based on the given conditions and properties. Reasonable estimates are entered for all unknown parameters. EES uses the embedded or entered lookup tables to find the thermo physical properties of the fluid and other materials of the PTS. The EES then automatically identifies and groups equations that must be solved simultaneously. Finally it solves the equations by adjusting the estimate values of the variable parameters to minimize the square of the residuals from the equations. In the developed model, there are some important properties of the material in the PTS impacting on the PTS s efficiency. Table 3-2 show these import parameters with the values used. Table 3-2 Parameters and values used in the PTS model Parameter Nomenclature Value used in the model α glass envelope absorptivity Initially assumed to be 0.03 (given by g BROAD), 0.1~0.15 by experiment τ glass envelope transmittance 0.82 g R glass envelope reflectance 0.07 g ε glass envelope emissivity 0.86 g α absorber tube absorptivity 0.96 a τ absorber tube transmittance 0.01 a R absorber tube reflectance 0.03 a ε absorber tube emissivity 0.43 a r reflector mirror s reflectance Solar irradiation absorption The developed PTS performance model does not include analysis of optical losses resulting from the parabolic trough reflector and its focus of the solar rays on the absorber pipe of the PTS. However, about 30% to 40% of the incident solar energy lost due to the optical losses. 55

72 Such losses normally are caused by the position of collector, the properties of the material used in the reflector, the focusing errors, tracking errors and errors in the formation and alignment of the reflector surface. So at the beginning of this section, solar irradiation analysis is first introduced in order to achieve an understanding of the optical losses Direct normal solar radiation On the earth s surface, the sum of the incident solar irradiation from all directions is called the global radiation. It mainly comprises the direct normal solar radiation from sun, and the diffuse radiation from all directions except directly from sun. Typically mid to high-temperature solar collectors employ reflectors to concentrate the solar irradiation onto a receiver of reduced area. For collectors with geometric concentration ratio of 10 or greater, the ratio of the reflector aperture surface to the receiver surface area, typically, only direct radiation can be used since the light must strike the reflector at a precise angle in order to be reflected to a predetermined point, the focus point [Stine 1985]. The installed solar system from the PTS study has a pyrheliometer to measure the direct normal radiation. This pyrheliometer is mounted on a solar tracker, which ensures that the sun's beam is always instantaneously directed into the instrument's 5.7 field of view and that the sensors are always located on the plane perpendicular to the sun rays Incident angle and incident modifier For the solar collector designed to operate with tracking rotation about only one axis, a tracking drive system rotates the collector about its axis until the sun central ray and the aperture normal are coplanar. The solar rays impinge on the aperture at an angle with its perpendicular. The intensity of solar radiation on the surface is reduced by a factor equal to the cosine of this angle. For the PTS, the angle between the sun rays and the normal direction of the aperture surface is the incident angle as shown in Figure 3-4. Normal to the aperture surface Sun ray θ Tracking axis Aperture surface Figure 3-4 Incident angle of the PTS 56

73 In addition to reducing the solar intensity, the incident angle also can cause other losses due to additional reflection and absorption by the glass envelope when the angle of incident angle is not equal to zero. The incident angle modifier (IAM) is the coefficient used to correct for these additional reflection and absorption losses. Normally, the IAM is achieved by means of empirical experimental data fitting. The PTS, fabricated by Industrial Solar Technology (IST), has the same materials for coating and glass as Broad PTS. Based on the IAM value given by the IST, the IAM value of the installed PTS in the IW is estimated as shown in Equation IAM = ( θ ) ( θ ) Equation 3-12 Figure 3-5 shows the comparison of the incident angle modifier, cosine value of the incident angle and the products of them. It indicates that the incident angle modifier does not greatly impact the overall solar irradiation intensity compared with the cosine factor of incident angle based on the IAM value End-loss In addition, a nonzero incident angle is also the cause of no solar radiation, reflected from the reflector mirror, on the absorber pipe of the PTS over a certain length. This loss is normally called end-loss. Figure 3-6 depicts the occurrence of end losses on the PTS when a nonzero incident angle. The endloss is related to the focal length of the parabolic trough and the incident angle. Its length for a given PTSs array can be determined by Equation The length of the installed solar array is 12 meter and focal length of the parabola is m. Figure 3-7 indicates that the length of the end-loss in the installed solar field in IW. It shows when the incident angle 1.2 Incident angle modifier and incident angle 1 IAM cos(θ) 0.4 IAM*cos(θ) incident angle Figure 3-5 Incident angle modifier and incident angle 57

74 the reflected sun rays lost at the left end no sun ray reflected on the right end of the receiver pipe f the incident angle Lendloss Lendloss 5.00 Figure 3-6 End-loss of the PTS The end-loss length (m) incident angle Figure 3-7 The length of the end-loss in the solar field is 80, almost 40% of total solar irradiation lost due to the effect of the end loss. Where, L endloss = f * tanθ Equation 3-13 L endloss = length of the end-loss, m f = parabola focal length, m 58

75 3.5.4 Shadow-loss The configuration of solar collector installation may introduce further losses due to the shading from the adjacent parallel row when solar angle is low. The upper part of Figure 3-9 shows how Wr a Sr a Wr ß Sr Figure 3-8 Shadow loss from the adjacent solar collector array to find the shadow angle,α, which can be calculated by Equation 3-14 [15]. The shadow angle is referred to the threshold angle in the shadow loss. Wr α = arcsin( ) Equation 3-14 Sr Where, Wr = the aperture opening of solar collector, m Sr = the horizontal distance between two adjacent solar collector arrays, m The width of collector shadow on the aperture surface of the PTS can be calculated according to Equation W_shadow_co llector = Wr - Sr *sin( β ) Equation 3-15 Where, β = tracking angle of the PTS, from the horizon to the normal direction of the aperture surface. 59

76 3.6 onclusion This chapter presents detail PTS model along with adjustments due to endloss and shadow loss that comprises an overall model of a PTS solar system. The next chapter presents the measured performance of the PTS and the results of model calculations, after the model has been adjusted and validated. 60

77 4 Model-based experimental data analysis of PTS The model developed for the PTS has been used to analyze the experimental data from the test program. The computational model used a number of measurements as inputs to calculate all working conditions. The discrepancies between the measurements and the model calculations have been found and minimized by adjusting the model assumptions and input parameters. The largest discrepancy between experiment measurements and model calculations has been resolved by an adjustment in the absorptivity of the glass envelope. With the adjustment the model has been validated and has been used to analyze the PTS s performance under the various operating and weather conditions. Some recommendations on the PTS s design are then addressed. 4.1 Analytical method First, the experimental data from the test program and equipment parameters from Broad, the PTS suppler, have been used as PTS model inputs to solve the model. The data used include ambient temperature, wind velocity, direct normal solar radiation, incident angle, the inlet temperature of fluid, the flow rate of the fluid, the effective aperture surface of the trough. The model calculated the outlet temperature of fluid, the surface temperatures of absorber pipe and glass envelope, the heat transfer in the receiver pipe, and the efficiency of the PTS. Second, the measured data and model solutions were compared. The model assumptions and input parameters were checked and adjusted to reduce the discrepancies between the calculated and measured data. Third, the validated model was used to analyze the performance of the PTS under the influence of the various weather conditions and operating parameters Model validation The experimental data from 55 steady states, described in hapter 2, were selected to validate the developed PTS model. The model was used to calculate the temperatures throughout the PTS and the heat transfer in the receiver pipe. It was found that the collector efficiency estimated by the model was higher than the experimental data, especially when the system was operated at a high temperature. In addition, the model calculation indicates that the temperature of glass envelope was low, from 20 to 28, and changed little while the operation temperature of fluid increased from ambient to an elevated high temperature. Regarding to this observations, three questions were raised: 61

78 Does the glass temperature greatly impact on the PTS efficiency? Is the temperature of glass envelope really so low? Is the temperature around glass envelope uniform? To answer these questions, additional experiments were designed and conducted to measure the temperature and distribution on glass envelope surface under various fluid operation temperatures. Five spots around the glass envelope and one spot marked as 6 on exposed absorber pipe between PTS sections were measured by the infrared temperature sensor and contact surface temperature sensors, as shown in Figure 4-1. The noncontact infrared temperature sensor was used to measure the temperature of the PTS while it was operating since the concentrated solar radiation significantly affects contact sensors. The range of the infrared sensor is -18 to 315 and accuracy range is ±2% of full range. The contact surface temperature sensors were used only when the PTS was defocused. During the experiments, the PTS tracked the sun and the HTF was circulated through by-pass and heated in the solar collection loop. Six spots were measured by both the infrared and the contact sensors at various HTF temperatures from ambient temperature to 150. There are total five sets of measurements. Figure 4-1 shows one set of measured data by the infrared sensor when the PTS was focusing and defocusing, respectively. The experimental condition and temperature measurements at six spots are shown in Table 4-1. Time: Time: : : Focusing 5 5 defocused Bare tube Bare tube T Fluid = T Fluid = Figure 4-1 Measured temperature distribution of the glass envelope Table 4-1 Glass temperature measurements in the test T_node 1 ( T_node 2 ( T_node 3 ( T_node 4 ( T_node 5 ( T_node 6 ( Type Time T Ambient ( ) T_PTS_in ( ) T_PTS_out ( ) F-1 (gpm) F-1 (kg/h) NIP (w/m2) wind (m/s) ) ) ) ) ) ) 10:58~11: :52~11: Focusing 13:42~13: :27~14: :55~15: Defocused

79 There are moderate differences between the temperatures at bottom, at the top, and on the left side on Figure 4-1; but in general the temperatures around the glass envelope are uniformly distributed. The assumption of the uniform temperature distribution of glass envelope appears justified. In addition, the measurements show that the glass envelops temperature is much higher than the model calculate values of 20 to 28, around 35 to 50. The higher glass temperature will induce greater thermal loss from the glass envelope to the surroundings by means of convection and radiation, so that the overall solar collector efficiency will be reduced. This explains why the calculations estimated a higher efficiency of the solar collector. The experimental data indicated that the calculated glass temperature was not correct. The glass envelope actually absorbs more solar energy than the model predicted and transmits it to the surroundings by convection and radiation as a result of its increased temperature. The analyses of the relationship between the glass temperature and the properties of glass and coating on the absorber pipe showed that the absorptivity of glass is the most critical property impacting on the glass temperature compared to other properties such as transmittance, reflectance, emissivity of glass; and the transmittance, absorptance, reflectance, and emissitivity of absorber pipe. The original value of absorptivity of glass in the model was 0.03, a constant based on the information from the manufacture. As Bouhuer s law states, the absorbed radiation is proportional to the local intensity in the medium and the distance the radiation travels in the medium; the absorptivity, transmittance and reflectance of the medium can be approximately calculated by given the extinction coefficient and the index of refraction of the glass envelope for the solar spectrum, when an unpolarized radiation passes through the receiver pipe [Beckman 1980]. The equations are presented that used to adjust the PTS model in Appendix 4.1. The adjustment of the absorptivity of glass minimizes the discrepancy between the experimental data and model calculations. omparison among the solutions calculated by the validated model and experimental data are shown both in Figure 4-2 and Table 4-2. In the Table, the values under the experiment columns are measured data at 55 steady states and the values under the model columns are the calculated values at the corresponding steady states. Tout deviation is the difference between the measured outlet temperature of solar field and calculated outlet temperature value by the model and error % is the ratio of Tout deviation to the difference between measurements of the solar array inlet and outlet and inlet temperature. The table shows that the calculated values are reasonably close to the measured values although error seems higher when the HTF at a high temperature. One of the reasons is that the temperature sensor of solar array inlet or outlet is located before or after the 63

80 flexible hose connections rather than the beginning or the end of the PTS solar arrays. Energy is lost through the piping connections of two arrays and the four flexible hoses. When the HTF is operated at higher temperature, this thermal loss will increase. In addition, sensor accuracy contributes to the deviation between the measured data and model solutions. Also the accuracy of the heat transfer equations used in the model is about 6%~10%. Taking all of these inaccuracies into account, it is reasonable to say that the calculated quantities are in reasonable agreement with the measured data. Table 4-2 omparison between measured values and model calculations SS # T_amb ( ) Wind_speed (m/s) NIP (W/m^2) incident angle Experiment Aa (m^2) solar loop flow F1 (kg/hr) Tin ( ) Tout ( ) Tout-Tin Eta T_glass_ outer ( ) Tout ( ) Eta omparison Q_delivered(k W) Eta_deviation Model 64

81 Measured data alculation PTS's efficiency Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-2 omparison between the measured data and calculation solutions 4.2 Model-based PTS performance analysis The PTS s performance has been analyzed under the various weather and operating conditions. Efficiency plots are addressed in following subsections showing the effects of those parameters: direct normal solar radiation, the incidence angle, the operating temperature, the presence of air in the annular space, flow type, flow rate, the presence of glass cover, and the wind speed Temperature distribution in the receiver pipe The PTS performance model calculates all of temperatures in the glass envelope and the absorber pipe. It is helpful to visualize these temperatures in order to understand the transfer of heat in the PTS. The temperature distribution in a cross section of the receiver pipe, when the HTF is operated at 140, is shown in Figure 4-3. In the Figure, the abscissa is the distance in the radial direction and the ordinate is the temperature. The highest temperature occurs at the outer surface of the absorber pipe. The lowest temperature is that of the ambient air and the sky. The direct normal solar radiation is primarily absorbed at outer surface of absorber pipe. The absorbed solar radiation, as heat, is then conducted through the absorber pipe and convected into the flowing HTF that delivers the thermal energy for use outside of the PTS. The glass envelope remains at a lower temperature and reduces heat transfer from the absorber pipe, so that it 65

82 improves the solar collector efficiency. Along the length of absorber pipe, the heat transfer to the fluid slowly increases its temperature as indicated in Figure T (in ) At a fixed Y Y 140 T At a fixed X X Y 60 Absorber pipe length fluid absorber tube X (mm) sky ambient air ambient air sky glass envelope Figure 4-3 Temperature distribution in the receiver pipe Thermal losses The thermal losses from the PTS to the surroundings include three parts: conduction, convection and radiation from the connections of absorber pipe to the surroundings. convection from the glass envelope to the ambient air. radiation from glass envelope to the sky. Both conduction and convection are proportional to the difference between the average operating temperature and the ambient temperature; and radiation is proportional to the difference in the fourth powers of the temperature. Model calculations, Figure 4-4, project the losses in recovered solar energy by the PTS due to conduction, radiation, convection and optical factors. The Figure shows that three parts of heat transfer are almost equal. In addition, the calculations also predict that both the efficiency of the PTS drops and the thermal loss to the surroundings as the 66

83 difference between the average operating temperature and ambient temperature increases as shown in Figure 4-5,4-5 and onvection from glass to ambient air Percent of energy to total collected SYLTHERM 800 Idn = 900 W/m^2 Incident angle =0 Losses due to optical reasons Loss due to convection from glass to ambient air Loss due to radiation from glass to sky Loss from connection structure between receiver pipes Radiation from glass to sky onvection, radiation, and conduction from connection structure Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-4 Thermal losses through the receiver pipe PTS's efficiency Syltherm 800 Incident angle =0 insolation 1100 W/m^2 insolation 1000 W/m^2 insolation 900 W/m^2 insolation 800 W/m^2 insolation 700 W/m^2 insolation 600 W/m^ Average operation temeprature of fluid above the ambient temperature ( ) Figure 4-5 PTS s efficiency and direct normal solar radiation at 0 incident angle PTS efficiency and solar radiation The model predicts that the useful solar gain from the PTS will increase with solar radiation increasing as indicated in Figure 4-5. For the same operating temperature of HTF, although there 67

84 is no change in the thermal loss from the receiver pipe, the total solar energy impinging on the receiver pipe is enhanced due to the increase of the solar radiation intensity PTS efficiency and incident angle of solar beam The incident angle plays a significant role on solar collector efficiency. The incident angle is defined as an angle between the sun ray and the normal direction of collector aperture surface. This angle causes foreshortening of the collector aperture as well as other effects such as these on transmittance of the glass envelope or the absorption of the selective surface, so that a larger incident angle can greatly reduce the solar collector performance. The end effect of the incident angle is to reduce solar radiation arriving at the receiver pipe. The efficiency of the PTS is basically proportional to the cosine of the incident angle. For a PTS operated at the same temperature, it will have the highest efficiency when the incident angle is zero and a relative lower efficiency when the solar incident angle is large, as illustrated in Figure 4-6. The calculation, for instance, predicts that the efficiency of the PTS will drop 15~20% when the incident angle increases from 0 to 45, as illustrated in Figures 4-5 and S N Syltherm 800 Idn =900W/m^2 PTS's efficiency θ Incident angle = 0 Incident angle = 10 Incident angle = 20 Incident angle = 30 Incident angle = 40 Incident angle = Average operation temeprature of fluid above the ambient temperature ( ) Figure 4-6 PTS s efficiency and incident angle PTS efficiency and wind speed Since the PTS is exposed to the surroundings, wind is one of important factors influencing convection heat transfer from the glass envelope or the PTS connections to the surroundings. The model indicates that the efficiency of the PTS decreases with the increasing of the wind 68

85 velocity as shown in Figure 4-8; the effect of wind on reducing collector performance is greater when the operation temperature of HTF is high. PTS's Efficiency Syltherm 800 Incident Angle =45 insolation 1100 W/m^2 insolation 1000 W/m^2 insolation 900 W/m^2 insolation 800 W/m^2 insolation 700 W/m^2 insolation 600 W/m^ Average operation temeprature of fluid above the ambient temperature ( ) Figure 4-7 PTS s efficiency and direct normal solar radiation at 15 incident angle PTS's efficiency Syltherm 800 Incident Angle =0 I_dn=900 W/m^2 Wind speed = 0 m/s Wind speed = 2 m/s Wind speed = 4 m/s Wind speed = 6 m/s Wind speed =8 m/s Wind speed =10 m/s Wind speed = 12 m/s Average operation temeprature of fluid above the ambient temperature ( ) Figure 4-8 PTS s efficiency and wind speed PTS efficiency and fluid type In order to explore the impact of the fluid type impacts on the PTS performance, six heat transfer fluids -- Syltherm 800, Dowtherm Q, Therminol 59, Therminol 66, Therminol XP, and 69

86 water -- were selected to analyze the PTS performance. All of fluids have almost same performance curve at the same operating conditions as shown in Figure 4-9. The fluid type is not a significant factor in determining PTS efficiency PTS's efficiency Incident Angle =0 I_dn=900 W/m^2 SYLTHERM 800 DOWTHERM Q THERMINOL 59 THERMINOL 66 THERMINOL XP WATER Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-9 PTS s efficiency and fluid type PTS efficiency and flow rate Turbulent flow of the heat transfer fluid must be sustained during operation of the PTS to ensure good heat transfer. Once the requirement of turbulent flow is satisfied, the results of calculations, shown in Figure 4-10, indicate that flow rate does not influence PTS performance PTS efficiency and air in the annular space The installed PTS has an evacuated annular space between the absorber pipe and the glass envelope in order to reduce conduction and convection between them; however, air may leak into the annular space. Thermal losses will increase because of the increasing of conduction and convection in the annular space. The results of the model calculations indicate that air at atmospheric pressure in the annular space results in a 2% decrease in the efficiency of the PTS, at 160, as indicated in Figure If there is a difference between the average operating temperature and the ambient temperature, thermal losses from the PTS to the surroundings always exist whether solar radiation is present or not. Figure 4-12 illustrates that thermal losses of the PTS have a similar shaped curve under sun or no sun condition; thermal losses with the sun are higher than with no sun. When solar radiation impinges on the PTS, the glass envelope 70

87 and the absorber pipe absorb solar radiation and have higher temperatures, so that the heat transferred from them to the surroundings increases PTS's efficiency Syltherm 800 Incident Angle =0 I_dn=900 W/m^2 Flow rate = 4 m3/h Flow rate= 6 m3/h Flow rate = 8 m3/h Flow rate = 10 m3/h Flow rate = 12 m3/h Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-10 PTS s efficiency & flow rate PTS's efficiency Syltherm 800 Incident Angle =0 I_dn=900 W/m^2 Optical losses vacuum Air Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-11 PTS s efficiency and air in the annual space 71

88 Syltherm 800 Incident angle = 0 with sun Heat loss, W/m NO SUN _ VAUUM NO SUN _ AIR SUN (900W/m^2) _ VAUUM SUN (900W/m^2) _ AIR Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-12 Thermal losses with Sun or No-sun PTS efficiency and glass envelope The outer glass envelope is significantly cooler than the absorber pipe. If the glass envelope of the PTS is removed, thermal losses of convection and radiation from the receiver pipe to the surroundings will be greatly increased. Figure 4-13 shows that the solar collector with glass cover has lower optical efficiency than one without glass cover due to the absorptivity of the glass; however, the thermal losses of the PTS without glass cover are much higher PTS's efficiency Syltherm 800 Incident Angle =0 I_dn=900 W/m^2 Optical losses wth glass cover Optical losses without glass cover With glass cover Without glass cover Average operating temperature of fluid above the ambinet temperature ( ) Figure 4-13 PTS s efficiency and glass cover 72

89 In the Figure, the efficiency of solar collector without glass cover drops rapidly along with the increasing of fluid operation temperature; when the fluid operation temperature is above 200, the efficiency of the solar collector without glass cover is less than 10%. With such low efficiency, the solar collector basically cannot collect any energy. Therefore, the glass cover plays a significant role in improving the solar collector efficiency. 4.3 Recommendations on the PTS s design Bellow design According to the experimental data and analyses of heat transfer in the PTS, the thermal loss from the bellows, which include flanges, bare flexible absorber pipe and the bracket as shown in Figure 3-3, is approximately one third of total losses from the PTS to the surroundings. In order to improve the performance of solar collector, measures in the bellow design could be adopted as follows: reduce the exposed area of the bellow reduce the dimension of the bellow contain the expansion piece inside of glass cover Figure 4-14 is an example of the bellow design from SOLEL. Figure 4-14 New bellow design from SOLEL Glass cover In the model validation process, it was found that the absorptance of the glass envelope of solar collector was apparently not as represented. The absorptance, transmittance, and reflectance of

90 the glass envelope play important roles in the collector performance. To improve the solar collector efficiency, the glass envelope material should have following features: high transmittance, at least greater than 0.9. low absorptivity, less than small thickness Diameter of the glass envelope The glass envelope should be designed with as small diameter as possible since the thermal losses from glass envelope increase with heat transfer surface area Diameter of the absorber pipe A perfect parabolic trough reflector and a well aligned absorber pipe ensure that the reflected solar rays impinge on the absorber pipes; however actual installations may not meet the requirement of accurate focusing. A large diameter of absorber pipe may compensate for inaccurate focus. On the other hand, the convection and radiation between the absorber pipe with larger size and the glass envelope will be greater; therefore the diameter of the absorber pipe should be determined by a tradeoff between solar radiation received by the absorber pipe and its thermal losses. 74

91 5 Solar absorption cooling and heating system simulation An overall solar building cooling / heating system performance simulation has been developed to assist in the installed system design: to evaluate the system performance, to optimize the system configuration, and to test guidelines for the design and operation of a variety of similar solar applications. The developed model is able to calculate the detailed system operating conditions under various weather and output conditions and to search for an optimized system, which has the lowest life cycle cost. The objective of modeling the solar driven cooling and heating system is to investigate the effectiveness of the solar driven cooling and heating system for a building: its ability to maintain comfort conditions in a building over the range of weather and occupancy conditions given the system configuration and equipment capabilities validate the system operating rules and control to adjust properly the system depending on the weather conditions and the building loads quantify the performance of the solar thermal system: the useful energy provided by the solar receivers explore the effects on performance of key system design parameters (receiver area, storage tank volume, and pipe size) and of key operating parameters (operational strategies) develop a tool and a technique for the synthesis and analysis of energy supply systems for buildings: selecting and optimizing systems configurations, equipment selection and sizing, and operations specifications The developed system simulations are based on the configuration of the solar system installed in the IW. The base case models have been evaluated by the experimental data. The annual performance of the solar cooling and heating system was calculated to estimate what fraction of the energy required operating the system could be provided by solar energy. Based on the base case, the system optimization and sensitivity analysis have been carried out, so that an optimized system was defined and system guidelines in design and operation were found for the similar applications. 75

92 5.1 Model approach The software selected to model the solar thermal system is Trnsys, developed by Solar Energy Laboratory of the University Wisconsin. It is a flexible tool designed to simulate the transient performance of thermal energy systems. Trnsys supports detailed simulations of multi zone buildings and their energy supply systems. It has the capability of modeling and interconnecting system components, of solving the corresponding differential equations, and of facilitating information output. The entire problem of the system simulation reduces to a problem of identifying all the components that comprise the particular system and formulating a general mathematical description of each. To supplement the available components in the Trnsys library, users also can use a basic component format to model their own new components either as a performance table or as a FORTRAN program. The approach used for the simulation was the predictive method, which first calculates the building cooling and heating loads and then calculates the required energy input to the cooling and heating system to meet the loads, either from available solar radiation or from other energy sources. So the developed system simulation includes two major parts: the building simulation and energy supply system (ESS) simulation. The information flow of annual system simulation is shown in Figure 5-1. The building simulation considers the configuration of the building; weather conditions; the schedules for occupancy, lighting, equipment; set points for temperature and humidity; and its conditioned fresh air, which is provided by a ventilation unit the Semco. (The solar cooling and heating system does not provide energy to condition fresh air by the Semco unit.) The output quantifies building thermal condition and building sensible loads. The energy supply system simulation consists of two loops as also shown in Figure 5-1: solar collection loop and load loop. Most of the major equipment components in the solar system are available in the Trnsys component library. The PTS component has been represented by a modified linear parabolic concentrator component available in the library of Trnsys. In addition, one new component, the overall system control, was written to integrate controls for the PTS, absorption chiller, pumps, and other devices. 76

93 Figure 5-1 Information flow of TRNSYS simulation 5.2 Model assumptions Weather The developed overall system simulation uses the typical meteorological year (TMY2) data to obtain the weather condition and geographic information for Pittsburgh (latitude N, longitude ). The TMY2 data sets were values of solar radiation and meteorological elements for a one-year period derived from the National Solar Radiation Data Base (NSRDB). They represent typical rather than extreme conditions [22]. Figure 5-2 shows the monthly outside air temperature in Pittsburgh. Figure 5-3, 5-4 are the solar radiation conditions in Pittsburgh. Figure 5-4 shows that the direct normal solar radiation has relative stable intensity throughout a year. omparing solar radiation in January and July, as shown in Figure 5-3, the direct normal solar radiation is higher in the winter than in summer. The cause of the low direct 77

94 normal solar radiation in summer is the humidity of the Pittsburgh summer; a large amount of direct normal solar radiation is diffused before it reaches the earth. In the winter, it is dry and chilled so that the intensity of direct normal solar radiation is relatively higher, sometimes up to 1000 W/m 2. Overall, the daily average solar radiation throughout a year in Pittsburgh is not higher than 420 W/m 2. Pittsburgh is not an ideal place for a PTS based application Dry bulb temperature in Month Mean_Temp Min_Temp Max_Temp Figure 5-2 Monthly average dry bulb temperature of Pittsburgh Figure 5-3 Direct normal solar radiation in Pittsburgh 78

95 1.20 Global solar radiation on ground Direct normal solar radiation Solar flux in kw/m Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Figure 5-4 Daily average solar radiation throughout a year in Pittsburgh Assumptions in the model of solar energy supply system Based on the IW solar cooling and heating system, the simulated energy supply system primarily consists of PTS s and a double effect absorption chiller. An aqueous solution containing 50% propylene glycol is the heat transfer fluid in the solar collection loop. Figure 5-5 and 5-6 are the Trnsys information flow diagram of the base cases for solar cooling and heating system modeling, respectively. Since installed absorption chiller is dual fired, two absorption chillers were applied in the solar cooling simulation: hot water driven absorption chiller and direct fired absorption chiller. These two chillers have been configured in parallel, and either of them provides the chilled water for the space cooling at a given instant. The solar heating base case uses the heat exchanger as heating device since the experimental data indicates that the heat exchanger based solar heating has better system performance than the absorption chiller based solar heating, as discussed in hapter 2. When solar energy is not available or adequate, an auxiliary electrical heater is used in the solar heating system. In the Trnsys solar energy system model, the heating and cooling devices were simplified by using LOAD component, which imposes a user specified load on a flow stream and calculates the resultant outlet fluid conditions. This simplification does not impact on the overall solar system 79

96 performance computations. In the collection loop we watch solar input with load by defocus at 95. Type 709 Return pipe Type 11 Flow diverter Type 11 Flow mixer Type 230 BROAD PTS Hot-water abs. chiller Direct fired abs. chiller Type 682 Heating load Type 677 HW abs. chiller Type 678 DF abs. chiller Type 709 Supply pipe Type 60 Storage tank for heat capacity Type 114 Pump S-1 Type 11 Flow mixer Type 11 Flow diverter Type 114 Pump S-4 Type 709 Load pipe The solar collection loop The load loop Figure 5-5 TRNSYS information flow diagram of solar cooling base case Type 709 Return pipe Type 11 Flow mixer Type 6 Heater Type 230 BROAD PTS Type 5 Heat exchanger HX-2 Type 682 Heating load Type 709 Supply pipe Type 60 Storage tank for heat capacity Type 114 Pump S-5 Type 11 Flow diverter Type 114 Pump S-4 Type 709 Load pipe The solar collection loop The load loop Figure 5-6 TRNSYS information flow diagram of solar heating base case Heat transfer fluid, HTF Although water is an economical and thermodynamically superior heat transfer fluid, it processes a relatively high freezing point and also promotes corrosion. Based on the lowest temperature, - 80

97 28 o [23], in Pittsburgh historic data, a mixture of 50/50 inhibited propylene glycol and water was selected as the heat transfer fluid, HTF, in the solar cooling and heating system installed for the IW because of its freeze protection temperature, -32. In order to prevent vaporization of 50% propylene glycol at 170, an operating pressure of 6 bar was adopted, based on Boiling temperature and pressure of aqueous propylene glycol solutions shown in Figure 5-7. To simplify the problem, the physical- thermal properties of 50% propylene glycol were constant value according to two typical operation temperatures: 150 and 60 for cooling and heating simulations, respectively, even though the properties of fluid are changed with the changes of HTF operational temperature. Figure 5-7 Boiling temperature and pressure of aqueous propylene glycol solutions Piping and insulation Steel pipe was selected for IW solar thermal system due to its 343 o highest working temperature in according with ASHRAE 2000 I hapter 41. In order to reduce pressure drop and thermal loss along 85 m long pipes connecting the PTS s with the chiller, 1 1/4 schedule 40 black iron pipe with 3 inch fiberglass insulation was selected. 81

98 5.3 System components and operation controls omponents in the solar heating base-case The components in the solar simulation based solar cooling and heating system installed in the IW include PTS, heat exchanger, circulation pumps, building heating load and controllers as shown in Figure 5-6. All of components are represented in the Trnsys simulation as indicated in the Figures 5-5 and 5-6. The linear parabolic concentrator, Type 536, in the Trnsys s library can not accurately simulate the performance of the installed PTS, since this component does not consider the influence on the solar collector performance of the end-loss and shadow loss discussed in hapter 3, and also tracking limitation of the installed PTS. Unlike Type 536, linear parabolic concentrator, the PTS installed can only rotate between 70 to -70 as shown in Figure 5-8. A new component, type 230, Broad PTS, was programmed including the endloss, shadow loss, and tracking limitations of the PTS. The performance of installed PTS is based on the efficiency Equation 5-1 generated by means of multiple regression of the experimental data. Figure 5-8 BROAD PTS tracking range η = 0.626IAM 1.47( Tin Tam) / Equation 5-1 I dn Where, IAM = incident angle modifier Tin = inlet temperature of solar field in Tam = ambient temperature in I = direct normal solar radiation in W/m 2 82

99 5.3.2 Operational controls in the solar heating base-case For the heat exchanger based solar system, the solar collection loop does not need to operate at a high temperature as long as the temperature requirement of heating device is met. In order to prevent HTF boiling without pressurization of the system, an upper limit of the solar field outlet temperature is set to 95. When the outlet temperature of solar field reaches this temperature, the PTS is defocused to prevent vaporization. This control feature is included in the Type 230, PTS model. A simple control has proposed in the system based on on/off with hysteresis. Table 5-1 ontrol mode in the base-case simulation of solar heating system Mode Solar collection loop A. On onditions Direct normal solar radiation NIP >300 W/m 2, Differential controller ON Output S-5 S-4 HX-2 Heater ON Off B. Off Load loop A. On W/m^2 NIP<250 W/m^2 There is load from IW Trnsys building model OFF ON ON B. Off No Load OFF OFF Heater A. OFF (T_sr_o: outlet temperature of solar receiver, T_HX-2_in: inlet temperature of water at the cold side of HX-2 T_heater_in:the inlet temperature of the water entering in the heater) B. ON There is load, and T_sr_o > T_HX2 _in +3, Differential controller, hysteresis ON OFF T_heater_in < 45 and There is load ON OFF ON OFF OFF ON There are three components: pump S-5 in the solar collection loop, pump S-4 in the load loop, and electric heater in the load loop, controlled by this simple control. Table 5-1 deals with the control of the pumps and the heater. There is a major circulation pump, S-5, in the solar collection loop. It is a constant frequency pump working with the PTS to collect solar energy when solar radiation is available. According to the experimental experience, the minimum direct normal solar 83

100 radiation to operate solar collection loop is 300 W/m 2, which is required to balance the heat loss in the collection loop. Similar to the collection loop, load loop also has a circulation pump, S-4, controlled by the load requirement. Whenever load is required, it delivers hot water either from heat exchanger or the auxiliary heater for heating. When the outlet temperature of solar collectors is at 3 above the inlet hot water of HX-2 and there is heating required, the heat exchanger will transfer the heat from the collection loop to the load loop for the heating device if useful solar energy is available. The electrical heater is triggered by the inlet temperature of hot water to the heater. If it is less than 45, the heater is turned on to provide the energy for space heating omponents and operation controls in the solar cooling base-case The components in the solar cooling simulation include PTS, hot water driven double effect absorption chiller, direct fired double effect absorption chiller, circulation pumps, space cooling load and controllers as shown in Figure 5-5. Similar to the solar heating base case, the system also includes two major loops: the solar collection loop and the load loop. In the solar collection loop, the PTS works with pump S-1 to deliver solar energy to the absorption chiller when solar radiation is available. In the early operation period, the HTF is continuously circulated and heated through bypass in the solar collection loop until the HTF reaches the temperature required by hot water chiller, 155. The HTF, then, is switched from the bypass to the Regenerator of absorption chiller by a three-way valve. This three-way valve also can be used to adjust HTF flow over the regenerator of the chiller. So the absorption chiller controls the HTF flow based on the chiller load by using the three-way valve and the defocus controls the temperature of the HTF assuming solar input meets or exceeds need. The absorption chiller installed in the IW could be driven either by the solar energy or by the natural gas, but it could not be driven simultaneously by both energy sources. In the real control logic, the switch between two energy sources depends upon the relationship between the temperature of HTF entering into the chiller model of chiller and the chiller load. The available Trnsys chiller model does not provide the relation. To simplify the problem, simulation used the control logical to switch the energy sources as follows. When the HTF from the solar field reaches 155, the chiller starts to use solar energy, and once it drops lower than 135, the chiller is switched to use natural gas.. The control mode of the solar cooling system base case is 84

101 shown in Table 5-2. The detail of components and simulation in solar cooling and heating could be found addressed in Appendix 5. Table 5-2 ontrol mode in the base-case simulation of solar cooling system Mode Solar collection loop A. ON onditions NIP >300 W/m^2, Differential controller, hysteresis ON Output S1 bypass HWchiller Firedchiller ON Off W/m^2 B. OFF Load loop A. ON NIP<250 W/m^2 1.Load >0 kj/h 2.T_sr_o >155 Differential controller, hysteresis OFF OFF ON OFF ON Off B. OFF 140 T_sr_o < ON OFF ON No load ON OFF OFF 5.4 Simulation evaluation It normally takes long time and money to build a simulation model. However, without experimental data to evaluate the simulation, it is difficult to judge the accuracy of a simulation because it embeds so many assumptions; therefore experimental based evaluation is a critical step for any simulation to ensure that reasonable simulation results are provided. The evaluations of solar cooling and heating simulation were carried out by using experimental data on 31 July 2007and 02 March 2007, respectively, address in solar heating and cooling testing described in hapter 2. This evaluation compares the simulation results to the experimental data with respect to the instant operational temperature conditions and energy quantities of the system. In the solar heating experiment on 02 March, the PTSs worked with the HX-2 to generate hot water for heating; the generated hot water was rejected heat through the HX-1, which used chilled 85

102 water at around from grid on the cold side of HX-1. The evaluation of solar heating simulation includes the measured weather data, models of the PTS and two heat exchangers, and the system operation procedure on 02 March. The simulation time period was 1440 minutes and the time step was 1 minute Simulation results T_HTF_HX_in Temperature in T_sr_out 9:30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 T_HX_HW_out THTFHXout T_HX_HW_in Experimental data 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Mar.02, 2007 Figure 5-9 Operation temperature comparison between solar heating evaluation simulation and experiment At the beginning, the simulation results showed that warm-up time of the system was much short than the actual time span although system pipeline was modeled with proper size and thermal loss in the simulation. In checking each of the components in simulation, it was found that the problem was from pipe component. The simulation of the PTS in the system did not consider 86

103 the energy used for heating the fluid and the pipe from the initial temperature to its operating temperature (its thermal heat capacity). To resolve this problem, a thermal storage tank was added and configured like the pipe: a horizontal cylinder with the same diameter as the piping line and same thermal loss coefficient. The pipe component was still retained without thermal loss in the simulation due to its contribution to the relation between the temperature change and time delay. After this modification, the results showed good agreement between the simulation result and the experimental data, as shown in Figures 5-9 and Simulation results Idn*Aa*cos(θ) Power rate in kw Q_solar_delivered Q_HX_solarinput Q_HX_heating :30 9:50 10:10 10:30 10:50 11:10 11:30 11:50 12:10 12:30 12:50 Experimental data 13:10 13:30 13:50 14:10 14:30 14:50 15:10 15:30 15:50 16:10 16:30 16:50 17:10 17:30 Local time on Mar.02,2007 Figure 5-10 Energy flow comparison between solar heating evaluation simulation and experiment In Figure 5-9, the five temperature curves represent the outlet temperature of solar field, the temperature of HTF entering into heat exchanger, the temperature of hot water leaving from the heat exchanger, the temperature of HTF leaving from the heat exchanger, and the temperature of 87

104 water entering into heat exchanger, from top to the bottom of the chart. Figure 5-10 shows the energy flow of the simulation and of the experiment. The curves in the charts show total solar energy rate received by the PTS, the energy transferred to the HTF, the energy input to the HX2, and the heat output of the HX-2 from top to the bottom of chart. Both figures show that the added thermal storage improves the system stability so that the simulation results have smoother curves than the data. The results basically confirm that the simulation well represents the experiment on 02 March Simulation results Temperature in T_sr_out 8:40 9:00 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 Experimental data T_chiller_HTF_in T_chiller_HTF_out T_chiller_HW_return T_chiller_HW_supply 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 17:00 Local time on Jul.31, 2007 Figure 5-11 Operating temperature comparison between cooling evaluation simulation and experiement 88

105 Simulation results Power rate in kw I DN *Aa*cos(θ) Q_chiller_cooling Q_solar_delivered Q_chiller_solarinp 8:40 9:00 9:20 9:40 10:00 10:20 10:40 11:00 11:20 11:40 12:00 12:20 12:40 13:00 13:20 13:40 14:00 14:20 14:40 15:00 15:20 15:40 16:00 16:20 16:40 17:00 Experimental data Local time on Jul.31, 2007 Figure 5-12 Energy flow comparison between solar cooling evaluation simulation and experiment In the solar cooling experiment on 31 July, the PTSs worked with the absorption chiller to generate chilled water for space cooling; the chilled water rejected heat in the HX-1 to hot water at around 35 ~ 40 from grid. The solar cooling evaluation simulation includes the measured weather data, the models of the PTS, two double effect absorption chillers and the system operation procedure on 31 July. The simulation time period was also 1440 minutes and time step was 1 minute. The storage tank simulated system heat capacity was also added in the evaluation cooling model. The results show good agreement between the simulation calculation and the experimental data. Figure 5-11, 5-12 are the comparisons of operational temperatures and energy flow between the simulation and the experiment. Since the simulation has used the chiller performance data provided by the Trnsys model, there is small difference the operational temperature of HTF 89

106 between the simulation result and experiment. However, the simulation still provides a reasonable accurate response of the system and reasonable performance of the solar cooling system. 5.5 Base-case result of solar cooling and heating simulation Building simulation results The solar cooling and heating system installed has been designed for IW south zone, whose net floor area is about 245 m 2. Its average height is about 3.1 m including the raised floor and average height of roof. The IW, an open space, is subdivided by partition walls and furniture in nine offices and one conference space. The IW has horizontal shading on the east and west facades. Its internal loads include lighting, computers, and occupants. The total heat gain of the artificial lighting is 17 W/m2. The equipment heat gain is 100 W per computer (one per person) and 50 W for one printer. It is occupied by 30 faculty members and students throughout the weekdays. Most of them arrive between 9AM and 10AM and leave between 5PM and 8PM. The equipment heat gain is based on the occupancy schedule and the lighting heat gain is based on solar radiation available. The lights are seldom turned on due to the architectural integration of day lighting features (skylight, windows) of the space. Figure 5-13 shows the cooling and heating loads of the IW south zone estimated by Trnsys Q_sensible_cooling Q_dehumidification Q_sensible_heating Q_humidification Load in kwh Month Figure 5-13 IW building heating and cooling load estimated by building simulation 90

107 5.5.2 Solar energy system simulation results Several definitions aid in understanding the expressed results of the Trnsys simulation of the performance of the IW solar cooling / heating system. The solar fraction is the ratio of the building cooling / heating loads shown in Table 5-3 provided by solar energy. The value of solar fraction represents the efficiency of solar cooling and heating system. The higher the solar fraction is, the higher the system performance Solar heating simulation The annual IW solar heating performance in a base case is detailed in Table 5-3. Table 5-3 System performance estimated by IW solar heating system base-case simulation Item Energy (kj) Solar total energy collected by the solar collectors (including dumped solar energy) 18,403,342 Solar energy dumped by the solar collectors (high cut limit temperature is 95 ) 2,933,399 Solar energy delivered to the solar collection loop 15,469,944 Energy added by circulation pump in the solar collection loop 125,874 Energy loss through pipe to the environment in the solar collection loop 14,497,792 Energy transffered to the load loop by HX2 994,196 Energy delivered to the load loop by the auxiliary heater 30,807,898 Energy loss through pipe to the environment in the load loop 1,155,017 Heating load 32,603,345 Energy added by the circulation pump in the load loop 1,697,190 Solar energy delivered to the collection loop / total solar energy collected 84.1% Energy loss through pipe in the collection loop / energy in the collecton loop 93.0% Energy transferred by HX / energy in the collection loop 6.4% Solar fraction 3.8% Only 3.8% of the IW heating load was provided by solar energy, even though the total solar useful energy was about half of total sensible heating load. Much of the solar heat collected 79% was lost to the surroundings from the system; 16% was rejected by defocusing. Pittsburgh has 49% days with solar availability throughout middle of October to the middle of April (182 days) in TMY2 weather data, but simulation results showed that solar energy can be used for space heating only about one or two hours during a day. The reason is that the solar 91

108 availability is out of phase with the IW heating load. The building heating load along with the available solar energy is shown in Figure 5-14 and Figure 5-15, the energy flow profile in the IW for a day in the heating season, illustrated the problem is solar heating of the space Q_useful_solar_kW Q_load(kW) Energy rate in kw /1 1/2 1/3 1/4 1/5 1/6 1/7 1/8 1/ /10 1/11 1/12 1/13 1/14 1/15 1/16 1/17 1/18 1/19 1/20 1/21 1/22 1/23 1/24 1/25 1/26 1/27 1/28 1/29 1/30 1/31 Figure 5-14 Useful solar energy and IW sensible heating load in January Q_useful_solar_kW Q_load(kW) Q_HX2_kW Date Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00 Solar time Figure 5-15 Useful solar energy and IW sensible cooling on 30 December First, much of the required heat flow for building heating occurs when solar energy is not available. Heat storage is required to overcome both this difficulty and the next. Second, when 92

109 solar energy is available it overwhelms that required to heat the space; the PTS reflectors must be defused to moderate the solar flux; solar energy is lost. Third, much heat is lost from the system both during its operation in the day and at night. The lost heat at night might be replaced at the beginning of the day to return the solar heating system to its operating conditions. Reducing the surface area, volume, and heat capacity of the system will reduce this third problem. In both storage and reduced system volume are needed to increase the solar fraction for IW heating above Solar cooling simulation Overall solar cooling performance in the IW calculated; the results are shown in Table 5-4. From the table, the solar fraction in cooling the IW is 12.7%; 76% of the collected useful solar energy by the system is rejected to the surroundings through the piping and solar collectors. Table 5-4 System performance estimated by IW solar cooling system base-case simulation Item Energy (kj) Solar total energy collected by the solar collectors (including dumped solar energy) 26,836,168 Solar energy dumped by the solar collectors (high cut limit temperature is 175 ) 476,504 Solar energy delivered to the solar collection loop 26,359,664 Energy added by circulation pump in the solar collection loop 277,922 Energy loss through pipe to the environment in the solar collection loop 20,242,840 Energy transffered to the load 6,220,705 ooling delivered by the hot water chiller 6,720,545 Energy delivered to the load loop by the gas fired chiller 49,253,322 Energy loss through pipe to the environment in the load loop -1,577,278 ooling load 52,761,101 Energy added by the circulation pump in the load loop 1,634,412 Solar energy delivered to the collection loop / total solar energy collected 98.2% Energy loss through pipe in the collection loop / energy in the collecton loop 76.0% Energy transferred HWchiller / energy in the collection loop 23.4% Solar fraction 12.7% Solar cooling is not well suited to a climate like Pittsburgh, which is humid in the summer with a high latent cooling load. The moisture in the air diffuses the direct normal solar radiation, and 93

110 reduces the solar energy collected by the PTS. According to the simulation results, solar cooling could work for about 69 days over 183 days from middle April to middle of October Q_useful_kW Q_load(kW) Q_Hchiller_chw_kw Power rate in kw /1 8/2 8/3 8/4 8/5 8/6 8/7 8/8 8/ /10 8/11 8/12 8/13 8/14 8/15 8/16 8/17 8/18 8/19 8/20 8/21 8/22 8/23 8/24 8/25 8/26 8/27 8/28 8/29 8/30 8/31Date Figure 5-16 Useful solar energy, cooling load and energy provided by chiller in August Q_useful (kw) Q_load (kw) Q_Hchiller_chw (kw) Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Solar time Figure 5-17 Useful solar energy, cooling load and energy provided by chiller on 09 August The average time span of the chiller using solar energy to generate the chilled water is about three to four hours over six to eight hours of solar time due to a large system heat capacity. A great 94

111 portion of useful solar energy is used to heat the system to the temperature required by the absorption chiller, between 135 to 170 ; the operational time is about three hours per day. In addition, for the high variable weather conditions during Pittsburgh summer, solar cooling is intermittent since absorption chiller is sensitive to the HTF temperature from the PTS s. Figure 5-16 and 5-17 show the system performance of solar cooling monthly and daily, respectively. Figure 5-16 indicates relatively few days that conditions are suitable for solar cooling in August. Figure 5-17 shows that in the morning it took 3-4 hours to heat up the HTF to the operation state. Next chapter shows how reduce the volume of piping or use a drain back system to solve the problem. In general, solar energy meets the load. 95

112 6 Simulation-based design and performance analysis on solar cooling and heating Model based calculations of the overall performance of the solar cooling and heating system installed in the IW have been described in hapter 5. In order to identify optimal system design and operation, performance calculations have been performed to assess the impact of the system configuration, operating conditions, and control strategies on the fraction of IW cooling and heating that might be met with solar energy to investigate the effects of: the orientation of parabolic trough solar collectors, PTS s. the system operation and control strategies the thermal storage provided. the inclusion of a gas fired auxiliary heater for HTF preheating. the pipe length and size in the solar heating loop. The area of PTS s provided. The results of model calculations quantify the effects of these variables on system performance. An optimized system configuration and operating condition has been proposed, and guidelines for the design and operation of a similar solar cooling and heating systems have been provided. 6.1 Orientation of PTS A parabolic trough solar collector tracks the altitude of sun in the sky along only a single axis. onsequently, the perpendicular to the aperture of the collector does not point directly at the sun throughout a day, some of the solar energy that might be collected is lost. The cosine of the angle between this perpendicular and the direct rays of the sun, the tracking angle, represents the fractional loss of solar energy due to the limitation of a singe axis tracking arrangement. As discussed in hapter 3, the tracking angle varies both the time of day and the orientation of the PTS: therefore, this orientation plays important role in the collection of solar energy. To determine the most efficient orientation of the PTS at a given location, twelve orientations were selected to analyze the relation between the orientation of the PTS and the solar energy 96

113 collected. In addition, the relation has been reassessed by considering the tracking limitations of the PTS, such as the Broad PTS as discussed in hapter Orientation of the PTS for increased, effective solar energy recovery The orientation of the PTS refers to the position of its tracking axis. Twelve orientations shown in Figure 6-1 have been selected to examine the relation between the orientation of the PTS and the solar energy collected. In the figure, the due north -south, NS, orientation is 0 degrees; on the right, the figure shows the primarily east - west, EW, orientation of the PTS installed in the IW. Typical simulation results are shown in Figures 6-2 and 6-3. The solar beam radiation, Watts/m 2, throughout a summer and a winter day are plotted for both a NS and an EW orientation of the PTS. Solar beam radiation on a NS oriented PTS has two peak values, one in the morning and the other in the afternoon of a day; an EW oriented PTS has one peak. The beam radiation in the summer day has a higher intensity and a longer time duration than in the winter day. These figures illustrate that in Pittsburgh on a summer day a PTS with a NS orientation can collect one third more solar beam radiation than one with a EW orientation, while on the winter day a NS oriented PTS will receive less than one half of the beam radiation on a EW oriented PTS. (165) (150) (135) (120) (105) N N (180 ) W E (90) (75) (60) (45) (30) (15) S (0) 105 S (0 ) Figure 6-1 Twelve orientations in the simulation If the PTS is not oriented due north or due east, the pattern of solar beam irradiation on the PTS is similar but the peak radiation varies with the angular difference between two orientations. For instance, the PTS installed on IW roof at a 105 orientation has solar energy 97

114 radiation pattern similar to the EW orientation, but the peak of solar beam radiation is delayed slightly due to 15 deviation toward north from due east. For solar cooling applications in Pittsburgh, model calculations show that NS is favored. Figure 6-4 shows that there is 25% more solar radiation available on a NS oriented PTS than on a EW PTS throughout the summer; however, as shown in Figure 6-5 for solar heating, an EW orientation is better than a NS because of 5% more solar radiation available from October to April. For combined solar cooling and heating, Figure 6-6 indicates that a NS oriented PTS receives more solar energy (927.7 kwh/m 2 ) than does an EW PTS (815.3 kwh/m 2 ) throughout the entire year in Pittsburgh, a difference of 14% Beam_0 Beam_90 Beam_105 Beam irradiation on the aperture in W/m Time of Jun.10 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 0:00 Figure 6-2 Orientation and solar beam irradiation on a PTS on 10 June in Pittsburgh Beam_0 Beam_90 Beam_105 Beam irradiation on the aperture in W/m Time of Dec. 02 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 0:00 Figure 6-3 Orientation and solar beam irradiation on a PTS on 2 December in Pittsburgh 98

115 900 Monthly solar beam radiations (kwh/m2) (N/S) 90 (E/W) 105 (IW) Direct normal_kwh/m ooling 0 (N/S) (E/W) (IW) Direct normal_kwh/m Figure 6-4 Orientation and solar beam irradiation on a PTS in summer of Pittsburgh (N/S) Monthly solar beam radiations (kwh/m2) (E/W) 105 (IW) Direct normal_kwh/m Heating 0 (N/S) (E/W) (IW) Direct normal_kwh/m Figure 6-5 Orientation and solar beam irradiation on a PTS in winter of Pittsburgh 99

116 Annual total solar beam radiation availability (kwh/m2) (N/S) (E/W) 105 (IW) Total solar gain Figure 6-6 Annual solar beam irradiation on PTS with different orientations in Pittsburgh Tracking Angle (degree) degree Time of Jun. 21 0:59 1:59 2:59 3:59 4:59 5:59 6:59 7:59 8:59 9:59 10:59 11:59 12:59 13:59 14:59 15:59 16:59 17:59 18:59 19:59 20:59 21:59 22:59 23: TA_0 TA_15 TA_30 TA_45 TA_60 TA_75 TA_90 TA_105 TA_120 TA_135 TA_150 TA_165 Figure 6-7 Tracking angle and orientation of the PTS on 21Jun Tracking Angle (degree) degree Time of Dec :59 1:59 2:59 3:59 4:59 5:59 6:59 7:59 8:59 9:59 10:59 11:59 12:59 13:59 14:59 15:59 16:59 17:59 18:59 19:59 20:59 21:59 22:59 23: TA_0 TA_15 TA_30 TA_45 TA_60 TA_75 TA_90 TA_105 TA_120 TA_135 TA_150 TA_165 Figure 6-8 Tracking angle and orientation of the PTS on 21 Dec 100

117 1200 Monthly solar beam radiations (kwh/m2) IDN_20_90_kWh/m2 IDN_20_105_kWh/m2 IDN_20_0_kWh/m2 Direct normal_kwh/m Total IDN_20_90_kWh/m IDN_20_105_kWh/m IDN_20_0_kWh/m Direct normal_kwh/m Figure 6-9 Solar beam irradiation and orientation of the PTS in Pittsburgh Orientation, tracking limitation, and solar beam irradiation on the PTS Due to a difficulty in its structure or drive mechanism, the reflector of a PTS may not be able to rotate over the full range required to follow the altitude of the sun. The range of the Broad PTS installed in the IW is addressed in hapter 5. Its tracking limitation reduces the solar collection time and the amount of the solar beam radiation collected. Analyses of the collection time and solar radiation availability have been carried out for this PTS installed in the IW. Model based simulation results for the required tracking angle shown in Figure 6-7 and 6-8 indicate that the tracking period for this PTS is reduced by about 40% in the typical winter day and about 25% in the typical summer day. The effect of this reduction in tracking time on the annual total solar energy collected by the IW PTS in its EW orientation is a 2.2% reduction; a NS orientation would have experienced a 13% in solar energy collected due to this tracking limitation and solar collection time reduction. In addition, the simulation indicated that the impact from the orientation on solar beam irradiation is on longer significant with this tracking limitation. In general, NS is the favored orientation of a PTS for the solar cooling or the combination of solar cooling and heating system. For the Broad PTS with its tracking limitation, the orientation does not seem to be a affect greatly the solar beam radiation collected. 101

118 In general, N/S is the favorite orientation of the PTS for the solar cooling or solar cooling and heating system. For the PTS with tracking limitation, the orientation may not be critical factor effect on the solar beam irradiation on the PTS s aperture Orientation and overall system performance Based on the system configuration of base cases, the overall system performance has been analyzed how orientation of the PTS effect on building cooling and heating system, the results show that NS orientated solar system has better performance in summer and worse in winter than the current orientated PTS. For solar cooling, NS orientation could provide 6% more energy for cooling than EW orientation; for solar heating, the difference on the solar fraction between NS orientation and EW is very small, as shown in Table 6-1. In addition, simulation indicated that the performance of the system with tracking limitation is almost same as the system without tracking angle limitation. When the PTS cannot track the sun, due to the cosine effect on solar radiation intensity is relative low, less than 300 W/m2, under which thermal loss is same as solar gain, the final effect of collected solar energy by the PTS is almost same for with and without tracking angle limitations. Therefore, NS orientation of the PTS shall be the favorable choice for solar cooling dominated system, to achieve a high system performance. Table 6-1 Effect of PTS s orientation on overall system performance Base case (105 ) ooling season N/S orientation (0 ) Base case (105 ) Heating season N/S orientation (0 ) Solar fraction 12.7% 18.7% 3.8% 2.0% ollected solar energy available to HTF (kj) 26,836,168 32,393,246 18,403,342 13,338,496 Dumped solar energy to avoid overheating(kj) 476, ,892 2,933, ,380 HTF tramsmitted to the chiller / HX (kj) 6,220,705 9,855, , , System operation and control onstant-flow or constant-outlet temperature control of the PTS The flow rate of the heat transfer fluid, the HTF, through the PTS and the solar collection loop is a significant operating condition of the solar cooling and heating system. Two alternatives to setting this HTF flow are: a constant flow or a flow varied to maintain a constant outlet temperature of the HTF from the PTS. onstant HTF flow control is common in flat plate 102

119 collector systems to minimize the costs of equipment and in solar collectors used in supplemental, preheat systems. ollectors with HTF temperature control are used in high temperature systems where excessive temperatures can degrade the HTF or cause high system pressure resulting in damage [Stine 1985]. These two control alternatives: constant HTF flow and constant HTF outlet temperature have been simulated in the solar cooling and heating model to compare their system performances. In this comparison, the HTF flow in the solar loop is adjusted between a maximum flow of 80 gpm determined by the pump capacity and a minimum flow required to maintain well developed turbulent flow in the solar loop, a Reynolds number of The outlet HTF temperature from the PTS is set at 75 for space heating, and 156 for solar cooling in order to operate the absorption chiller effectively. The model based results show that there is little difference in system performance between these two alternatives in solar heating; but in solar cooling, constant HTF outlet temperature control improves the solar fraction attained by factor 2 as indicated in Table 6-2. This performance improvement is due to the longer time period when the absorption chiller directly uses solar energy to generate the chilled water as shown in Figure This figure indicates that the constant outlet temperature control shortens the preheat time prior to chiller operation for about one hour, so that more useful solar energy is directly used for cooling rather than for preheating the HTF in the solar loop. Figure 6-11 shows that the HTF temperature is not uniform throughout the system with constant outlet temperature control. Table 6-2 Effect of flow controls on overall system performance onstant flow (base case) ooling season onstant temperature onstant flow (base case) Heating season onstant temperature Solar fraction 12.7% 21.8% 3.8% 3.6% ollected solar energy available to HTF (kj) 26,836,168 29,246,233 18,403,342 19,229,768 Dumped solar energy to avoid overheating(kj) 476, ,716 2,933,399 5,234,940 HTF tramsmitted to the chiller / HX (kj) 6,220,705 10,504, , ,

120 Q_useful_onsFR Q_Hchiller_chw_onsFR Q_useful_onsT Q_Hchiller_chw_onsT Q_load Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Time on Aug.09 Figure 6-10 System performance comparison of alternate controls on 9 August 2007 When the HTF outlet temperature from the PTS s reaches the desired value for chiller operation, the HTF inlet temperature entering the PTS is at a lower value. This reduced HTF temperature in the return pipeline indicates that less solar energy has been used in heating the system and fluid at the time the chiller operation is initiated for generating chilled water. The operating period for the chiller is in this way extended. In solar heating, however, the IW heat load is out of phase with the availability of solar energy availability, the reduction of system preheat time does not increase the operational period of IW heating. Table 6-2 provides a quantitative comparison of system performance between the two HTF flow control alternatives for solar cooling and heating in the IW. Temperature in :00 1:00 T_sr_in_onsFR T_sr_out_onsFR T_ambient T_sr_in_onsT T_sr_out_onsT 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17: :00 19:00 20:00 21:00 22:00 23:00 Time on Aug.09 Figure 6-11 System operating temperature comparison of alternate controls on 9 August 2007

121 In general, constant outlet temperature control of HTF flow is recommended for solar cooling systems with a large heat capacity; this approach improves the system performance in solar cooling by reducing the preheat time of the system. 6.3 Storage tank requirements Thermal energy storage is employed in solar cooling and heating systems primarily to shift excess solar energy recovered during periods of high solar availability to periods of reduced solar availability. In addition, a small amount of energy storage, called buffer storage, can be provided to smooth irregularities in the solar supply and building loads. Without storage, some of the solar thermal energy available from the PTS s may have to be discarded and therefore may not be available to provide the energy for the building cooling and heating loads, because the time - rate profiles of incident solar energy supply may not coincide with the profiles of the building cooling and heating loads. The appropriate amounts of PTS capacity and thermal energy storage depend primarily on these time profiles. If thermal storage is installed in a solar cooling/heating system, there are two alternative locations for the storage: in the solar collection loop or in the load loop. There are also alternate choices for the storage media: the HTF in the collection loop or the fluid circulated in the load loop or a separate phase change medium. The choice among these alternates depends on the volume, weight, cost, and energy loss, of the required storage; on the capacities of the chiller and heat recovery exchanger components of the system, and on the flexibility required in the operation of the overall cooling/heating system, particularly in the choice of operating temperatures. Based on all these factors, thermal storage in HTF in the solar collection loop of the IW cooling/heating system has been selected for model simulations to calculate the performance benefits of storage. Ordinarily, a solar collection loop is operated at as low a temperature and pressure as is compatible with the required cooling/heating components of the system. In order to reduce the volume and weight of the storage, however, the operating temperature might be elevated to increase the amount of sensible heat available from the stored HTF as it delivers heat to the cooling/heating system components at the required temperature. The appropriate operating temperature and pressure for the solar collection loop including its storage depends both on the heat losses from the PTS s and the system and on the temperature limitations of equipment and materials comprising the system. 105

122 6.3.1 The volume of the storage tank As mentioned above, the proper amount of thermal storage is determined by the time - rate profiles of the building cooling and heating loads and of the solar incident radiant solar energy. But also simple empirical rules can be applied to provide a preliminary estimate of thermal storage capacity to be included in the solar cooling/heating system of the IW. A simple empirical rule suggests that the most practical storage capacity for a solar thermal energy system used for space heating with water as the HTF is approximately 1-2 gal of water per square foot of collector area [19]. Based on this rule, the storage tank volume for the IW solar energy system would be 3.2 m 3 as indicated in Equation V _ st = 1.5( gal / ft )* Aa Equation 6-1 = 1.5 (gal/ft2 collector area) * (ft2)*10.76 (ft2/m2) = 846 gal = 3.2 m3 Where, Aa =aperture area of solar collector, ft 2 The capacity of the storage tank might also be calculated from the solar energy to be stored, 100 kwh based on simulation results, and the temperature difference between the heat storage and the heat delivery temperatures of the HTF storage media, 90 and 60. The volume of the storage is estimated in this way is 3.19 m 3 according to Equation 6-2. V _ st = Q ρ * p * T Equation 6-2 = 100kWhr *3600sec/ hr 3 o o ( kg / m )*3.67( kj / kg. )*(90 60)( ) Where, = 3.19 m 3 Q = daily solar energy input from simulation result, in kwh/hr ρ = density of the HTF at 35 in kg/m 3 106

123 p = specific heat of HTF at 35 in kj/kg T = the difference between storage and the delivery temperatures Model based simulations of the annual performance of the IW solar cooling/heating systems incorporating HTF thermal storage volumes of 0.5 m 3 to 12 m 3 have been carried out to estimate quantitatively the effects of thermal storage Storage used for shifting energy for later use in solar heating In the Trnsys model of the IW solar cooling/heating system, a storage tank was placed parallel to the by pass and a pump was added for discharging the storage tank, as indicated in Figure 6-12, the information flow diagram of the system. In the morning, the HTF is heated in the PTS s and circulated through the by pass until the desired operating temperature, 90 o, is reached. Then the HTF is directed through the storage tank to charge it. If there is heating demand, the storage tank is discharged by pump S-6 until the solar energy stored in the tank is exhausted. The system control is addressed in Appendix 6. The simulation results show that the storage tank significantly improves the overall system performance of solar heating system. The storage tank shifts the excess solar energy during time of high solar availability to time of low solar availability. Due to the time lag between heating demand and solar availability, the excess solar energy is collected and stored during the day and used in the evening and the next morning, as shown in Figure Type 709 Return pipe Type 11 Flow diverter Type 6 Heater Type 230 BROAD PTS Type 534 Storage tank Type 682 Heating load Storage Tank HX-2 Type 5 Heat exchanger Type 709 Supply pipe Type 60 Storage tank for heat capacity Type 114 Pump S-5 Type 11 Flow mixer Type 114 Pump S-6 Type 114 Pump S-4 The solar collection loop The load loop Figure 6-12 Trnsys information flow diagram of solar heating system with storage 107

124 In addition, the simulation results show that for the currently installed IW solar heating system, a 4.5 m 3 storage tank is the optimal volume because of its highest quantity of solar energy used, as indicated in Table 6-3. Incrementally larger storage volumes lose more heat than they store Q_useful_solar_kW Q_load(kW) Q_HX2_kW Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 0:00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Solar time Figure 6-13 Solar energy collected, heating load, and energy provided on 14, 15 November Storage used for shifting energy for later use in solar cooling In the Trnsys model of the IW solar cooling system, a storage tank was located in parallel with and between the by pass and the absorption chiller as shown in the information flow diagram, Figure A variable frequency pump was added to discharge the storage tank into the chiller. Table 6-3 Effect of storage volume on solar heating system performance Volume of tank (m3) ollected solar energy available to HTF (kj) Dumped solar energy to avoid overheating (kj) HTF transmitted to HX (kj) V ( m3) solar fraction Base case % 18,403,342 2,933, , % 18,505,478 1,663,272 3,388, % 18,752,238 1,081,677 4,556, % 18,882, ,298 5,229, % 19,020, ,389 5,746, % 19,147, ,060 6,144,554 Storage tank Heating % 19,265, ,209 6,404, % 19,373, ,974 6,609, % 19,464, ,748 6,722, % 19,559, ,320 6,730, % 19,633, ,677 6,685, % 19,804,859 60,026 6,678, % 20,045,108 1,664 6,430, % 20,409, ,623,

125 In the morning, the HTF was circulated through the by pass and heated until the operating temperature of chiller, 160 o, was reached. The PTS s were then operated at constant outlet temperature, and the HTF was diverted through the regenerator of the absorption chiller. If the flow from the solar field was greater than the flow required by the chiller, the storage tank was charged. When the flow from the solar field was not adequate for the chiller and the HTF stored in the storage tank was at the operating temperature, the storage tank was discharged until it was exhausted. The simulation detail can be found in Appendix 6. Type 709 Return pipe Type 11 Flow diverter Type 11 Flow mixer Type 11 Flow mixer Type 236 PTS combined with VFD pump Type 677 HW abs. chiller Type 534 Storage tank Storage Tank Hot-water abs. chiller Direct fired abs. chiller Type 682 Heating load Type 110 VFD pump Type 678 DF abs. chiller Type 709 Supply pipe Type 60 Storage tank for heat capacity Type 11 Flow mixer Type 11 Flow diverter Type 11 Flow diverter Type 114 Pump S-4 The solar collection loop The load loop Figure 6-14 Trnsys information flow diagram of solar cooling system with storage for shifting energy Temperature in :00 1:00 T_sr_in_onsFR T_sr_out_onsFR T_ambient T_sr_in_onsT T_sr_out_onsT T_sr_in_onsT_0.5Tank T_sr_out_onsT_0.5Tank 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time on Aug.09 Figure 6-15 Operating temperature of solar cooling system with and without storage on 9 August The simulation results indicate that the storage tank did not improve the performance of the system operated at a constant outlet temperature from the PTS s. The reason is that the large collection loop volume of the IW solar cooling/heating system provides the storage required by 109

126 the system in cooling and that additional storage is not required. It merely results in additional heat loss. Figure 6-15 shows that the HTF in the system with storage flows into the PTS at a lower temperature than the system without storage. This circumstance indicates that heated HTF is being stored in the tank and replaced by cooler HTF flowing from it. In addition, the model solar cooling system results also show that the optimal volume of storage tank for the IW solar cooling system is less 0.5 m 3, as indicated in Table 6-4 A large storage tank does not improve the system performance, increasing the usage of the solar energy. The excess solar energy is limited. On the other hand, additional storage tank volume, and area, increases thermal losses. Table 6-4 Effect of storage volume on solar cooling system performance Volume of tank (m3) ollected solar energy available to HTF (kj) Dumped solar energy to avoid overheating(kj) HTF tramsmitted to the chiller (kj) V ( m3) solar fraction Base case 0 (ons_fr) 12.74% 26,836, ,504 6,720,545 0 (ons_t) 21.77% 29,246, ,716 10,504, % 30,523, ,368,464 Storage tank for shifting with constatn temperature control ooling % 30,570, ,293, % 30,574, ,276, % 30,576, ,276, % 30,576, ,276, % 30,576, ,276, % 30,576, ,276, % 30,576, ,276, % 30,577, ,276, Storage used for preheating The solar cooling/heating system installed in the IW requires a long time for heat up due to its large heat capacity. It takes three or more hours to heat the system to the temperature required to operate the absorption chiller. As mentioned before, the storage tank can be used to minimize the time required for preheating. At the end of the day heated HTF in the solar collection loop is directed to the storage tank, replaced by cool HTF fluid from the tank. The heated fluid is stored overnight in the tank whose heat loss is significantly lower that that of the loop. In the next morning HTF in the storage tank, still hot, is returned to the loop, replaced by the cool HTF from loop. This procedure limits the overnight heat loss from the system to that from the structure of the solar loop and from the reduced losses from the HTF stored in the tank. Based on this idea, the system with storage tank used for preheating has been simulated to estimate the performance improvement of system performance due to this procedure. 110

127 Two assumptions made in the system simulation: charging or discharging the storage tank occurs in a short time period. the storage tank is only used for preheating not for shifting the useful energy. Figure 6-16 show the Trnsys information flow diagram for the model simulation. The storage tank remains in parallel with the by pass, but an additional by pass and a pump, S-7, were added for charging the storage tank when the PTS s were stowed at the end of the day. The storage tank is discharged into the solar collection loop at the beginning of the next day. The solar collection loop is then operated under constant outlet temperature control. The details of the simulation are presented in Appendix 6. Type 11 Flow mixer Type 709 Return pipe Type 11 Flow diverter Type 11 Flow diverter Type 11 Flow mixer Type 236 PTS combined with VFD pump Type 677 HW abs. chiller Type 534 Storage tank Storage Tank Hot-water abs. chiller Direct fired abs. chiller Type 682 Heating load Type 114 Pump S-7 Type 678 DF abs. chiller Type 11 Flow diverter Type 60 Storage tank for Type 709 heat capacity Supply pipe Type 11 Flow mixer Type 11 Flow mixer Type 11 Flow diverter Type 114 Pump S-4 The solar collection loop The load loop Figure 6-16 Trnsys information flow diagram for solar cooling with storage for preheating Table 6-5 Effect of Preheat storage tank volume on solar cooling performance Volume of tank (m3) Dumped solar energy to avoid overheating(kj) HTF tramsmitted to the chiller (kj) V ( m3) solar fraction ollected solar energy available to HTF (kj) Base case 0 (ons_fr) 12.74% 26,836, ,504 6,720,545 0 (ons_t) 21.77% 29,246, ,716 10,504, % 30,253, ,562 9,617,850 Storage tank with constatn temperature control for preheating ooling % 30,242, ,377 9,589, % 30,487, ,890 9,448, % 31,064, ,647 9,314, % 31,631, ,270 9,084, % 32,225, ,118 8,629, % 32,874, ,295 8,202, % 33,418, ,484 7,707, % 34,667, ,368 6,703, % 37,258, ,354 3,907,

128 The simulation results showed that the storage tank used for preheating did not improve the system performance as much as expected, as shown in Table 6-5. The major reason is that the mixing of the HTF from the solar collection loop and storage tank is not able to avoid during charging and discharging process. However, a drain-back system could work well because it can prevent mixing and fulfill the function of the replacement between the cold fluid and hot fluid. 6.4 Auxiliary heater for preheating in the solar collection loop An auxiliary heater might be used to heat rapidly the HTF to the required operating temperature in order to minimize the warm up time at the beginning of the day. A gas fired preheater was added in the solar collection loop prior to the PTS s in the Trnsys model of the IW solar cooling/heating system, and performance calculations were carried out. The simulation and its results are presented in Appendix 6. These results show that the auxiliary heater improves significantly, the fraction of the IW cooling/heating loads carried by solar energy. For solar heating, if the HTF set temperature is 50, the overall solar fraction is improved from 3.8% to 7.1%; for solar cooling, if the heater set temperature is 120, from 12.7% to 30.4%. The energy quantities and the operational temperatures of solar cooling system with a heater for preheating are shown in Figure 6-17 and 6-18, respectively. These figures compare the system without a heater using constant-outlet-temperature control. Figure 6-17 indicate that the chiller starts to generate chilled water using solar energy one hour earlier than one in the system without a heater. The use of a gas fired heater substitutes heat from natural gas for solar heat in HTF preheating. It is not clear that this is a desirable substitution. In addition, the capital and operating cost of the system will be increased when an auxiliary heater is added to the system. A decision on this feature requires further analysis. According to the effect of the auxiliary heater for preheating on the system performance, a drainback system for the HTF in the solar collection loop could be a reasonable design to reduce the night time heat loss from the system and the warm-up in the next morning. Almost providing same function as auxiliary heater here, instead of using heater, a drain-back system could drain all of HTF from the solar collection loop at about 130 (cooling) / 50 (heating) in the well insulated 112

129 drain-back tank and use this hot fluid in the next morning. It has about 4-9% potential improvement in solar fraction for both solar heating and solar cooling Q_useful_heater Q_Hchiller_chw_heater Q_useful_onsT Q_Hchiller_chw_onsT Q_load Q_dumped_heater Q_dumped_constT Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Time on Aug.09 Figure 6-17 Effect of a heater on energy flow for solar cooling on 09 August Temperature in :00 1:00 T_sr_in_heater T_sr_out_heater T_ambient T_sr_in_onsT T_sr_out_onsT 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Time on Aug.09 Figure 6-18 Effect of a heater on operating temperature for solar cooling on 9 August 6.5 The length and diameter of collection loop pipe and solar system performance The length and diameter size of the pipe comprising the solar collection loop play an important part in the system performance. Greater pipe lengths and diameters increase the heat loss from the system and increase also the overall heat capacity of the system and of the HTF it contains. Both of these factors increase the amount of time and the quantity of energy consumed in preheating the system to its operating temperature at the beginning of the day. Greater pipe 113

130 lengths also increase the pressure loss in circulating the HTF, while greater diameters reduce these losses. In general, reducing the length and diameter of the loop pipes is an effective means to improve the solar cooling/heating system performance at least until pressure loss and pump energy for HTF circulation in the collection loop become appreciable. The IW solar cooling/heating system Trnsys model has been used to predict the system performance based on the HTF volume of the solar collection loop. A loop with one fourth of the volume of the current IW collection loop, for instance, would almost double the heating energy provided by solar energy as indicated in Table 6-6. With a 4 m 3 HTF thermal storage tank, the system could provide almost 36% of the IW s heating load. Further decreases in IW collection loop volume are limited by the HTF volume in the 52 m 2 Broad PTS s currently installed. Table 6-6 Effect of collection loop volume on solar heating system performance Base case Pipeline 1/4V pipe + Storage tank Volume solar fraction ollected solar energy available to HTF (kj) Heating Dumped solar energy to avoid overheating (kj) HTF transmitted to HX (kj) V ( m3) 105m long, OD=0.35m 3.84% 18,403,342 2,933, ,196 1/2 V 5.58% 16,942,898 6,856,225 1,558,454 1/3 V 6.55% 16,389,211 8,701,500 1,876,762 1/4 V 6.86% 16,194,339 9,385,102 1,974, % 16,626,052 6,622,184 5,341, % 16,893,591 5,261,024 7,355, % 17,235,390 3,763,539 9,656, % 17,465,762 2,864,593 10,871, % 17,713,285 2,323,940 11,451, % 17,965,478 1,871,082 11,588, % 18,143,975 1,439,610 11,619, % 18,435, ,945 11,314, % 18,971, ,458 10,270,337 Table 6-7 Effect of collection loop volume on solar cooling system performance Base case onstant FR+Pipeline Volume of tank (m3) ollected solar energy available to HTF (kj) ooling Dumped solar energy to avoid overheating(kj) HTF tramsmitted to the chiller (kj) V ( m3) solar fraction 105m long, OD=0.35m 12.74% 26,836, ,504 6,720,545 1/2 V 18.81% 22,431,818 1,670,502 9,210,723 1/3 V 20.82% 20,865,671 2,217,887 10,232,847 1/4 V 22.37% 19,802,927 2,639,001 11,030,277 constant temperature control 0 (ons_t) 21.77% 29,246, ,716 10,504,396 1/2 V 26.72% 25,622,199 2,770,240 12,748,661 onstant T+Pipeline 1/3 V 28.66% 24,181,580 3,569,758 13,605,490 1/4 V 29.86% 23,531,796 4,004,222 14,153, % 25,226, ,845 14,534,410 1/4V pipe + Storage tank % 25,302, ,374 14,398, % 25,397,568 38,389 14,118,

131 Reducing the HTF volume of IW solar collection loop also significantly improves system performance as indicated in Table 6-7 and Figure Again, a loop with one fourth of the volume of the current IW collection loop would almost double the energy provided to the chiller by solar energy as indicated in the table. Table 6-7 also indicates that a storage tank for shifting energy or for preheating does not significantly improve the system performance when the collection loop has a small volume. And finally, if the system were operated at a constant outlet temperature, as illustrated in Figure 6-19, it could provide 30% of the IW s cooling. Figures 6-20 and 6-21 show the energy quantities and operating temperatures in the system with a reduced collection loop volume. 35% 30% 1/4 V 1/3 V 1/2 V 25% V 20% onstant FR onstant T 15% 10% 0 1/5 2/5 3/5 4/ /5 Percent of the current pipe volume Figure 6-19 Solar fraction and pipe size under two control strategies Q_useful_onsT Q_Hchiller_chw_onsT Q_load Q_useful_onsT_1/4V Q_Hchiller_chw_onsT_1/4V Energy rate in kw :00 3:00 6:00 9:00 12:00 15:00 18:00 21:00 Time on Aug.09 Figure 6-20 System energy performance and pipe size on 9 August 115

132 Temperature in T_ambient T_sr_in_onsT T_sr_out_onsT T_sr_in_onsT_1/4v T_sr_out_onsT_1/4v 0:00 1:00 2:00 3:00 4:00 5:00 6:00 7:00 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 Figure 6-21 Operating temperature and pipe size on 9 August Time on Aug.09 Figure 6-20 showed that system with ¼ volume of installed pipe operates the absorption chiller by solar energy an hour earlier than current installation controlled by constant-outlet-temperature since the system with a small heat capacity heats up rapidly. In addition, the instant loss from the system to the ambient during preheat period is grater than one in the system with a large volume of pipe. Both Figure 6-20 and Figure 6-21 indicated that a small volume of the system has more useful solar energy utilized to provide building heating and cooling. 6.6 The area of solar collector and storage tank The area of the PTS s and the capacity of the thermal storage provided are important factors impacting the performance of a solar cooling/heating system. Increasing the area of the collectors and the capacity of the storage increase the quantity of solar heat made available for cooling/heating, but also increase the cost of the installation; an economic decision is required. The Trnsys model of the IW solar cooling/heating system has been used to calculate the performance of the system with 52 m 2 and with 81m 2 of PTS s. Figure 6-22 shows that the solar fraction has been improved by 10 % in solar cooling and 15 % in solar heating of the IW. In the Figure, in both solar cooling and heating is achieved by increasing the solar collector area from 52 m 2 to 81 m 2. In addition, the figure shows that the storage tank does not improve the system performance in solar cooling although it may be useful in minimizing fluctuations in system operation. In solar 116

133 heating, the greater PTS area collects more solar energy in the relatively short solar day in winter. And the storage tank makes the solar energy available when it is needed in the evening, at night, and in the early morning. Figure 6-22 also indicates that the larger area of solar collector requires a larger the storage tank to achieve the maximum solar fraction. For 81 m 2 of PTS s, 5 m 3 is the optimal volume of the storage tank. 40% 35% 30% Solar fraction 25% 20% 15% 10% 5% Solar heating (52 m2) Solar cooling (52 m2) Solar heating (81m2) Solar cooling (81 m2) 0% Volume of tank (m 3 ) Figure 6-22 Effect of PTS area and storage volume on the solar fraction in IW cooling and heating 6.7 Guidelines for design and operation of solar cooling and heating system onclusions from the results of calculations provided by the Trnsys model of solar cooling/heating in the IW throughout a year regarding the design and operation of such systems are summarized as follows: The orientation of the solar collectors The orientation of the PTS s impacts the performance of PTS based solar cooling system. A NS axis for PTS tracking is the orientation that provides the greatest amount of solar energy collected throughout the year for cooling and heating. PTS area The area of the PTS for the IW is reasonably determined so that the solar energy collected during a design day is that required by its cooling/heating load. The thermal storage stores throughout the day solar energy when it is in excess then makes it available when there is a 117

134 deficiency of incident solar energy. PTS s of larger area with adequate storage tank capacity improve the system performance for both cooling and heating on days when the solar incidence is intermittent or low. The selection of PTS area and thermal capacity is properly determined by an economic analysis. Storage tank capacity A storage tank located in the collection loop stores, during the day, thermal energy in excess of that required for cooling/heating and makes it available for cooling/heating at a later time. The storage can also be used to minimize the effects of short term fluctuations in the solar radiation or the cooling/heating load. Since the solar supply and the cooling load of the IW are nearly coincident, only a small storage tank is required. The solar supply and the heating load of the IW during the evening, night, and early morning are far from coincident. A sizable storage tank can significantly improve the system performances, the solar fraction, by storing the excess solar energy during the day and making it available in the evening and early morning. Drain-back system for preheating A gas fired auxiliary heater reduces the start up time and increase the time duration for solar cooling and heating. But this feature merely substitutes the energy of natural gas for solar energy, and it increases the capital and operating costs of the solar system. Therefore, based on similar principle, a drain-back system could be recommended for preheating. The length and diameter of pipe in the solar collection and the load loops Decreasing the length and diameter of the pipe in the solar collection loop reduces the system heat capacity, the heat loss, and the preheat time and energy required by the system. And these dimensions also affect the pressure loss and the energy required for circulating the HTF in the solar collection loop. A length and diameter of this pipe line as small as feasible, limiting the pressure loss and pumping energy for the loop, optimizes the system performance in both solar cooling and heating. Operating strategy Operating a PTS at a constant outlet temperature results in a higher solar fraction than operating at a constant HTF flow in solar cooling. This procedure effectively shortens the preheating time of system. Its advantage, however, is dependent on the heat capacity of the system. 118

135 In conclusion, additional system simulations have been implemented to predict the system performance of IW solar cooling and heating system applying the recommended design parameters and operating controls. The results indicated in Figure 6-23 show that the IW solar cooling/heating system with 52 m 2 of PTS s oriented NS, a 4 m 3 storage tank, a collection loop volume one fourth the current value, operated at a constant collector outlet temperature would provide 40% of the IW cooling load and 20% heating load by use of solar energy. But the orientation of the PTS in the IW is EW; the IW solar cooling/heating system with a 4 m 3 storage tank and one forth of the current collection loop volume would provide about 30% of its cooling by solar energy and 37% of its heating. 45% 40% Solar cooling (N/S, 52 m2, 1/4 V) 35% Solar heating (E/W, 52 m2, 1/4 V) 30% Solar cooling (E/W, 52 m2, 1/4 V) Solar fraction 25% 20% 15% Solar heating (E/W, 52 m2) Solar heating (N/S, 52 m2, 1/4 V) Solar cooling (E/W, 52 m2) 10% 5% 0% Volume of tank (m 3 ) Figure 6-23 Idealized IW solar cooling/heating system performance and system sensitivity analysis In general, the design and operation of a solar thermal cooling/heating system should be based on the building and its load profiles, the climate conditions and the incident solar radiation profiles, the physical limitations of the situation, and the economics of the energy supply. 119

136 7 ontributions and areas of future research 7.1 ontributions This thesis presents methods developed for the effective design, operation, and evaluation of parabolic trough solar collectors, PTS s, and also of the systems using them for providing cooling and heating for buildings. Such systems will reduce the primary energy consumption, and improve environmental benefits in buildings by using renewable energy, solar energy. The work reported in this thesis comprises 1. The establishment of a test bed PTS based solar cooling and heating system for a building A solar absorption cooling and heating test system was installed, and tested in the IW. Its primary components are 52 m 2 of parabolic trough solar collectors, PTS s; a 16 kw double effect absorption chiller; and a heat recovery exchanger together with a heat exchanger that simulates building cooling/heating loads for solar collector and system test and performance evaluation. A web based control and data acquisition system was developed to operate the solar thermal system while storing and displaying the test measurement data. The PTS s were tested at various operating conditions: direct normal solar irradiation; heat transfer fluid, HTF, flow and temperature; and wind velocity. Tests throughout 37 days involving solar cooling and heating were conducted at various weather conditions: clear days, mostly sunny days, mostly cloudy and overcast days in winter and summer. The analyses and evaluation of the experimental data from the tests of solar collectors and system were carried out using statistical analyses to define the efficiency and heat capacity of solar collectors, the heat and pressure losses of the solar collector loop, the OP and capacity of the absorption chiller, and the overall transfer coefficient of the heat recovery exchanger. In the future, the operation of this solar absorption cooling and heating test system will be integrated with the cooling and heating units in the IW, with the campus chilled water and heated water grids, and with the ventilation air supply unit. 2. The development and programming of a comprehensive solar collector model for analysis of experimental data obtained from the parabolic trough solar collectors and for the design of improved collectors. A comprehensive mathematical, PTS performance model based on the fundamental scientific and engineering principles was developed and programmed to analyze the experimental data from 120

137 the test program, to assist in the PTS design, and to assist in the evaluation of the system performance of solar cooling and heating. The model incorporates the energy balance relations for the absorber tube and the glass tube envelope and the heat transfer correlations among them and the surroundings. When appropriate assumptions, design parameters, operating conditions, and material properties are provided, the model can be solved calculating the efficiency, the heat capacity, and heat and pressure losses of the collector, and the temperatures throughout the collector. The solar collector model considers the effects of the intensity and direction of the normal solar radiation, the collector design and dimensions, the HTF fluid properties, and the ambient and operating conditions on the performance of the collector. 3. Development of an overall solar based, building cooling/heating system performance model for evaluating experimental data and for system design An overall system performance simulation was developed to assist the system design, to evaluate the system performance, and to optimize the system configuration. The model developed is able to calculate and consider in detail the working conditions of each system component (PTS s, chiller, recovery exchanger, heat storage, piping, and controls) under various ambient weather and operating conditions to investigate the effectiveness of the solar based system in cooling and heating a building space (the IW), to validate the system operation, to quantify the system performance, to provide a tool and technique for analysis of system, and to assist in the optimization of the design and operation of solar absorption cooling and heating systems. 4. Analysis of the experimental data, refinement of the solar collector model, and recommended provisions for the improvement of PTS design The model developed for the PTS s has been used to analyze the experimental data from the solar collector test program. Discrepancies between the measurements and the model calculations, most important low measured efficiencies and high measured glass tube temperature, have been found and used to adjust model assumptions, calculations, and parameter estimates. The discrepancies found were primarily associated with the heat losses from connectors and supporting arrangements of the absorber pipes and with an assumed high value of the transmissivity of the glass tube of the collectors. The validated solar collector model has been applied to project PTS performance under the various ambient weather and operating conditions. Significant recommendations concerning the PTS s design have been provided 121

138 regarding the designs of the collector tracking arrangements, the glass tube, and the absorber pipe with its connectors and supports 5. Analysis of the experimental data on solar absorption cooling and heating, and optimization of the design and operation of the solar absorption cooling/heating system Performance projections by the overall solar building cooling/heating system model were compared with experimental data on the cooling and heating of the IW obtained throughout several days in the test program. Reasonable agreement was observed. The model was used in sensitivity studies to explore what design and operation modifications in the installed solar cooling/heating system might be most effective in improving the performance of the system, primarily the fraction of the cooling/heating loads of the IW that could be met by solar energy system. The modifications explored in these studies include: the orientation and area of the solar collectors; the provision of thermal storage in the solar loop; changes in the solar loop pipe length, diameter, and insulation; modified system operation and control strategies. Finally, guidelines have been formulated to provide a basis for the preliminary selection of components, a configuration, and an operating approach in the design of an effective solar absorption cooling and heating system for a particular building. 7.2 Areas of future research Improving the tracking system of the PTS The installed PTS s track the sun in one dimension according to the tracking angle calculated by a set of equations that consider of the relative positions of the earth, the sun, and the PTS s at each instant throughout the year. The equations after an initial correction was made appear to be accurate. But experiments have showed that the PTSs did not accurately track the sun throughout a day, especially in the early morning or late afternoon. The difficulties in tracking and focus of the solar reflectors may be due to the tracking mechanisms and/or to structural inaccuracies of the reflector position and shape and of the support for the absorber pipe. Accurate tracking is the key factor for a PTS based solar thermal system in achieving a high efficiency for the recovery of solar energy throughout a day. Two approaches can be considered to improve the current tracking system. One is adding radiation focus sensors of various types. This approach adjusts the tracking based on a measurement at a single point along the length of the collector. 122

139 Another approach to improved tracking uses feedback control based on maximizing the temperature difference between the outlet and inlet HTF temperatures over the PTS s by adjusting the position of the reflector. This approach needs accurate temperature measurements at appropriate time intervals, but it does deal with the integrated performance of the whole of the PTS Extending the operational controls of the PTS, the absorption chiller, and the heat recovery exchanger The current solar cooling and heating system was tested in space cooling and heating; its operation now needs to be automated and integrated with the cooling/heating units of the IW and with the chilled and heated water grids of the campus. It still requires a lot of manual control to fulfill the task. In order to have a completely automatic control, both the controls of absorption chiller and the PTS need to be improved at following aspects. tracking offset control of the PTS At an instance, if the solar energy is gained by the system more than the system could receive, a tracking offset would be a better idea to limit the receiver temperature than fully defocusing the receiver tube off the focal line like current situation. Since a small tracking offset could not only reduce the solar gain of the system, but also stabilize the system operation. Therefore after the quantification of the relation between the solar gain and a tracking offset under a certain direct normal solar radiation, a tracking offset control of the PTS could be developed to efficiently control solar thermal system. Develop drain-back system A system with a large system heat capacity, like IW solar thermal system, normally has a long preheating time to heat the HTF in the pipe to the useful temperature. This long preheating process not only consumes a lot of useful solar energy but also shortens the effective solar cooling and heating time. So reducing this preheating time of the system will definitely improve the system performance. A drain-back system can be developed in the IW to shorten preheating period. There is a storage tank installed indoor for filling and draining the HTF. After well insulated, this storage tank could be utilized as a drain-back tank. So all of hot HTF from solar collection loop could be collected back to this tank when the HTF could not be directly used by 123

140 the absorption chiller / HX and then the stored hot HTF could be discharged from the tank back to the collection loop in the next sunny day. The substantial control strategy will also be developed to effectively operate the solar thermal system. Develop automated operation of solar cooling and heating system The current solar cooling and heating system installed in the IW is still operated in a partially manual way. A completely automated operational control system will be developed to operate the system without human being involving. This automated operation control includes start-up and shut-down the operation of solar cooling and heating system, operating absorption chiller / HX under various condition, and safely operating PTS. According to the cooling and heating demands and solar availability, this automated control can start to operate the absorption chiller by using natural gas or HX to meet the cooling or heating requirement and operate the PTS to collect solar energy if the conditions to operate the PTS are satisfied. Before operation of the PTS, the control system can refill the solar collection loop with the hot HTF from the drainback tank. It, then, can circulate and heat the HTF to the temperature desired by the absorption chiller / HX. The operation of the PTS can be protected by this control system without any damage from preheating. When the HTF cannot provide the energy required by the chiller / HX, the control system could drain the hot HTF from the solar collection loop back to the drain-back tank and shut-down the PTS. If there are no load demands, the system will shut-down by this automated control system. This automated control system integrates all of components in the solar cooling and heating system and safely operate it in completely automatic way Integrate thermal storage in the cooling/heating system Thermal storage can be an important component in solar based building cooling and heating systems that minimizes fluctuations in the energy supply and compensates for displacements in time of day between the solar heat supply and the cooling/heating energy requirements. Thermal storage can be provided either in the form of HTF at high temperature or of a material that changes phase absorbing/desorbing heat at a selected temperature. HTF based storage requires a relatively large volume of fluid in order to provide a high capacity. Phase change based storage reduces the volume and weight of storage, but it operates effectively only at the temperature of the phase change. In the future work, a phase change storage device will be integrated with the current IW solar cooling and heating system. 124

141 7.2.4 ost model A comprehensive cost model should be provided to support decision making in applying a solar cooling and heating to a particular building. This model should facilitate the comparison of the PTS based solar cooling and heating system with other alternative solar systems, and with the traditional cooling and heating system. As the basis of economical analysis, a cost model should be developed to predict the capital, operating, and maintenance costs of a solar cooling and heating system. The model will use these costs and the interest rates to provide the expected economic value added the return on investment, and the overall system economic performance of a solar cooling and heating systems. 125

142 8 References Active Solar Heating Systems Design Manual, ASHRAE Standard Test Method for Determining Thermal Performance of Tracking oncentrating Solar ollectors omparative limatic Data. 2001, National limatic Data enter. 8. ASHRAE handbook Assilzadeh, F., et al., Simulation and optimization of a LiBr solar absorption cooling system with evacuated tube collectors. Renewable energy, : p asals, X.G., solar absorption cooling in Spain: perspectives and outcomes from the simulation of recent installations. Renewable energy, : p ohen, G.E., D.W. Kearney, and G.J. Kolb, Final report on the operation and maintenance improvement program for concentrating solar power plants, in SAND Dudley, V.E., SEGS LS-2 solar ollector Duff, W.S., et al., Performance of the Sacramento demonstration IP collector and double effect chiller. Solar Energy, : p Duffie, J.A. and W.A. Beckman, Solar Engineering of Thermal Processes. 1980, New York: John wiley & sons. 15. Florides, G.A., et al., Modeling, simulation and warming impact assessment of a domesticsize absorption solar cooling system. Applied thermal engineering, : p

143 16. Florides, G.A., et al., Modeling and simulation of an absorption solar cooling system for YPRUS. Solar Energy, (1): p Hansen, E.G., Hydronic system design and operation, a guide to heating and cooling with water. 1985, New York: McGraw-Hill, Inc Henning and H.m., Air conditioning with solar energy Hewett, R., Solar absorption cooling: An innovative use of solar energy. AIhE SYMPOSIUM SERIES, (306): p Holman, J.P., Heat transfer. 8th ed. 1997: McGraw-Hill, Inc. 21. Ibrahim Atmaca and A.Y., Simulation of solar-powered absorption cooling system.. Renewable Energy, : p Ileri, A., A discussion on performance parameters for solar-aided absorption cooling systems. Renewable Energy, (4): p Incorpera, F.P., Fundamentals of heat and mass transfer. 3th ed. 1990: John Wiley & Sons. 24. Iqbal, M., An introduction to solar radiation. 1983: Academic Press anada. 25. Joudi, K.A. and Q.J.a. G., Development of design charts for solar cooling systems. Part I: computer simulation for a solar cooling system and development of solar cooling design charts. Energy conversion & management, : p Kalogirou, s., s. lloyd, and j. ward, modeling, optimisation and performance evaluation of a parabolic trough solar collector steam generation system. solar Energy, (1): p Kaushik, S.. and Y.K. Yadav, Thermodynamic design and assessment of hybrid double absorption solar cooling systems. Heat recovery system & HP, (4): p Knapp,.L. and T.L. Stoffel, Direct Normal Solar Radiation Data Manual. 1982, SERI report 29. Kulkarni, P.P., Solar Absorption ooling for Demand-side Management. Energy Engineering, (5): p

144 30. Lokurlu, A., et al., A new kind of steam supply and air conditioning in a hotel in antalya, Turkey, by application of parabolic trough collectors (SOLITEM PT) combined with double effect absorption chiller. VDI-BERIHTE, Mendes, L.F., Supply of cooling and heating with solar assisted absorption heat pumps: an energetic approach. Int J. Refrig., : p Mendes, L.F., M.. P., and F.Ziegler, Supply of cooling and heating with solar assisted absorption heat pumps: an energetic approach.. International Journal of Refrigeration, (2): p Odeh, S.D., G.L.M., and M.Behnia, Modelling of parabolic trough direct steam generation solar collectors. Solar energy, (6): p Patnode, A.M., simulation and performance evaluation of parabolic trough solar power plants. 2006, University of Wisconsin-madison 35. Price, H., Advanced in parabolic trough solar power technology. Journal of solar energy engineering, : p Stine, W.B., Energy Fundamentals and Design: With omputer Applications. 1985, New York: John Wiley and Sons. 37. Stine, W.B., Solar industrial process heat project. 1989, Sandia National laboratories. 38. Stine, W.B. and M. Geyer, Power from the sun: Principles of high temperature solar thermal technology. 1987: Solar Energy Research Institute. 39. Syed, A., optimal solar cooling systems Wardono, B., Simulation of a solar-assisted LiBr/H2O cooling system. ASHRAE transactions, (1): p Yin, H., An Absorption hiller in a Micro BHP Application: Model based Design and Performance Analysis. 2006, arnegie Mellon University: Pittsburgh. 128

145 Appendix i

146 Table of ontents of Appendix Appedix 1 The control of PTS Appedix 1.1 Sun position in the literatures A1.1.1 Solar constant A1.1.2 Solar angles Appedix 1.2 alculation in broad tracking program Appedix 1.3 Broad solar control system Appedix 2 The experimental performance data of the PTS Appedix 3 The PTS mathematic model Appedix 3.1 ode of the PTS model Appedix 4 Validation of the PTS model Appedix 4.1 Radiation transmission through covers and absorption by collectors Appedix 5 Solar cooling and heating system simulation Appedix 5.1 Building simulation interface A5.1.1 Geometry A5.1.2 Materials A5.1.3 Simulation of the ventilation Appedix 5.2 Solar heating evaluation simulation A5.2.1 Definition of the components in the model Appedix 5.3 Solar cooling evaluation simulation Appedix 5.4 Solar heating base-case Appedix 5.5 Solar cooling base-case Appedix 5.6 ode of parabolic trough solar collector Appedix 5.7 ode of main control of solar cooling base-case Appedix 6 System optimization and sensitivity analysis Appedix 6.1 Solar heating system with constant-outlet-temperature control Appedix 6.2 Solar cooling system with constant-outlet-temperature control Appedix 6.3 Solar heating system with storage tank A6.3.1 ontrol of solar heating system with storage tank Appedix 6.4 Solar cooling system with storage tank for shifting energy A6.4.1 ontrol of solar cooling system with storage tank for shifting energy ii

147 Appedix 6.5 Solar cooling system with storage tank for preheat A6.5.1 ontrol of solar cooling system with storage tank for preheat Appedix 6.6 Solar heating system with auxiliary heater for preheat Appedix 6.7 Solar cooling with auxiliary heater for preheat Appedix 6.8 ode of type 236: PTS with constant-outlet-temperature control Appedix 6.9 ode of type 237: main control of solar cooling for constant-outlettemperature control Appedix 6.10 ode of type 243: main control of solar cooling with storage tank for shifting energy by constant-outlet-temperature Appedix 6.11 ode of type 242: main control of solar cooling with storage tank control for preheat controlled by constant-outlet-temperature Appedix 6.12 code of type 245: ontrol of solar heating with auxiliary heater controlled by constant-outlet-temperature Appedix 6.13 ode of type 244: control of solar cooling with auxiliary heater for preheat 6-47 iii

148 Appedix 1 The control of PTS The data acquisition and control system, provided by Broad o, is able to operate the PTS to track sun throughout a day, protect the PTS, and record and display the experimental data as well. This control system calculates the tracking angle and sends commands to the PTS tracking the sun, based on solar field location and orientation (the longitude, altitude), date and time, tracking setting, operation condition and schedule, as indicated in the inputs block of Figure 1:1. Figure 1:1 Broad solar control inputs and outputs Appedix 1.1 Sun position in the literatures A1.1.1 Solar constant Solar energy approaches the earth as electromagnetic radiation, with wavelengths ranging from 0.1 um (x-rays) to 100 m (radio waves). Approximately 99% of sun s radiation energy has wavelengths between 0.28 and 4.96 um. The current value of the solar constant, which is defined 1-1

149 as the intensity of solar radiation on a surface normal to the sun s rays, just beyond the earth s atmosphere at the average earth-sun distance, is 433 Btu/h. ft 2 (1367 W/m 2 ). Due to absorption and scattering in the atmosphere, solar radiation is attenuated as it reaches the earth s surface [ASHRAE handbook 2003]. A1.1.2 Solar angles The earth rotates about its axis which is tilted at an angle of to the plane of the earth s orbital plane and the sun s equator. Due to this tilted axis, the position of the sun varies throughout a day. The declination angle is used to describe the angular position of the sun at solar noon with respect to the plane of the equator. The following expression for declination angle was developed by P.I. ooper in 1969 (Beckman 1980) N δ = 23.5 sin[360 ] (Equation 1:1) 365 Where, N= the day of the year, for Jan 1 st, N=1 Because the earth s daily rotation and its annual orbit around the sun are regular and predictable, the solar path can be calculated for any desired time of day if altitude, longitude, and date (declination) are specified. The hour angle determines the position of the sun, calculated by Equation 1:2. It is the angular displacement of the sun east or west of the local meridian. H = (12-AST)*15 (Equation 1:2) Where, AST= Apparent solar time in hour, which is claculated based on Equation 1:3. AST=LST+EOT+(LST meridian-local longitude)/15 (Equation 1:3) Where, LST=local stand time EOT = Equation of time [min], calculated based on Equation 1:4 LST meridian= standard meridian for the local time zone [deg] Local longitude = the local meridian of the site [deg] EOT=(229.2*(7.5*10^(-5) *OS(B) *SIN(B) *OS(2B) *SIN(2B)))/60 (Equation 1:4) ( Iqbal 1983) 1-2

150 Where, B=360*(N-1)/365, where N= the day of the year, for Jan 1 st, N=1 So solar altitude( β ), the angle between the line of signt to the sun anf the horizont, can be calculated based on decilination angle(δ ), latitude location of the site (LAT), and hour angle (H) as shown in Euqation 1:5 sin β = cos( LAT ) * cosδ * cos H + sin( LAT ) * sin δ (Equation 1:5) Figure 1:2 ingle axis tracking system coordinates Solar altitude s complement, the angle between the line of sight to the sun and the vertical, is the zeith angle. Solar azimuth (φ ) can be calculated according to Equation 1:6 sinφ = cosδ *sin H / cos β (Equation 1:6) 1-3

151 The tracking angle ( ρ ) measures rotation about the tracking axis, with ρ =0 when N is vertical as shown. It can be calculated based on Equation A1.7 as indicated Figure 1:2. (Stine 1987) ρ = S S b tan (Equation 1:7) u So the general case of a collector aperture tracking about a single, horizontal axis, the tracking angle tan ρ = S S b u cos β * cos( φ γ ) = sin β cos( φ γ ) = tan β (Equation 1:8) Where, γ : Orientation (azimuth of the surface normal) from true SOUTH, toward east -, toward west + And the incident angle, which the angle between the sun ray and the normal to the aperture surface of collector, is 2 2 cosθ i = 1 cos β *sin ( φ γ ) (Equation 1:9) 1-4

152 Appedix 1.2 alculation in broad tracking program Reference name ASHRAE BROAD Solar declination (same) Equation of time (hour) (Same) Solar time (same) N δ = 23.5sin[360 ] 365 Where, for Jan 1 st, N=1 EOT= 0.165*SIN(2B)-0.126*OS(B)-0.025*SIN(B) Where, B= 360*(N-81)/364 AST=LST+EOT+(LST meridian-local longitude)/ N δ = 23.5sin[360 ] 365 Where, for Jan 1 st, N=1 EOT=(229.2*(7.5*10^(-5) *OS(B) *SIN(B) *OS(2B) *SIN(2B)))/60 Where, B=360*(N-1)/365 AST=LST+EOT +(Local longitude- LST meridian-)/15 (East global) AST=LST+EOT -(Local longitude- LST meridian-)/15 (West global) Hour angle ( ) (inverse sign) H=(12-AST)*15 East : + ; West: - H=(AST-12)*15 Moring: - ; Afternoon: + Solar latitude ( ) (same) Solar azimuth ( ) Morning inverse sign; afternoon, same sin β = cos( LAT ) * cosδ * cos H + sin( LAT ) * sin δ Where, LAT= local latitude sinφ = cosδ *sin H / cos β OR sin β *sin( LAT ) sinδ cosφ = cos β *cos( LAT ) sin β = cos( LAT ) * cosδ * cos H + sin( LAT ) * sin δ Where, LAT= local latitude if δ LAT, then1 = 1 EW = arccos(tan( δ ) / tan( LAT ) if H EW, then1 = 1; otherwise1 = 1 iflat * ( LAT δ ) 0, then2 = 1; otherwise2 = 1 ifh 0, then3 = 1; otherwise3 = 1 FF = sin H * cosδ / cos β φ = 1* 2 * arcsin( FF) * 3* (1 1* 2) / 2 1-5

153 Vertical Shadow Angle ( ) Horizontal axis, the tracking angle Incident angle tan( VSA ) = tan( β ) / cos( φ γ ) Where, γ : orientation (azimuth of the surface normal) S tan ρ = S cosθ = i b u cos β * cos( φ γ ) cos( φ γ ) = = sin β tan β 1 cos 2 β *sin 2 ( φ γ ) cosθ = cos ρ * cos(90 β ) + sin ρ *sin(90 β ) * cos( φ γ ) i if φ > 90, RS = arctan(tan β / cos( φ γ ); otherwise RS = arctan(tan β / cos( φ γ ) 1-6

154 Appedix 1.3 Broad solar control system The PTS s control system monitors the instant operation temperature, pressure, solar radiation and wind velocity. Three operation modes are available in Broad solar control system: commissioning, automation, linkage. The commissioning mode is used to check if the main solar circulation pump and the PTS work well individually. The automation mode is used to automatically operate the PTS throughout a day under protection. The linkage mode is used to operate the PTS and absorption chiller together. This model basically is same as automation model, but it allows PTS and absorption chiller working together to provide cooling for a space through the information communication between them. In addition, Broad solar control system can also defocus the PTS to prevent the damage of the PTS or system when the operating temperate or pressure in the solar loop is too high. Figure 1:3 shows the control logic of Broad solar collector control. Figure 1:3 Broad PTS control block diagram 1-7

155 Appedix 2 The experimental performance data of the PTS Table 2:1 PTS performance experimental data in the steady states Date Time T_amb ( ) Wind_speed W1 (m/s) NIP (W/m^2) T1 T2 T2b solar loop flow F1 (gpm) solar loop flow F1 (kg/hr) two array Heat gain I*Aa*cos(Th Average Operated (kw) (mcpt2 eta)(kw) temp. [] 4/20/ :45~14: /20/ :15~14: /21/ :40~14: /21/ :10~14: /22/ :00~13: /22/ :30~14: /22/ :20~14: /22/ :50~15: /22/ :25~16: /22/ :49~16: /24/ :42~11: /24/ :53~12: /24/ :52~13: /24/ :22~13: /2/ :07~12: /2/ :37~13: /2/ :08~13: /2/ :38~14: /4/ :03~12: /4/ :33~13: /4/ :12~13: /4/ :42~14: /4/ :25~14: /4/ :55~15: /6/ :00~13: /6/ :30~14: /6/ :08~14: /6/ :38~15: /6/ :34~16: /6/ :04~16: /8/ :39~13: /16/ :33~14: /16/ :44~14: /16/ :10~15: /16/ :23~15: /31/ :01~13: /31/ :29~14: /31/ :02~14: /31/ :32~15: /31/ :03~15: /31/ :25~16: /9/ :30~14: /9/ :00~14: /9/ :31~15: /2/ :20~16: /8/ :30~15: /8/ :30~15: /20/ :45~15: /20/ :15~15: /20/ :46~16: /20/ :03~16: /11/ :35~15: /11/ :04~15: /11/ :36~15: /11/ :55~16:

156 Table 2:2 The flow rate required to ensure the turbulent flow Temperature Pipe size ID Dynamic viscosity Start Tubulent flow Tubulent flow Density Tubulent flow Tubulent flow Transition flow Transition flow ( ) (m) kg/m.s (Red=2300 )kg/s kg/h kg/m3 m3/h (2300)gpm (Red=4000 )kg/s (4000)gpm 3/4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2" /4" /4" /2"

157 Appedix 3 The PTS mathematic model Figure 3:1 Interface of the PTS mathematic model 3-1

158 Appedix 3.1 ode of the PTS model {Parabolic Trough Solar ollector Model By Ming Qu Feb.16,2006 Updated Sep.11,07 omments on Sep.11,07} {function for identifying the conductivity of glass, } Procedure Glassfeature(T_gi,T_go:A_g,T_g,R_g,E_g,k_g) $OMMON Glass$,THETA,OD_g,ID_g T_g:=(T_gi+T_go)/2 IF (Glass$='Pyrex glass') THEN IF(THETA=0) THEN A_g:=0.11 T_g:=0.82 R_g:=0.07 E_g:=0.86 ELSE {the extinction coefficient of the glass} K:=23 "[m-1]" THK_g:=OD_g-ID_g {the average index of refraction of glass for the solar spectrum is Based on "Solar Engineering of Thermal Processes" P173} Index_g:=1.526 Theta2:=arcsin(sin(THETA)/Index_g) r_perp:=(sin(theta2-theta)/sin(theta2+theta))^2 r_para:=(tan(theta2-theta)/tan(theta2+theta))^2 Tau_g_a:=EXP((-1)*K*THK_g/cos(Theta2)) A_g:=1-Tau_g_a T_g_r:=1/2*((1-r_para)/(1+r_para)+(1-r_perp)/(1+r_perp)) T_g:=T_g_r*Tau_g_a R_g:=Tau_g_a-T_g E_g:=0.86 k_g:=1.125 IF (Glass$='Pyrex AR glass') THEN A_g:=0.01 T_g:=0.94 R_g:=0.03 E_g:=0.86 IF(THETA=0) THEN A_g:=0.11 T_g:=0.82 R_g:=0.07 E_g:=0.86 ELSE {the extinction coefficient of the glass} K:=21 "[m-1]" THK_g:=OD_g-ID_g 3-2

159 {the average index of refraction of glass for the solar spectrum is Based on "Solar Engineering of Thermal Processes" P173} Index_g:=1.526 Theta2:=arcsin(sin(THETA)/Index_g) r_perp:=(sin(theta2-theta)/sin(theta2+theta))^2 r_para:=(tan(theta2-theta)/tan(theta2+theta))^2 Tau_g_a:=EXP((-1)*K*THK_g/cos(Theta2)) A_g:=1-Tau_g_a T_g_r:=1/2*((1-r_para)/(1+r_para)+(1-r_perp)/(1+r_perp)) T_g:=TAU_g_r*Tau_g_a R_g:=Tau_g_a-TAU_g E_g:=0.86 k_g:=1.125 END {function for idetifying the characteristics of absorber tube} Procedure Absorbtubefeature(T_ai,T_ao:A_a,T_a,R_a,E_a,k_a,e) $OMMON Steel$,oating$ T_a:=(T_ai+T_ao)/2 IF (oating$='black chrome') THEN A_a:=0.94 T_a:=0.01 R_a:=0.05 E_a:= *(T_a+T_zero#) IF(E_a<0.11) THEN E_a:=0.11 IF (oating$='black nickel') THEN A_a:=0.96 T_a:=0.01 R_a:=0.03 E_a:=0.43 IF (Steel$='Stainless Steel 304L') THEN k_a:=k_('stainless_aisi304', T_a) IF (Steel$='Stainless Steel 316L') THEN k_a:=k_('stainless_aisi316', T_a) e= [m] END "[W/m-K]" "[W/m-K]" {roughness of the absorber tube} {function for identifying the properties of fluid} Procedure Fluidproperties(T_f,P_w:_f,D_f,OND_f,V_f,P_f) $OMMON Fluid$ IFNOT(Fluid$='WATER') THEN _f:=interpolate(fluid$,'temperature','p',temperature=t_f) "[kj/kg-k]" {heat capacity} 3-3

160 D_f:=INTERPOLATE(Fluid$,'TEMPERATURE','DENSITY',TEMPERATURE=T_f) "[kg/m^3]" {density} OND_f:=INTERPOLATE(Fluid$,'TEMPERATURE','ONDUTIVITY',TEMPERATURE=T_f) "[W/mK]" {conductivity} V_f:=INTERPOLATE(Fluid$,'TEMPERATURE','VISOSITY',TEMPERATURE=T_f) "[kg/ms]" {viscosity} P_f:=INTERPOLATE(Fluid$,'TEMPERATURE','PRESSURE',TEMPERATURE=T_f) "[kpa]" {pressure} IF(Fluid$='WATER') THEN _f:=speheat(water,t=t_f,p=p_w) {heat capacity} D_f:=DENSITY(Water,T=T_f,P=P_w) {density} OND_f:=ONDUTIVITY(Water,T=T_f,P=P_w) {conductivity} V_f:=VISOSITY(Water,T=T_f,P=P_w) {viscosity} P_f:=P_w {pressure} END {function of the convection from the absorber tube inner surface to the heat transfer fluid} Procedure onv_f(tf,t_wall : Re_f,Pr_f,Q_12,FT$,m_t,h_f) $OMMON m1,dh_1,p_water, L_aper,S_a ALL Pro_conv_f(Tf,T_wall,Dh_1,L_aper,S_a: p_f_b,ft$,re_f,pr_f,m_t,h_f) Q_12:= h_f*pi*dh_1*(t_wall-tf) END {function of identifying the heat transfer coefficient of the convection from the absorber tube to the heat transfer fluid} Procedure Pro_conv_f(Tf,T_wall,Dh,L,S: p_rf_b,ft$,re_rf,pr_rf,m_rt,h_rf) $OMMON m1,p_water, ALL Fluidproperties(Tf,P_water:p_rf_b,rho_rf_b,KAPPA_rf_b,MU_rf_b,P_rf_b) ALL Fluidproperties(T_wall,P_water:p_rf_w,rho_rf_w,KAPPA_rf_w,MU_rf_w,P_rf_w) Re_rf:=Dh*m1/S/MU_rf_b Pr_rf:=1000*p_rf_b*MU_rf_b/KAPPA_rf_b converter} m_rt:=2300*s*mu_rf_b/dh flow which reyolds number >2300} { Kappa unit is W/m, 1000 is unit {the mininum flow rate to have turbulent IF(Re_rf>2300) THEN FT$:='Turbulent Flow' {Turbulent flow based on Frank P. Incorpera 3th "Fundamentals of heat and mass transfer" PP497} IF (2000>Pr_rf) AND (Pr_rf>0.5) AND (5*10^6>=Re_rf) THEN {for 6~10 percent accuracy} f:=(0.79*ln(re_rf)-1.64)^(-2) 3-4

161 Nus_rf_b:=(f/8)*(Re_rf-1000)*Pr_rf/(1+12.7*(f/8)^0.5*(Pr_rf^(0.6667)-1)) h_rf:=nus_rf_b*kappa_rf_b/dh "[w/m^2-k]" ELSE ALL WARNING('THERE IS NO AURATE EQUATION FOR THIS TYPE OF FLOW') ELSE {Laminar flow based on J.P.Holman 8th "Heat Transfer"PP289} FT$:='Laminar Flow' Nus_rf_b:=3.66+(0.0668*(Dh/L)*Re_rf*Pr_rf)/(1+0.04*((Dh/L)*Re_rf*Pr_rf)^(0.6667)) h_rf:=nus_rf_b*kappa_rf_b/dh END {function of convection between the absorber tube and glass envelope} Procedure onv_a_g(t_f_g,t_f_a,p_air_ga:q_conv_34) $OMMON OD_a, ID_g,T_std T_f_ga=(T_f_g+T_f_a)/2 p_air_ga:=speheat(air,t=t_f_ga) v_air_ga:=v(air,t=t_f_ga) MU_air_ga:=VISOSITY(Air,T=T_f_ga) KAPPA_air_ga:=ONDUTIVITY(Air,T=T_f_ga) KAPPA_std:=ONDUTIVITY(Air,T=T_std) RHO_air_ga:=DENSITY(Air,T=T_f_ga,P=P_air_ga) NU_air_ga:=MU_air_ga/RHO_air_ga ALPHA_air_ga:=KAPPA_air_ga/(RHO_air_ga*1000*p_air_ga) T_inK:=T_f_ga+T_zero# BETA_air_ga:=1/T_inK Pr_air_ga:=NU_air_ga/ALPHA_air_ga Gr_air_ga:=g#*BETA_air_ga*ABS(T_f_a-T_f_g)*OD_a^3/NU_air_ga^2 Ra_air_ga:=Pr_air_ga*Gr_air_ga Ra_air_ga_L=g#*BETA_air_ga*ABS(T_f_a-T_f_g)*((ID_g- OD_a)/2)^3/(NU_air_ga*ALPHA_air_ga) Q_cond_34:= 0 Ra_air_ga_star:=(LN( ID_g/OD_a))^4*Ra_air_ga_L/( ((ID_g-OD_a)/2)^3*((OD_a)^(- 0.6)+(ID_g)^(-0.6))^5) IF(Ra_air_ga_star<100) THEN KAPPA_eff:=KAPPA_air_ga {Raithby and hollands version,based on Frank P. Incorpera 3th "Fundamentals of heat and mass transfer" PP563 } Q_conv_34:=2*PI*KAPPA_eff*(T_f_a-T_f_g)/LN( ID_g/OD_a) ELSE {the gas in the annular region is circulating by laminar natural convection. in laminar flow, the fluid does not mix and the heat transfer mechanism is molecular conduction, for 100<Rac<10^7, Based on Vernon E. Dudley, " SEGS LS-2 solar ollector", SAND } {free-molecular heat transfer for annular space between horizontal cylinders} {ga in mmhg, DELTA_air in cm} DELTA_air := 3.53*10^(-8) [cm] {molecular diameter of the gas in cm} 3-5

162 a:=1 {accommodation coefficient is defined as the ratio of the energy actually transferred between impinging gas molecules and a surface and the energy which would be theoretically transferred if the impinging molecules reached complete thermal equilibrium with the surface. a=1 unless the surface were extremely cleanned. this is what was assumed} gamma:=p_air_ga/v_air_ga {the ratio for specific heats for the gas inside the annulus for air equal to 1.4} b:=(2-a)/a*(9*gamma-5)/(2*(gamma+1)) {BELOW, is for unit converter 1kpa= mmHg} LAMBDA:=2.331*10^(-20)*T_inK/(P_air_ga* *DELTA_air^2) h_air_ga:=kappa_std/(od_a/2*ln( ID_g/OD_a)+b*LAMBDA*(OD_a/ID_g+1)) Kineticq_conv_34:= h_air_ga*pi*od_a*(t_f_a-t_f_g) {Raithby and hollands version,based on Frank P. Incorpera 3th "Fundamentals of heat and mass transfer" PP563 } KAPPA_eff:=KAPPA_air_ga*0.386*(Pr_air_ga/(0.861+Pr_air_ga))^(0.25)*(Ra_air_ga_star)^(0.25) Natq_conv_34:=2*PI*KAPPA_eff*(T_f_a-T_f_g)/LN( ID_g/OD_a) {use the larger value between natural convection heat transfer and free-molecular heat transfer} IF(Kineticq_conv_34>Natq_conv_34) THEN Q_conv_34:=Kineticq_conv_34 ELSE Q_conv_34:=Natq_conv_34 END {function of convection from the glass envelope to the surrounding air} Procedure onv_g_air(t_f_og,t_s,u_wind, OD:Q_conv_56,h_f) ALL Property_air_sur(T_f_og,T_s,U_wind,OD:Pr_g_air,Gr_g_air,Ra_g_air,Nus_g_air,h_g_air) h_f:=h_g_air Q_conv_56:= h_f*pi*od*(t_f_og-t_s) END {function of identifying the heat transfer coeeficient of the convection from the glass envelope to surrounding} Procedure Property_air_sur(T_sur,T_air,U_wind,OD:Pr_sur_air,Gr_sur_air,Ra_sur_air,Nus_s_air,h_s_air) $OMMON wind$ T_sur_air:=(T_sur+T_air)/2 P_air:=Po# p_air:=speheat(air,t=t_sur_air) MU_air:=VISOSITY(Air,T=T_sur_air) KAPPA_air:=ONDUTIVITY(Air,T=T_sur_air) RHO_air:=DENSITY(Air,T=T_sur_air,P=P_air) 3-6

163 NU_air:=MU_air/RHO_air ALPHA_air:=KAPPA_air/(RHO_air*1000*p_air) T_inK_sur_air:=T_sur_air+T_zero# BETA_sur_air:=1/T_inK_sur_air Pr_sur_air:=NU_air/ALPHA_air Gr_sur_air:=g#*BETA_sur_air*ABS(T_sur-T_air)*OD^3/NU_air^2 Ra_sur_air:=Pr_sur_air*Gr_sur_air IF(wind$='NO') THEN U_wind=0 IF(Ra_sur_air<=10^(-5)) or (Ra_sur_air>=10^12) Then ALL WARNING('The result may not accurate, since equation does not hold.') Nus_s_air:=( *Ra_sur_air^(0.1667)/(1+(0.559/Pr_sur_air)^(0.5625))^(0.2963))^2 h_s_air:=nus_s_air*kappa_air/od {Forced convection, based on J.P.Holman, "heat transfer", 8th edition, P302,303} IF(wind$='YES') THEN Re_sur_air:=U_wind*OD/NU_air Nus_s_air:=0 IF (Re_sur_air=<4) AND (Re_sur_air=>0.4) THEN Nus_s_air:=0.989*(Re_sur_air)^0.33*Pr_sur_air^(1/3) IF (Re_sur_air=<40) AND (Re_sur_air>4) THEN Nus_s_air:=0.911*(Re_sur_air)^0.385*Pr_sur_air^(1/3) IF (Re_sur_air=<4000) AND (Re_sur_air>40) THEN Nus_s_air:=0.683*(Re_sur_air)^0.466*Pr_sur_air^(1/3) IF (Re_sur_air=<40000) AND (Re_sur_air>4000) THEN Nus_s_air:=0.193*(Re_sur_air)^0.618*Pr_sur_air^(1/3) IF (Re_sur_air=<400000) AND (Re_sur_air>40000) THEN Nus_s_air:=0.0266*(Re_sur_air)^0.805*Pr_sur_air^(1/3) h_s_air:=nus_s_air*kappa_air/od END {function of convection from the glass envelope to surrounding} Procedure Property_air_brac(T_sur,T_air,U_wind,OD:Pr_sur_air,Gr_sur_air,Ra_sur_air,Nus_s_air,h_s_air) $OMMON wind$ T_sur_air:=(T_sur+T_air)/2 P_air:=Po# p_air:=speheat(air,t=t_sur_air) MU_air:=VISOSITY(Air,T=T_sur_air) KAPPA_air:=ONDUTIVITY(Air,T=T_sur_air) 3-7

164 RHO_air:=DENSITY(Air,T=T_sur_air,P=P_air) NU_air:=MU_air/RHO_air ALPHA_air:=KAPPA_air/(RHO_air*1000*p_air) T_inK_sur_air:=T_sur_air+T_zero# BETA_sur_air:=1/T_inK_sur_air Pr_sur_air:=NU_air/ALPHA_air Gr_sur_air:=g#*BETA_sur_air*ABS(T_sur-T_air)*OD^3/NU_air^2 Ra_sur_air:=Pr_sur_air*Gr_sur_air IF(wind$='NO') THEN U_wind=0 IF(Ra_sur_air<=10^(-5)) or (Ra_sur_air>=10^12) Then ALL WARNING('The result may not accurate, since equation does not hold,wind=no.') Nus_s_air:=( *Ra_sur_air^(0.1667)/(1+(0.559/Pr_sur_air)^(0.5625))^(0.2963))^2 h_s_air:=nus_s_air*kappa_air/od IF(wind$='YES') THEN {for noncircular cylinders, reference: Heat transfer,j.p. Holman P307} Re_sur_air:=U_wind*OD/NU_air IF(Re_sur_air<5000) or (Re_sur_air>100000) THEN ALL WARNING('The result may not accurate, since equation does not hold, bracket convection.') Nus_s_air:=0.102*(Re_sur_air)^0.675*Pr_sur_air^(1/3) h_s_air:=nus_s_air*kappa_air/od END {function of conduction through bracket} {for infinite fin, reference: Heat transfer,j.p. Holman P47} Procedure ond_bracket(t_ab,t_s,u_wind:q_cond_38,h_f_38) $OMMON wind$, OD_brac,P_brac,A_cs_brac,k_brac T_base=T_ab-15 T_support=(T_base+T_s)/2 about average T_base and ambient} {assume temperature od support is ALL Property_air_brac( T_support,T_s,U_wind,OD_brac:Pr_b_air,Gr_b_air,Ra_b_air,Nus_b_air,h_b_air) h_f_38:=h_b_air Q_cond_38:= SQRT( h_f_38*p_brac*a_cs_brac*k_brac)*(t_base-t_s) END {function of conduction through flange} Procedure ond_fin(t_ab,t_s,u_wind,l_fin,t_fin,r1,k_fin:q_cond_fin,q_cyl_permeter,h_f_fin) $OMMON wind$ Lc:=L_fin+t_fin/2 r2c:=r1+lc x_value:=r2c/r1 Am:=t_fin*(r2c-r1) 3-8

165 OD=2*r1 ALL Property_air_sur( T_ab,T_s,U_wind,OD:Pr_s_air,Gr_s_air,Ra_s_air,Nus_s_air,h_s_air) y_value:=lc^(3/2)*(h_s_air/k_fin/am)^(1/2) e:=interpolate2d('cir fin efficiency', 'x','y', 'eta_fin', x=x_value, y=y_value) h_f_fin:=h_s_air Q_fin_max=2*PI*(r2c^2-r1^2)*h_s_air*(T_ab-T_s) Q_cond_fin:= e* Q_fin_max Q_cyl_permeter:=2*PI*r1*h_s_air*(T_ab-T_s) END {Heat loss thru piping } Procedure Pipeloss(pmaterial$,insulations$,T_f,T_oa,L:k_p,k_ins,Q_pipe,rcr,t_1p,t_2p,t_3p,t_4p,R_01,R_12,R_23,R_34,R_total) $OMMON ID_pipe,OD_pipe,S_pipe,r_pipe_in,r_pipe_out,r_insult_out,V_wind OD_ins:=r_insult_out*2 ALL Property_air_sur( T_oa+0.5,T_oa,V_wind,OD_ins:Pr_pb_air,Gr_pb_air,Ra_pb_air,Nus_oa,h_oa) ALL Pro_conv_f(T_f,T_f-0.4,ID_pipe,L,S_pipe: p_p_f,ft$,re_p_f,pr_p_f,m_p_f,h_f) k_p:=k_(pmaterial$, t_f) k_ins:=interpolate(insulations$,'temperature','k',temperature=( t_f+t_oa+3)/2) "[W/mk]" R_01:=1/(2*PI*h_f*r_pipe_in) R_12:=LN(r_pipe_out/r_pipe_in)/(2*PI*k_p) R_23:=LN(r_insult_out/r_pipe_out)/(2*PI*k_ins) R_34:=1/(2*PI*h_oa*r_insult_out) R_total:=R_01+R_12+R_23+R_34 Q_pipe:=(T_f-T_oa)/R_total rcr=k_ins/h_oa t_1p:=t_f-q_pipe*r_01 t_2p:=t_1p-q_pipe*r_12 t_3p:=t_2p-q_pipe*r_23 t_4p:=t_3p-q_pipe*r_34 END {Heat loss thru piping } Procedure PipePro(pipingnd$,schedules$,TH_ins,t_f:ID_pp,OD_pp,S_pp,r1,r2,r3) ID_pp:=INTERPOLATE2D('PIPES', 'ND', 'schedule', 'ID', ND=pipingnd$, schedule=schedules$,1 )/100 "[m]" OD_pp:=INTERPOLATE2D('PIPES', 'ND', 'schedule', 'OD', ND=pipingnd$, schedule=schedules$,1) /100 "[m]" S_pp:=INTERPOLATE2D('PIPES', 'ND', 'schedule', 'FA', ND=pipingnd$, schedule=schedules$,1 ) /10000 "[m^2]" {fiberglass inch m} {cellularglass...4.5inch m} r1:=id_pp/2 r2:=od_pp/2 r3=r2+th_ins END 3-9

166 {function of presure in pipe} {for pressure loss in the pipe, reference: Introduction to fluid mechanics, Robert W. Fox P357~361} Procedure Pressure_f(Dh,S,m,T_f,P_f,e_p,ID,L :Re_f,Lambda_f,Delt_Pressure, Delt_H) ALL Fluidproperties(T_f,P_f:p_f,rho_f,KAPPA_f,MU_f,P_f_f) Re_f:=Dh*m/S/MU_f Pr_f:=1000*p_f*MU_f/KAPPA_f W_f:=m/rho_f/S IF(Re_f>2300) THEN {Turbulent flow} Lambda_f:=0.25/(log10(e_p/ID/ /Re_f^0.9))^2 ELSE {Laminar flow} Lambda_f:=64/Re_f Delt_Pressure=Lambda_f*W_f^2*L/ID*rho_f/2/1000 "[kpa]" {"1000" for unit conversion} Delt_H=Delt_Pressure*1000/g#/rho_f "[m]" {"1000" for unit conversion} END "************************************************************************************************************** {calculation of the solar energy absorbered by glass envelope and absorber tube} {Incidence angle modifer is equal to consin incident angle} IAM=cos(THETA) *(THETA) *(THETA)^2 ALL Glassfeature(T_4,T_5:ALPHA_g,TAU_g,RHO_g,EPSILON_g,KAPPA_g) E_op_beforein=E_op*R_mirror ALL Absorbtubefeature(T_2,T_3:ALPHA_a,TAU_a,RHO_a,EPSILON_a,KAPPA_a,e_a) q_sol_total=i_dn*w q_focus=i_dn*e_op_beforein*iam*w q_solab_g=q_sol_total*e_op_beforein*cos(theta)*1.01*tau_g*alpha_g q_solab_a=q_focus*tau_g*alpha_a {onvection of fluid} Dh_1=ID_a S_a=PI*ID_a^2/4 ALL onv_f(t1_avg,t_2:re_f_b,pr_f_b,q_conv_f,flowtype$,m_turb,h_f_b) m_turb_h=m_turb*3600 {conduction through absorber tube} q_cond_a=2*pi*kappa_a*(t_2-t_3)/ln(od_a/id_a) {convection between absorber tube and glass envelope} T_std=25 [c] "standard temperautre" ALL onv_a_g(t_4,t_3,p_air_34:q_conv_a) {radiation between absorber tube and glass envelope, based on J.P.Holman, "heat transfer", 8th edition, P430} 3-10

167 T_3_k=T_3+T_zero# T_4_k=T_4+T_zero# q_rad_a= sigma#*pi*od_a*(t_3_k^4-t_4_k^4)/(1/epsilon_a+od_a/id_g*(1/epsilon_g-1)) {conduction through glass envelope} q_cond_g= 2*PI*KAPPA_g*(T_4-T_5)/LN(OD_g/ID_g) {onvection from glass envelope to surrounding air} ALL onv_g_air(t_5,t_s_air,v_wind,od_g:q_conv_g_air,h_f_56) {Radiation from glass envelope to sky} T_5_k=T_5+T_zero# T_s_air_k=T_s_air+T_zero# T_sky_k=0.0552*T_s_air_k^1.5 T_sky=T_sky_k-T_zero# q_rad_g= sigma#*pi*od_g*epsilon_g*(t_5_k^4-t_sky_k^4) {Heat transferred by conduction through rectangle bracket cross-section} N_module=N_a*N_s "[-]" OD_brac=0.03 P_brac=0.08 A_cs_brac= k_brac=kappa_a ALL ond_bracket(t1_avg,t_s_air,v_wind:subq_cond_bracket,h_f_bracket) {Heat transferred by conduction through flange} L_flange=25/1000 [m] t_flange=20/1000 [m] r1=21/1000 [m] ALL ond_fin(t1_avg,t_s_air,v_wind,l_flange,t_flange,r1,kappa_a:q_cond_flange,q_c_flange,h_f_f lange) {Heat transferred by conduction through flexible end of solar collector} L_fle_end=7/1000 [m] t_fle_end=2/1000 [m] ALL ond_fin(t1_avg,t_s_air,v_wind,l_fle_end,t_fle_end,r1,kappa_a:q_cond_fle_end,q_c_fle_end, h_f_fle_end) {Heat transferred by conduction through hard end of solar collector} L_h_end=37/1000 [m] t_h_end=1/1000 [m] ALL ond_fin(t1_avg,t_s_air,v_wind,l_h_end,t_h_end,r1,kappa_a:q_cond_h_end,q_c_h_end,h_f_ h_end) {Heat transferred by conduction through connection flexible spring tube between solar modules} L_connect=10/1000 [m] t_connect=8/1000 [m] ALL ond_fin(t1_avg,t_s_air,v_wind,l_connect,t_connect,r1,kappa_a:q_fle_connect,q_c_connect, h_f_connect) 3-11

168 {although the bracket stainless steel is trapezoid, assumed that it is circumferential fin, the curve part bracket is negalitable in heat transfer view} L_bracket=73/1000 [m] t_bracket=1/1000 [m] ALL ond_fin(t1_avg,t_s_air,v_wind,l_bracket,t_bracket,r1,kappa_a:q_brac_c,q_c_brac_c,h_f_br ac) {Heat transferred by conduction through one module} {2mm is the space between the curves, there are total 13 curves at one end} q_fend=13*q_cond_fle_end+q_cond_h_end+q_c_h_end*13*2/1000 {4mm is the space between the curves, there are total 13 curves,40mm bare receiver tube } q_connect=13*q_fle_connect+q_c_connect*13*4/ /1000*q_c_flange q_hend=q_cond_h_end {25mm is the bare receiver tube at the short end} q_end_short=q_hend+q_cond_flange+25/1000*q_c_flange {35mm is the bare receiver tube at the long end} q_end_long=q_fend+q_cond_flange+35/1000*q_c_flange q_connect_sub=q_fend+q_end_short+35/1000*q_c_flange q_module_stru=q_end_short+2*q_connect_sub+q_end_long+4*subq_cond_bracket {305mm is the bare receiver tube at the end conneting to the flexible hose} q_array_end=360/1000*q_c_flange+q_cond_flange q_array_stru=n_a*q_module_stru+q_connect*(n_a-1)+2*q_array_end q_stru_total=n_s*q_array_stru q_cond_bracket=q_stru_total/l {Heat transferred by radiation from the connection pieces to the sky} EPSILON_stainlesssteel=0.79 OD_flange=92/1000 "[m]" OD_flex_1=60/1000 "[m]" OD_flex_2=65/1000 "[m]" OD_hend=120/1000 "[m]" LT_flange=5*2*2*20/1000 "[m]" L_flex_1=3*2*2*60/1000 "[m]" L_flex_2=2*65/1000 "[m]" L_hend=2*3*2*2*1/1000 "[m]" L_bare=2*2*360/1000+3*2*2*(35+25)/1000+2*40/1000 "[m]" q_rad_flange= sigma#*pi*od_flange*epsilon_stainlesssteel*(t_3_k^4-t_sky_k^4) q_rad_flex_1= sigma#*pi*od_flex_1*epsilon_stainlesssteel*(t_3_k^4-t_sky_k^4) q_rad_flex_2= sigma#*pi*od_flex_2*epsilon_stainlesssteel*(t_3_k^4-t_sky_k^4) q_rad_hend= sigma#*pi*od_hend*epsilon_stainlesssteel*(t_3_k^4-t_sky_k^4) q_rad_bare= sigma#*pi*od_a*epsilon_stainlesssteel*(t_3_k^4-t_sky_k^4) q_rad_connection=(q_rad_flange*lt_flange+q_rad_flex_1*l_flex_1+q_rad_flex_2*l_flex_2+q_r ad_hend*l_hend+q_rad_bare*l_bare)/l {Solar collector efficiency} q_heatloss=q_rad_g+q_conv_g_air+q_cond_bracket+q_rad_connection {Energy balance} q_solab_g+q_rad_a+q_conv_a=q_rad_g+q_conv_g_air q_cond_g=q_conv_a+q_rad_a {conduction from the absorber tube to the glass envelope is included in convection 3-12

169 calculation between the absorber tube and the glass envelope} q_solab_a=q_conv_a+q_rad_a+q_cond_a+q_cond_bracket+q_rad_connection q_cond_a=q_conv_f {absorbed by receiver including absorber tube and galss tube} q_sol_abd=q_solab_g+q_solab_a {solar energy deliveried} q_sol_deli=q_conv_f m1=m1_h/3600 {unit conversion} T1_avg=(T1_in+T1_out)/2 ALL Fluidproperties(T1_avg,P_water:p_f_b,rho_f_b,KAPPA_f_b,MU_f_b,P_f_b) m1*p_f_b*(t1_out-t1_in)=q_conv_f*l/1000 ETA=q_sol_deli/q_sol_total {pressure drop in one module} L_module=6 ALL Pressure_f(Dh_1,S_a,m1,T1_avg,P_water,e_a,ID_a,L_module :Re_sc,Lambda_sc,Delt_P, Delt_H_sc) {solar energy delivery from modules} T_out_modules=T1_out Q_out_modules=q_sol_deli*L/1000 kw" P_loss_modules=Delt_P*N_module H_loss_modules=Delt_H_sc*N_module "1000 is coversion factor ro convert W to {piping supply calculation} e_pipe= [m] {piping return calculation} ALL PipePro(pipe$,schedule$,Thickness_ins,T_p_retn:ID_pipe,OD_pipe,S_pipe,r_pipe_in,r_pipe_out, r_insult_out) T_p_retn=T_out_modules T_p_supy=T1_in ALL Pipeloss(pipematerial$,insulation$,T_p_retn,T_s_air,L_piping_r:k_pipe_retn,k_ins_retn,q_pipe_re tn,rcr_rent,t_1pr,t_2pr,t_3pr,t_4pr,r_01r,r_12r,r_23r,r_34r,r_total_retn) ALL Pipeloss(pipematerial$,insulation$,T_p_supy,T_s_air,L_piping_s:k_pipe_supy,k_ins_supy,q_pipe _supy,rcr_supy,t_1ps,t_2ps,t_3ps,t_4ps,r_01s,r_12s,r_23s,r_34s,r_total_supply) Q_loss_piping=(L_piping_s*q_pipe_supy+L_piping_r*q_pipe_retn)/1000 coversion factor ro convert W to kw" "1000 is ALL Pressure_f(ID_pipe,S_pipe,m1,T1_in,P_water,e_pipe,ID_pipe,L_piping_s:Re_sp,Lambda_sp,Del t_p_spiping, Delt_H_spiping) ALL Pressure_f(ID_pipe,S_pipe,m1,T_out_modules,P_water,e_pipe,ID_pipe,L_piping_r :Re_rp,Lambda_rp,Delt_P_rpiping,Delt_H_rpiping ) Delt_P_piping=Delt_P_rpiping+Delt_P_spiping Delt_H_piping=Delt_H_rpiping+Delt_H_spiping 3-13

170 {Overall performance} Q_overall=Q_out_modules-Q_loss_piping P_overall_loss=P_loss_modules+Delt_P_piping H_overall_loss=H_loss_modules+Delt_H_piping {for model presentation} Delt_T=(T1_out+T1_in)/2-T_s_air Delt_T_I=Delt_T/I_dn m1_vol=m1_h/rho_f_b 3-14

171 Appedix 4 Validation of the PTS model Appedix 4.1 Radiation transmission through covers and absorption by collectors The glass cover properties of the PTS play an important role in the PTS s efficiency. These properties, including transmittance, reflectance, and absorptivity, are functions of the incoming radiation, and glass thickness, refractive index, and the extinction coefficient of the material. When direct normal solar irradiation impinges on a solar collector with a glass cover, the absorbed solar radiation can be calculated as the Equation 4:1. S = * cosθ1 *( τα) (Equation 4:1) I DN Where, I DN : direct normal solar radiation in W/m2 θ 1: incident angle between direct normal solar irradiation and the normal of glass envelope (τα): transmittance-absorptance product of glass cover, of the radiation passing through the cover system and striking the plate, some is reflected back to the cover system. However, all this radiation is not lost since some of it is, in turn, reflected back to the plate. The value of (τα), for most practical solar collectors, is a reasonable approximation of 1.01 times of the product of τ times α. According to Bouguer s law, the absorbed radiation is proportional to the intensity and the distance traveled (x) in the medium and can be expressed as Equation 4:2 di = I DN * K * dx (Equation 4:2) Where, K: extinction coefficient, which is assumed to be constant in the solar spectrum. The value of K varies from 4 m -1 for water white glass to 32 m -1 for poor glass (which appears greenish when views on the edge). Integrating along the actual path length in glass yields τ = / cosθ 2 a e KL (Equation 4:3) Where, τ a : transmittance regarding to absorption losses 4-1

172 L: thickness of glass θ 2 : angle of refraction in the medium. It can be calculated based on Equation 4:4 n n 1 2 sinθ1 = sinθ 2 (Equation 4:4) Where, n1: index of refraction of air, 1 n2: index of refraction of medium, for glass, the average index of refraction of glass is

173 The simplified equation for the transmittance, absorptance, and reflectance of a collector cover can be calculated based on Equation 4:5, 4:6, and 4:7, respectively. τ τ a τ r (Equation 4:5) α 1 τ a (Equation 4:6) ρ τ a τ (Equation 4:7) Where, τ r : transmittance regarding to reflected losses, it is related the perpendicular component of unpolarized radiation, r and the parallel component of unpolarized radiation, r. Equation 4:8, 4:9, 4:10 can be used to find τ r, r and r, respectively. 1 1 r11 1 r τ r = [ + ] (Equation 4:8) 2 1+ r 1+ r 11 2 sin ( θ2 θ1) = 2 sin ( θ + θ ) 2 1 r (Equation 4:9) 11 2 tan ( θ2 θ1) = 2 tan ( θ + θ ) r (Equation 4:10) 2 1 Therefore, according to the extinction efficient and index of glass cover, for a certain incident angle, the transmittance, absorptance, and reflectance can be identified by addressed equations. 4-3

174 Appedix 5 Solar cooling and heating system simulation Appedix 5.1 Building simulation interface Figure 5:1 Building simulation interface 5-1

175 The thermal behavior of the building is modeled by the TRNSYS type To perform the load calculation, the building characteristics have to be defined in TRNBuild. A5.1.1 Geometry The net floor area of the south zone is about 245 m 2 (10.2 m * 24 m) and the average height is about 4.8 m 2. The building has horizontal shadings (catwalk) on the east and west facades (1m * 24 m). The whole zone is divided in 5 bays (approximate 49 m 2 / bay). The quasi open space is subdivided by partition walls and furniture in about two times five areas (office or conference zones approximate 22m 2 ) on the east and west façade. The hallway in the middle zone is separated to the zones only by open partition walls and furniture. As the IW south part is an open space office, the model in TRNBuild is a single zone model. Figure 5:2 shows the roof plan of the IW and the IW s orientation. Solar ollectors Building North 15 Location of chiller and control box 57'7"(17500mm) 39'4"(12000mm) 31'-6" (9600mm) 9@15'9" (9@ 4800 mm) 17'5"(5310mm) Figure 5:2 IW roof plan and orientation 1 The level of detail of TRNSYS building model is compliant with the requirements of ANSI/ASHRAE standard The average height includes the raised floor and average height of roof. 5-2

176 A5.1.2 Materials There are three different types of wall: roof: WALL ROOF ; floor : WALL BST_H_FLO ; vertical wall: WALL 001. And there are two types of windows. The specifications of these walls and windows are described in the Table 5:1. Table 5:1 Specifications of the material in the building Vertical wall WALL 001 Insulated metal panel Layers Thickness (m) U value (W/m 2.K) Inside surface resistance - Aluminum or steel siding R-19 bat insulation Aluminum or steel siding Outside surface resistance - Roof WALL ROOF Insulated metal panel Layers Thickness (m) U value (W/m 2.K) Inside surface resistance - Steel deck R-38 bat insulation Build up roofing 0.01 Outside surface resistance - Floor WALL BST_H_FLO Layers Thickness (m) U value (W/m 2.K) oncrete stab Resistance - U-value (BTU/hr*ft2 F) U-value (W/m2K) Shade coefficient g-value (solar heat gain coefficient) Window Window A5.1.3 Simulation of the ventilation A Description of the ventilation The ventilation deals with the sensible, latent loads and hygienic air change. onditioned outside air is supplied by a desiccant wheel unit (SEMO unit) via a main duct placed in the center of the plenum distributed to each zone and controlled by a constant volume device. The outside air volume rate is about 20 cfm (34m 3 /h) per person based on the ASHRAE standards 3. The outside 3 A regular office, without point sources (copiers, laser printers, faxes) may be adequately supplied with 20 cfm per person. 5-3

177 air is supplied throughout whole year by under floor air system with a proper relative humidity to maintain the room conditions in the comfort range. The conditioned outside air properties are listed in Table 5:2. Table 5:2 onditioned outside air properties onditioned outside air properties Temperature ( ) Relative humidity (%) Heating period ooling period A Model of the ventilation The SEMO unit is modeled by the TESS type 696 component. Type696 models a simple air conditioning device that adds or removes sensible and latent energy from an air stream to meet specified set point conditions of temperature and or humidity. In this device, the sensible condition controls the latent decisions. In other words the device cannot heat and dehumidify or cool and humidify the air stream. It can, however, heat and humidify or cool and dehumidify. A Simulation of the air conditioning ASHRAE Standard Thermal Environmental onditions for Human Occupancy recommends: temperature ranges of 20 to 23.5 elsius for winter and 23 to 26 elsius for summer at 50% relative humidity; relative humidity levels of 30% to 60%. During the heating period (from 16 th October to 15 th April), the room set point temperature is 20 for the day time period (7:00 AM - 8:00 PM every day). During the night, the set back temperature is 15 and the air room temperature decreases depend on the outside conditions. During the cooling period (from 16 th April to 15 th October), the room set point temperature is 22 for the day time period (7:00 AM - 8:00 PM every day). During the night, the set back temperature is 40 and the air room temperature increases or decreases depend on the outside conditions. To meet these room temperature requirements, the air conditioning system heats and cools the room. TRNSYS calculates the heating and cooling loads required to meet the demand. 5-4

178 Appedix 5.2 Solar heating evaluation simulation Figure 5:3 Interface of solar heating evaluation simulation 5-5

179 This simulation includes the PTSs, two heat exchangers and two circulation pumps. Figure 5:3 is the snap shot of solar heating evaluation model. A5.2.1 Definition of the components in the model A Pipes in the solar collection loop The pipe is used only for the time delay effect. The input of pipe for the model is given in table 5:3. Table 5:3 Pipe specification in the evaluation simulation Solar collection loop pipe (Type 709) Inside diameter Outside diameter Pipe length Pipe thermal conductivity Fluid density Fluid specific heat Fluid thermal conductivity Fluid viscosity Initial fluid temperature Number of nodes Insulation thickness onductivity of insulation [m] [m] 55 [m](return line) / 50 [m](supply line) 216 [kj/hr.m.k] [kg/m^3] [kj/kg.k] [kj/hr.m.k] [kg/m.hr] 8 [] [m] [kj/hr.m.k] Outer surface convection coefficient Loss temperature for node 0 [kj/hr.m^2.k] Tam (from the weather file) In order to simulate the system heat capacity, a thermal storage tank was included in the model. Its specification is given in table 5:4. Table 5:4 Thermal tank specification in the evaluation simulation Thermal storage tank - horizontal cylinder (Type 60I) Tank volume Tank height Height of flow inlet 1 Height of flow outlet [m 3 ] [m] [m] [m] Fluid specific heat Fluid density Tank loss coefficient Fluid thermal conductivity Destratification conductivity Boiling temperature [kj/kg.k] [kg/m^3] 15 [kj/hr.m2.k] [kj/hr.m.k] 0 [kj/hr.m.k] 300 [ ] 5-6

180 A irculation pump in the solar collection loop Solar main circulation pump is using a constant speed pump. Type 114 was used in the simulation. It sets the downstream flow rate based on its rated flow rate parameter and the current value of its control signal input. The specification of the pump is given in table 5:5. Table 5:5 Solar pump specification in the evaluation simulation Solar pump (Type 114) Rated flow rate Fluid specific heat Rated power Motor heat loss fraction 1372 kg/hr [kj/kg.k] 2520 [kj/hr] 0 - Overall pump efficiency Motor efficiency A Heat exchanger between solar loop and load loop The HX-2 is a brazed plate heat exchanger, which offers the highest level of thermal efficiency and durability in a compact, low cost unit. The corrugated plate design provides very high heat transfer coefficients, resulting in a more compact design. According to the specification sheet, overall heat transfer coefficient, service is 359 Btu/hr-ft2-F, which is equal to 2.05 kw/m^2- (8560 kj/h.k). the specification of HX2 is given in table 5:6. Table 5:6 HX - 2 specification in the evaluation model Heat exchanger (counter flow) (Type 5b) ounter flow mode Specific heat of hot side fluid Specific heat of cold side fluid [kj/kg.k] Overall heat transfer coefficient of exchanger 8600 [kj/h.k] A Parabolic Trough Solar collector The PTS is one-axis tracking solar collector. It is basically comprised of a reflector mirror, a receiver tube, supporting structure, and tracking mechanism. Its concentrate rate is 19.6 (the aperture area is m^2 and receiver area is 0.68 m^2.). The heat transfer fluid flows from one end of the absorber tube to another to convert solar energy into thermal energy. 5-7

181 It is able to provide a high temperature fluid for both industry and residential thermal usage. The specification of modeled PTS is given in Table 5:7. Table 5:7 PTS specifications in the evaluation simulation Linear Parabolic oncentrator Solar ollector (Type 230 based on type 536) Number of collectors in series 4 Number of collectors in parallel 1 Aperture area [m 2 ] oncentration ratio 22 Intercept efficiency (FrTan) Efficiency slope (FrUl) kj/hr. m^2 K Fluid specific heat kj/kg.k (at 54) Number of IAM points 8 Tested flow rate kg/hr.m^2 (Flow rate/aa of one module) Number of modules in an array Number of arrays Parabola focal length Distance between adjacent arrays Max, outlet temperature IAM [m] 4.8 [m] Heating: 95 0,10,20,30,40,50,60,70/1,0.984,0.93,0.84,0.715,0.559,0.376,0.169 A irculation pump in the load loop Main circulation pump in the load loop is a constant speed pump. Type 114 was used in the simulation. The specification of the pump is given in Table 5:8. Table 5:8 Loop pump specification in the evaluation model Load pump (Type 114) Rated flow rate Fluid specific heat Rated power Motor heat loss fraction 1480 kg/hr [kj/hr] 0 - Overall pump efficiency Motor efficiency A Heat exchanger as false load The HX - 1 as the false load consists of one shell pass and two tube passes. It is water-to-water heat transfer style. Heat transfer area is 0.68 m 2 and overall heat transfer coefficient is 1.06 kw/m 2, which is proved accurate by experimental data. 5-8

182 The specification of this heat exchanger is given in Table 5:9. Table 5:9 Heat exchanger 1 in the evaluation simulation Heat exchanger as false load (Type 5g) Shell and Tube mode Specific heat of hot side fluid Specific heat of cold side fluid Number of Shell Passes [kj/kg.k]@ [kj/kg.k]@7 1 Overall heat transfer coefficient of exchanger 2600 [kj/h.k] A Pipes in the load loop The pipe is used in the load loop. Since the length of the load pipe is not very long, the heat capacity of the load loop did not concern specially. The input of pipe in the load loop is given in table 5:10. Table 5:10 Specification of the pipe in the load loop for the evaluation simulation Load loop pipe (Type 709) Inside diameter Outside diameter Pipe length Pipe thermal conductivity Fluid density Fluid specific heat Fluid thermal conductivity Fluid viscosity Initial fluid temperature Number of nodes Insulation thickness onductivity of insulation [m] [m] 5 [m] 1436 [kj/hr.m.k] [kg/m^3] 4.18 [kj/kg.k] [kj/hr.m.k] 2.15 [kg/m.hr] 20 [] [m] [kj/hr.m.k] Outer surface convection coefficient Loss temperature for node 10 [kj/hr.m^2.k] Tam (from the weather file) A Solar pump control and load pump control Based on the operation of experiment, the schedule of the solar pump followed operational procedure from 9:37 AM to 3:22 PM On March 02, 07. The load pump was controlled by a differential controller. When the outlet temperature of the solar field is greater than 93, load pump is turned on till the outlet temperature is lower than

183 Appedix 5.3 Solar cooling evaluation simulation Figure 5:4 Solar cooling evaluation mode 5-10

184 Appedix 5.4 Solar heating base-case Figure 5:5 Solar heating base-base simulation interface 5-11

185 Appedix 5.5 Solar cooling base-case Figure 5:6 Solar cooling base-case simulation interface 5-12

186 Appedix 5.6 ode of parabolic trough solar collector SUBROUTINE TYPE230 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Linear Parabolic oncentrator Solar ollector IISiBat Model: Type230 DESRIPTION: THIS SUBROUTINE MODELS A LINEAR PARABOLI ONENTRATING SOLAR OLLETOR. THIS SUBROUTINE REQUIRES THE DYNAMIDATA UTILITY SUBROUTINE TO DETERMINE THE INIDENE ANGLE MODIFIERS THAT ARE INTERPOLATED FROM A USER-SUPPLIED DATA FILE. THE USEFUL ENERGY EQUATION AND MODIFYING FUNTIONS FROM "SOLAR ENGINEERING OF THERMAL PROESSES" BY DUFFIE AND BEKMAN. SPEIFIALLY HAPTER 7 OF THE SEOND EDITION. Editor: Ming Qu Date: 1/1/97 LAST UPDATED:: May 2007 The Linear Parabolic oncentrator model is based on equations taken from Duffie and Beckman's "Solar Engineering of Thermal Processes" Ming Qu modified TESS type 536 in three aspects 1. when FR_solar is equal to 0 and there is direct solar radiation, it is assumped that the PTS is not tracking, so that the outlet temperature of solar collector is equal to the ambient temperature. 2. add the end-loss into model due to the incident angle existed 3. add the shadow loss caused by the adjacent solar collector array *** *** Model Parameters *** Number of collectors in series - [1;+Inf] Number of collectors in parallel - [1;+Inf] Aperture length m [0.0;+Inf] Aperture width m [0;+Inf] oncentration ratio - [0.0;+Inf] Intercept efficiency (FrTan) - [0.0;1.0] Efficiency slope (FrUl) kj/hr.m^2.k [-Inf;+Inf] Fluid specific heat kj/kg.k [0.0;+Inf] Logical unit - [10;30] Number of IAM points - [2;10] Tested flow rate kg/s [0.0;+Inf] Number of modules in an array - [1;+Inf] Number of arrays - [1;+Inf] Parabola focal length m [0;+Inf] Distance between adjacent arrays m [0;+Inf] *** *** Model Inputs *** Inlet temperature [-Inf;+Inf] Inlet flow rate kg/hr [0.0;+Inf] Ambient temperature [-Inf;+Inf] 5-13

187 Incident beam radiation kj/hr.m^2 [0.0;+Inf] Incidence angle degrees [0.0;90.0] Maximum outlet temperature [-Inf;+Inf] Tracking angle degrees [0.0;90.0] ****** *** *** Model Outputs *** Outlet temperature [-Inf;+Inf] Outlet flow rate kg/hr [0.0;+Inf] Useful energy gain kj/hr [-Inf;+Inf] Dumped energy kj/hr [0.0;+Inf] Theoretical temperature [-Inf;+Inf] Length of the end loss m [0;+Inf] Height of the shadow from the adjacent array m [0;+Inf] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE230!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS IMPLIIT NONE DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NP,NI,NOUT,ND,IUNIT,ITYPE,INTRL HARATER*3 YHEK,OHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=15,NI=7,NOUT=8,ND=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NOUT),PAR(NP),YHEK(NI),OHEK(NOUT) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES DOUBLE PREISION BEAM,X(1),Y(1),XNS,XNP,AAP_TOTAL,ON,FRTAN,FRUL DOUBLE PREISION GTEST,TIN,FLW,TAMB,THETA,TMAX,XKAT,FTEST,FPUL,RTEST DOUBLE PREISION QU,TAL,TOUT,QDUMP,LFOAL,LSPAN,HSHADOW,AAL,AAW 5-14

188 DOUBLE PREISION RDONV,XK,R2,AAP,PF,LENDLOSS,R1 DOUBLE PREISION TANGLE,SANGLE,OTANGLE INTEGER NX(1),LU,NPOINT,NUM_S,NUM_P,NUM_M,NUM_A ONSTANTS REQUIRED BY THE MODEL DATA RDONV/ D0/ GET GLOBAL TRNSYS SIMULATION VARIABLES TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN DO ALL THE VERY LAST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(8).EQ.-1) THEN RETURN PERFORM ANY "AFTER-ITERATION" MANIPULATIONS THAT ARE REQUIRED IF(INFO(13).GT.0) THEN RETURN DO ALL THE VERY FIRST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(7).EQ.-1) THEN RETRIEVE THE UNIT NUMBER AND TYPE NUMBER FOR THIS OMPONENT FROM THE INFO ARRAY IUNIT=INFO(1) ITYPE=INFO(2) SET SOME INFO ARRAY VARIABLES TO TELL THE TRNSYS ENGINE HOW THIS TYPE IS TO WORK INFO(6)=NOUT INFO(9)=1 INFO(10)=0!STORAGE FOR VERSION 16 HAS BEEN HANGED ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS OMPONENT REQUIRES TO WHAT IS SUPPLIED IN THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND) SET THE YHEK AND OHEK ARRAYS TO ONTAIN THE ORRET VARIABLE TYPES FOR THE INPUTS AND 5-15

189 OUTPUTS DATA YHEK/'TE1','MF1','TE1','IR1','DG1','TE1','DG1'/ DATA OHEK/'TE1','MF1','PW1','PW1','TE1','LE1','LE1'/ ALL THE RHEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT TYPES FOR THIS OMPONENT ALL RHEK(INFO,YHEK,OHEK) ALL LINKK TO TELL THE UISER THIS TYPE REQUIRES TAU_ALPHA ALL LINKK('TYPE 536',' TAU_ALPHA',4,INFO(1)) RETURN TO THE ALLING PROGRAM RETURN DO ALL OF THE INITIAL TIMESTEP MANIPULATIONS HERE - THERE ARE NO ITERATIONS AT THE INTIAL TIME IF (TIME.LT.(TIME0+DELT/2.D0)) THEN SET THE UNIT NUMBER FOR FUTURE ALLS IUNIT=INFO(1) READ IN THE VALUES OF THE PARAMETERS IN SEQUENTIAL ORDER NUM_S=JFIX(PAR(1)+0.1) XNS=DBLE(NUM_S) NUM_P=JFIX(PAR(2)+0.1) XNP=DBLE(NUM_P) AAL=PAR(3) AAW=PAR(4) ON=PAR(5) FRTAN=PAR(6) FRUL=PAR(7) PF=PAR(8) LU=JFIX(PAR(9)+0.1) NPOINT=JFIX(PAR(10)+0.1) GTEST=PAR(11) NUM_M=JFIX(PAR(12)+0.1) NUM_A=JFIX(PAR(13)+0.1) LFOAL=PAR(14) LSPAN=PAR(15) HEK THE PARAMETERS FOR PROBLEMS AND RETURN FROM THE SUBROUTINE IF AN ERROR IS FOUND IF(NUM_S.LT.1) ALL TYPEK(-4,INFO,NI,1,0) IF(NUM_P.LT.1) ALL TYPEK(-4,INFO,NI,2,0) IF(AAL.LE.0.) ALL TYPEK(-4,INFO,NI,3,0) IF(AAW.LE.0.) ALL TYPEK(-4,INFO,NI,4,0) IF(ON.LE.0.) ALL TYPEK(-4,INFO,NI,5,0) IF(FRTAN.LE.0.) ALL TYPEK(-4,INFO,NI,6,0) IF(FRUL.LE.0.) ALL TYPEK(-4,INFO,NI,7,0) IF(PF.LT.0.) ALL TYPEK(-4,INFO,NI,8,0) IF(LU.LT.10) ALL TYPEK(-4,INFO,NI,9,0) IF(NPOINT.LT.2) ALL TYPEK(-4,INFO,NI,10,0) IF(GTEST.LE.0.) ALL TYPEK(-4,INFO,NI,11,0) IF(NUM_M.LT.1) ALL TYPEK(-4,INFO,NI,12,0) IF(NUM_A.LT.1) ALL TYPEK(-4,INFO,NI,13,0) IF(LFOAL.LE.0.) ALL TYPEK(-4,INFO,NI,14,0) IF(LSPAN.LE.0.) ALL TYPEK(-4,INFO,NI,15,0) 5-16

190 PERFORM ANY REQUIRED ALULATIONS TO SET THE INITIAL VALUES OF THE OUTPUTS HERE OUT(1)=XIN(1) OUT(2)=0. OUT(3)=0. OUT(4)=0. OUT(5)=0. OUT(6)=0. OUT(7)=0. OUT(8)=0. RETURN TO THE ALLING PROGRAM RETURN *** ITS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED IF(INFO(1).NE.IUNIT) THEN RESET THE UNIT NUMBER IUNIT=INFO(1) ITYPE=INFO(2) READ IN THE VALUES OF THE PARAMETERS IN SEQUENTIAL ORDER NUM_S=JFIX(PAR(1)+0.1) XNS=DBLE(NUM_S) NUM_P=JFIX(PAR(2)+0.1) XNP=DBLE(NUM_P) AAL=PAR(3) AAW=PAR(4) ON=PAR(5) FRTAN=PAR(6) FRUL=PAR(7) PF=PAR(8) LU=JFIX(PAR(9)+0.1) NPOINT=JFIX(PAR(10)+0.1) GTEST=PAR(11) NUM_M=JFIX(PAR(12)+0.1) NUM_A=JFIX(PAR(13)+0.1) LFOAL=PAR(14) LSPAN=PAR(15) RETRIEVE THE URRENT VALUES OF THE INPUTS TO THIS MODEL FROM THE XIN ARRAY IN SEQUENTIAL ORDER TIN=XIN(1) FLW=XIN(2) TAMB=XIN(3) BEAM=XIN(4) THETA=XIN(5) TMAX=XIN(6) TANGLE=XIN(7) 5-17

191 HEK THE INPUTS FOR PROBLEMS IF(FLW.LT.0.) ALL TYPEK(-3,INFO,2,0,0) IF(BEAM.LT.0.) ALL TYPEK(-3,INFO,4,0,0) IF(ERRORFOUND()) RETURN PERFORM ALL THE ALULATION HERE FOR THIS MODEL. FIND THE LENGTH OF REEIVER TUBE WITH NO SUN RAY DUE TO ENDLOSS SANGLE=DASIN(AAW/LSPAN) OTANGLE=(90.-TANGLE)*RDONV IF(SANGLE.GT.OTANGLE) THEN HSHADOW=AAW-LSPAN*DSIN(OTANGLE) ELSE HSHADOW= 0. LENDLOSS=LFOAL*DTAN(THETA*RDONV) IF (LENDLOSS.GT.NUM_M*AAL)THEN AAP_TOTAL = 0 ELSE AAP_TOTAL=NUM_A*(NUM_M*AAL-LENDLOSS)*(AAW-HSHADOW) AAP=AAL*AAW DETERMINE INIDENE ANGLE MODIFIER FROM ALL TO DATA IF((BEAM.GT.0.).AND.(THETA.LE.90.)) THEN X(1)=THETA NX(1)=NPOINT ALL DYNAMIDATA(LU,1,NX,1,X,Y,INFO,*10) ALL LINKK('TYPE 230','DYNAMIDATA',1,99) 10 IF(ERRORFOUND()) RETURN 1 XKAT=Y(1) IF(XKAT.LT.0.) XKAT=0. ELSE XKAT=0. HEK TO SEE IF THERE IS FLOW IF(FLW.GT.0.) THEN ALULATE F'UL FTEST=FRUL/GTEST/PF/ON IF(FTEST.GE.1.) THEN FPUL=FRUL ELSE FPUL=-GTEST*PF*DLOG(1.-FRUL/GTEST/PF/ON) DETERMINE MODIFIERS FOR OFF-TEST FLOW RATE AND OLLETORS IN SERIES RTEST=GTEST*PF*(1.-DEXP(-FPUL/GTEST/PF)) R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP)/PF)) 1 /RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK 5-18

192 ALULATE USEFUL ENERGY GAIN AND OLLETOR TEMP QU=R1*R2*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) TAL=QU/FLW/PF+TIN TOUT=DMIN1(TAL,TMAX) QDUMP=FLW*PF*(TAL-TOUT) ELSE TAL=TAMB TOUT=DMIN1(TAL,TMAX) TAL=TAMB+FRTAN*XKAT*BEAM*ON/FRUL TOUT=DMIN1(TAL,TMAX) QU=0. QDUMP=0. FLW= SET THE OUTPUTS FROM THIS MODEL IN SEQUENTIAL ORDER AND GET OUT OUT(1)=TOUT OUT(2)=FLW OUT(3)=QU OUT(4)=QDUMP OUT(5)=TAL OUT(6)=LENDLOSS OUT(7)=HSHADOW OUT(8)=AAP_TOTAL EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END Appedix 5.7 ode of main control of solar cooling base-case SUBROUTINE TYPE235 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE235 Author: Ming Qu Editor: Date: TRNSYS 7.5 last modified: Jun 2007 NOTE: This controller can only be used with Solver 0 (Successive substitution) *** 5-19

193 *** Model Parameters *** No. of oscillations - [1;+Inf] High limit cut-out of NIP kj/hr.m^2 [-Inf;+Inf] High limit cut-out of circulating fluid temperature [-Inf;+Inf] *** *** Model Inputs *** Direct normal solar radiation on horizontal surface kj/hr.m^2 [-Inf;+Inf] Upper dead band of NIP kj/hr.m^2 [-Inf;+Inf] Lower dead band of NIP kj/hr.m^2 [-Inf;+Inf] Input control function of solar radiation - [0;1] ooling load kj/hr [-Inf;+Inf] HTF temperature at the inlet of the chiller [-Inf;+Inf] Upper dead band of inlet temperature [-Inf;+Inf] Lower dead band of inlet temperature [-Inf;+Inf] Input control function of inlet temperature - [0;1] hilled water supply temperature of HW chiller in the last timestep [-Inf;+Inf] *** *** Model Outputs *** Output control signal of NIP - [0.0;1.0] ounter of stick in solar raidiation control - [0.0;1.0] Osillation number of solar radiation control - [0.0;1.0] Lastcall output of solar radiation control - [0.0;1.0] Output control signal of HT ctrl - [0.0;1.0] ounter of stick in temperature control - [0.0;1.0] Osillation number of the temperature control - [0.0;1.0] Lastcall output of temperature control - [0.0;1.0] Output control of bypass - [0.0;1.0] Output control of chiller - [0.0;1.0] Output control of fired chiller - [0.0;1.0] Output control of cooling load - [0.0;1.0] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE235!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE 5-20

194 PARAMETER (NP=3,NI=11,NO=14,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE LOAL VARIABLE DELARATIONS HARATER*12 INTStr HARATER*160 WarnMsg INTEGER NSTK,!No of_oscillations 1 INT_1,INT_2, 1 NTOLD_1,NTOLD_2, 1 IOS_1,IOS_2!Osillation number DOUBLE PREISION RMAX,!direct normal solar radiaition max value 1 TMAX,!max temperature at inlet of the chiller 1 LAST_1,LAST_2, 1 NIP,!Direct_normal_solar_radiation_on_horizontal_surface 1 UP_NIP,!Upper_dead_band_of_NIP 1 LOW_NIP,!Lower_dead_band_of_NIP 1 LOAD,!cooling load 1 T_chillerin,!HTF_temperature_at_the_inlet_of_the_chiller 1 UP_T,!Upper_dead_band_of_inlet_temperature 1 LOW_T,!Lower_dead_band_of_inlet_temperature 1 LOAD_TRL,!LOAD ONTROL SIGNAL 1 HILLER_TRL,!HW HILLER ONTROL 1 T_HL_PREV_O,!OUTLET TEMPERATURE OF HW FROM THE HILLER 1 IF_1, IF_2,!ONTROL FUTION INPUTS 1 T_HW_SETPOINT,!HILLED WATER SETPOINT 1 R_BYPASS,!RATIO OF HTF OVER THE BYPASS 1 R_LDIVERTER!RATIO OF HW OVER THE HW HILLER DATA STATEMENTS DATA YHEK/'IR1','IR1','IR1','F1','PW1','TE1','TE1','TE1', 1 'F1','TE1','TE1'/ DATA OHEK/'F1','DM1','DM1','F1','F1','DM1','DM1','F1', 1 'F1','F1','F1','F1','F1','F1'/ TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN PERFORM LAST ALL MANIPULATIONS IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN 1 INT_1=JFIX(OUT(2)*DELT/(TFINAL-TIME0)*100.d0) INT_2=JFIX(OUT(6)*DELT/(TFINAL-TIME0)*100.d0) 5-21

195 IF(INT_1.GE.10) THEN WRITE (INTStr,*) INT_1 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) IF(INT_2.GE.10) THEN WRITE (INTStr,*) INT_2 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) RETURN FROM THIS MODEL AS NO "AFTER-ONVERGENE" MANIPULATIONS ARE REQUIRED IF(INFO(13).GT.0) RETURN PERFORM FIRST ALL MANIPULATIONS IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required!set THE PROPER NUMBER OF PARAMETERS AORDING TO THE MODE NSTK=JFIX(PAR(1)+0.01) IF(NSTK.LT.0) ALL TYPEK(4,INFO,0,1,0)!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN 1 PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN!set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values NSTK = JFIX(PAR(1)+0.01) RMAX = PAR(2) TMAX = PAR(3) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** 5-22

196 RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) ITYPE=INFO(2)!reread the parameter values NSTK=JFIX(PAR(1)+0.01) RMAX = PAR(2) TMAX = PAR(3) PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of solar radiation control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of solar radiation control OUT(7),A.K.A. IOS_2, IS STIK OUNTER of by-pass valve control OUT(6) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of by-pass valve control LEAR STIK OUNTER ON FIRST ALL OF T-S IF(INFO(7).EQ.0) THEN OUT(3)=0.d0 OUT(7)=0.d0 NIP = XIN(1) UP_NIP = XIN(2) LOW_NIP = XIN(3) IF_1= XIN(4) LOAD = XIN(5) T_chillerin = XIN(6) UP_T = XIN(7) LOW_T = XIN(8) IF_2= XIN(9) T_HL_PREV_O = XIN(10) T_HW_SETPOINT = XIN(11) ONTROL SOLAR PUMP IT ONTROL NTOLD_1=OUT(4)! control signal for the last time step LAST_1=OUT(1) IOS_1=JFIX(OUT(3)+0.01) IF(IOS_1.EQ.NSTK) OUT(2)=OUT(2)+1.d0 IF(IOS_1.GE. NSTK) GO TO 15 IF(INFO(7).EQ.0) THEN IF (XIN(4).GT. 0.5) THEN NTOLD_1 = 1 OUT(4)=DBLE(NTOLD_1) ELSE NTOLD_1 = 0 OUT(4)=DBLE(NTOLD_1) 5-23

197 IF (NTOLD_1.GT.0.5) GOTO 10!OUTPUT WAS 0 LAST ALL OUT(1)=0.d0 IF (NIP.GT.UP_NIP) OUT(1)=1.d0 IF (NIP.GT.4680.) OUT(1)=0.d0 GO TO 15!OUTPUT WAS 1 LAST ALL 10 OUT(1)=1.d0 IF (NIP.LT.LOW_NIP) OUT(1)=0.d0 IF (NIP.GT.4680.) OUT(1)=0.d0 15 IF((ABS(LAST_1-OUT(1)).LT.1.E-06).AND.(IOS_1.NE.NSTK)) THEN OUT(3)=OUT(3) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(3)=OUT(3)+1.d0 LOADS ONTROL IF (LOAD.LE.0.) THEN LOAD_TRL = 0. ELSEIF (LOAD.GT.0) THEN LOAD_TRL = 1. ******************************** ONTROL HT at inlet of chiller ******************************** TEMPERATURE ONTROL NTOLD_2=OUT(8)! control signal for the last time step LAST_2=OUT(5) IOS_2=JFIX(OUT(7)+0.01) IF(IOS_2.EQ.NSTK) OUT(6)=OUT(6)+1.d0 IF(IOS_2.GE.NSTK) GO TO 25 IF(INFO(7).EQ.0) THEN IF (XIN(9).GT.0.5) THEN NTOLD_2 = 1 OUT(8)=DBLE(NTOLD_2) ELSE NTOLD_2 = 0 OUT(8)=DBLE(NTOLD_2) IF (NTOLD_2.GT.0.5) GOTO 20!OUTPUT WAS 0 LAST ALL OUT(5)=0.d0 5-24

198 IF (T_chillerin.GT.UP_T) OUT(5)=1.d0 IF (T_chillerin.GT.175.) OUT(5)=0.d0 GO TO 25!OUTPUT WAS 1 LAST ALL 20 OUT(5)=1.d0 IF (T_chillerin.LT.LOW_T) OUT(5)=0.d0 IF (T_chillerin.GT.175.) OUT(5)=0.d0 25 IF((ABS(LAST_2-OUT(5)).LT.1.E-06).AND.(IOS_2.NE.NSTK)) THEN OUT(7)=OUT(7) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(7)=OUT(7)+1.d0 IF THE OUTLET TEMPERATURE OF HILLER IS HIGHER THATN 14 IF(T_HL_PREV_O.LT.13.1) THEN HILLER_TRL = OUT(5)*LOAD_TRL*OUT(1) ELSE HILLER_TRL = 0.0 BYPASS ONTROL BASED ON THE HILLED WATER OUTLET TEMPERATURE IF(HILLER_TRL.GE. 0.5) THEN IF(T_HL_PREV_O.GE.(T_HW_SETPOINT+4.0)) THEN R_BYPASS=0.0 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+3.0)) THEN R_BYPASS=0.2 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+2.0)) THEN R_BYPASS=0.4 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT-1.0)) THEN R_BYPASS=0.65 ELSE R_BYPASS=1.0 ELSE R_BYPASS=1.0 OUTPUT DELARATION OUT(9) = OUT(1)- HILLER_TRL OUT(10)= HILLER_TRL OUT(11)= LOAD_TRL - HILLER_TRL OUT(12)= LOAD_TRL OUT(13)= R_BYPASS OUT(14)= HILLER_TRL EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END

199 Appedix 6 System optimization and sensitivity analysis Appedix 6.1 Solar heating system with constant-outlet-temperature control Figure 6:1Interface of solar heating system with constant-outlet-temperature control 6-1

200 Appedix 6.2 Solar cooling system with constant-outlet-temperature control Figure 6:2 Interface of solar cooling system with constant-outlet-temperature control 6-2

201 Appedix 6.3 Solar heating system with storage tank Figure 6:3 Interface of solar heating system with storage tank 6-3

202 A6.3.1 ontrol of solar heating system with storage tank Table 6:1 ontrol of solar heating system with storage tank for shifting energy Mode Solar collection pump A. On onditions Direct normal solar radiation NIP >300 W/m 2, Differential controller ON Output S5 S6 HX2 Heater ON B. Off Storage tank A. harging ON B. harging OFF Off 250 NIP<250 W/m^2 300 W/m^2 T_sr_out > T_st_bottom_out + 4 T_sr_out < T_st_bottom_out + 3 OFF. Discharging Heater A. ON T_st_top_out > T_lpump_out +3 Load is ON T_mixer_out < 45 and There is load Load is ON ON ON ON B. OFF T_st_top_out > T_lpump_out +3 OFF Where, T_sr_out: the outlet temperature of the solar collector T_st_bottom_out: the outlet temperature at the bottom of storage tank T_st_top_out: the outlet temperature on the top of storage tank T_lpump_out: the outlet temperature of the fluid from the pump in the load loop 6-4

203 Appedix 6.4 Solar cooling system with storage tank for shifting energy Figure 6:4 Interface of solar cooling system with storage tank for shifting energy 6-5

204 A6.4.1 ontrol of solar cooling system with storage tank for shifting energy Table 6:2 ontrol of solar cooling system with storage tank for shifting energy Mode Solar collection pump A. On onditions Direct normal solar radiation NIP >300 W/m 2, Differential controller ON Output S5 HWchiller VFD pump ON Off B. Off Storage tank A. harging ON 250 NIP<250 W/m^2 FR_sr >FR_chiller 300 W/m^2 OFF B. Discharging FR_sr >FR_chiller or T_sr_out < 130 ON ON ON Off T_st_top 130 Load is ON 155 Where, T_sr_out: the outlet temperature of the solar collector FR_sr: the flow rate of the solar collection loop FR_chiller: the flow rate required by the absorption chiller T_st_top: the temperature at the top of the storage tank NIP: the direct normal solar radiation 6-6

205 Appedix 6.5 Solar cooling system with storage tank for preheat Figure 6:5 Interface of solar cooling system with storage tank for preheat 6-7

206 A6.5.1 ontrol of solar cooling system with storage tank for preheat Table 6:3 ontrol of solar cooling system with storage tank for preheat Mode Solar collection pump A. On onditions Direct normal solar radiation NIP >300 W/m 2, Differential controller ON Output Solar main S7 VFD pump ON Off B. Off Storage tank A. harging 250 NIP<250 W/m^2 300 W/m^2 Solar loop at last time step is ON Solar loop at current time step is OFF T_st_top < T_H_AVE-3 OFF OFF ON B. Discharging Solar loop at last time step is OFF Solar loop at current time step is ON T_st_bottom -3 > T_H_AVE ON OFF Or Discharging is ON Solar loop is ON Mass in the storage tank has not fully been discharged Where, T_st_top: the temperature at the top of the storage tank T_st_bottom: the temperature at the bottom of the storage tank T_H_AVE: the average temperature of the HTF in the pipeline NIP: the direct normal solar radiation 6-8

207 Appedix 6.6 Solar heating system with auxiliary heater for preheat Figure 6:6 Interface of solar heating system with auxiliary heater for preheat 6-9

208 Appedix 6.7 Solar cooling with auxiliary heater for preheat Figure 6:7 Interface of solar cooling with auxiliary heater for preheat 6-10

209 Appedix 6.8 ode of type 236: PTS with constant-outlet-temperature control SUBROUTINE TYPE236 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Linear Parabolic oncentrator Solar ollector IISiBat Model: Type236 Author: From Duffie and Beckman "Solar Engineering of Thermal Processes" Editor: Ming Qu Date: last modified: Oct 2007 The Linear Parabolic oncentrator model is based on equations taken from Duffie and Beckman's "Solar Engineering of Thermal Processes" Description: this subroutine models a linear parabolic concentrating solar collector with a variable speed pump to keep the collector outer at the user specified condition, but if the he flow reaches its maximum condition, dump collected energy to keep the outlet at the desired condition. Ming Qu modified TESS type 536 at four aspects 1. when FR_solar is equal to 0 and there is direct solar radiation, it is assumped that the PTS is not tracking, so that the outlet temperature of solar collector is equal to the ambient temperature. 2. add the end-loss into model due to the incident angle existed 3. add the shadow loss caused by the adjacent solar collector array 4. onstant temperature control added in the calculation 5. include the viscosity of propylene glycol look up table *** *** Model Parameters *** Number of collectors in series - [1;+Inf] Number of collectors in parallel - [1;+Inf] Aperture length m [0.0;+Inf] Aperture width m [0;+Inf] oncentration ratio - [0.0;+Inf] Intercept efficiency (FrTan) - [0.0;1.0] Efficiency slope (FrUl) kj/hr.m^2.k [-Inf;+Inf] Fluid specific heat kj/kg.k [0.0;+Inf] Logical unit - [10;30] Number of IAM points - [2;10] Tested flow rate kg/s [0.0;+Inf] Number of modules in an array - [1;+Inf] Number of arrays - [1;+Inf] Parabola focal length m [0;+Inf] Distance between adjacent arrays m [0;+Inf] Maximum flow ratekg/hr [0;+Inf] Diameter of the pipe m [0;+Inf] Viscosity of the fluid kg/m.s [0;+Inf] *** *** Model Inputs *** Inlet temperature [-Inf;+Inf] Inlet flow rate kg/hr [0.0;+Inf] Ambient temperature [-Inf;+Inf] Incident beam radiation kj/hr.m^2 [0.0;+Inf] Incidence angle degrees [0.0;90.0] Maximum outlet temperature [-Inf;+Inf] Tracking angle degrees [0;+Inf] 6-11

210 Pump control specification - [0;+Inf] Outlet temperature setpoint [0;+Inf] *** *** Model Outputs *** Outlet temperature [-Inf;+Inf] Outlet flow rate kg/hr [0.0;+Inf] Useful energy gain kj/hr [-Inf;+Inf] Dumped energy kj/hr [0.0;+Inf] Theoretical temperature [-Inf;+Inf] Length of the end loss m [0;+Inf] Height of the shadow from the adjacent array m [0;+Inf] Actual aperture area m^2 [0;+Inf] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE236!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS IMPLIIT NONE DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT,STORED INTEGER*4 INFO(15),NP,NI,NOUT,ND,IUNIT,ITYPE,INTRL,NSTORED HARATER*3 YHEK,OHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=19,NI=9,NOUT=8,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NOUT),PAR(NP),YHEK(NI),OHEK(NOUT), 1 STORED(NSTORED),T(ND),DTDT(ND) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE DOUBLE PREISION BEAM,X(1),Y(1),XNS,XNP,AAP_TOTAL,ON,FRTAN,FRUL, 1 Z(1),W(1) DOUBLE PREISION GTEST,TIN,FLW,TAMB,THETA,TMAX,XKAT,FTEST,FPUL DOUBLE PREISION RTEST,TWANT,FLOWMAX,FLOWMIN,DH,S,MUF,RATIO,MIDPAR DOUBLE PREISION QU,TAL,TOUT,QDUMP,LFOAL,LSPAN,HSHADOW,AAL,AAW 6-12

211 DOUBLE PREISION RDONV,XK,R2,AAP,PF,LENDLOSS,R1,R1A,PI,MIDPAR2 DOUBLE PREISION TANGLE,SANGLE,OTANGLE,IDNMIN INTEGER NX(1),LU,NPOINT,NUM_S,NUM_P,NUM_M,NUM_A,IPUMP,IOUNT,LU1, 1 NPOINT1,NW(1) HARATER*12::FAILOUNTSTR HARATER(LEN=MAXMESSAGELENGTH)::WARNMESS ONSTANTS REQUIRED BY THE MODEL DATA RDONV/ D0/, PI/ / GET GLOBAL TRNSYS SIMULATION VARIABLES TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN DO ALL THE VERY LAST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN DO ALL THE VERY FIRST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(7).EQ.-1) THEN RETRIEVE THE UNIT NUMBER AND TYPE NUMBER FOR THIS OMPONENT FROM THE INFO ARRAY IUNIT=INFO(1) ITYPE=INFO(2) SET SOME INFO ARRAY VARIABLES TO TELL THE TRNSYS ENGINE HOW THIS TYPE IS TO WORK INFO(6)=NOUT INFO(9)=1 INFO(10)=0!STORAGE FOR VERSION 16 HAS BEEN HANGED ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS OMPONENT REQUIRES TO WHAT IS SUPPLIED IN THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND) SET THE YHEK AND OHEK ARRAYS TO ONTAIN THE ORRET VARIABLE TYPES FOR THE INPUTS AND OUTPUTS DATA YHEK/'TE1','MF1','TE1','IR1','DG1','TE1','DG1','DM1','TE1'/ DATA OHEK/'TE1','MF1','PW1','PW1','TE1','LE1','LE1','AR1'/ 6-13

212 ALL THE RHEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT TYPES FOR THIS OMPONENT ALL RHEK(INFO,YHEK,OHEK) ALL LINKK TO TELL THE USER THIS TYPE REQUIRES TAU_ALPHA ALL LINKK('TYPE 536',' TAU_ALPHA',4,INFO(1)) RETURN TO THE ALLING PROGRAM RETURN DO ALL OF THE INITIAL TIMESTEP MANIPULATIONS HERE - THERE ARE NO ITERATIONS AT THE INTIAL TIME IF (TIME.LT.(TIME0+DELT/2.D0)) THEN SET THE UNIT NUMBER FOR FUTURE ALLS IUNIT=INFO(1) READ IN THE VALUES OF THE PARAMETERS IN SEQUENTIAL ORDER NUM_S=JFIX(PAR(1)+0.1) XNS=DBLE(NUM_S) NUM_P=JFIX(PAR(2)+0.1) XNP=DBLE(NUM_P) AAL=PAR(3) AAW=PAR(4) ON=PAR(5) FRTAN=PAR(6) FRUL=PAR(7) PF=PAR(8) LU=JFIX(PAR(9)+0.1) NPOINT=JFIX(PAR(10)+0.1) GTEST=PAR(11) NUM_M=JFIX(PAR(12)+0.1) NUM_A=JFIX(PAR(13)+0.1) LFOAL=PAR(14) LSPAN=PAR(15) FLOWMAX=PAR(16) DH=PAR(17) LU1=JFIX(PAR(18)+0.1) NPOINT1=JFIX(PAR(19)+0.1) HEK THE PARAMETERS FOR PROBLEMS AND RETURN FROM THE SUBROUTINE IF AN ERROR IS FOUND IF(NUM_S.LT.1) ALL TYPEK(-4,INFO,NI,1,0) IF(NUM_P.LT.1) ALL TYPEK(-4,INFO,NI,2,0) IF(AAL.LE.0.) ALL TYPEK(-4,INFO,NI,3,0) IF(AAW.LE.0.) ALL TYPEK(-4,INFO,NI,4,0) IF(ON.LE.0.) ALL TYPEK(-4,INFO,NI,5,0) IF(FRTAN.LE.0.) ALL TYPEK(-4,INFO,NI,6,0) IF(FRUL.LE.0.) ALL TYPEK(-4,INFO,NI,7,0) IF(PF.LE.0.) ALL TYPEK(-4,INFO,NI,8,0) IF(LU.LT.10) ALL TYPEK(-4,INFO,NI,9,0) IF(NPOINT.LT.2) ALL TYPEK(-4,INFO,NI,10,0) IF(GTEST.LE.0.) ALL TYPEK(-4,INFO,NI,11,0) IF(NUM_M.LT.1) ALL TYPEK(-4,INFO,NI,12,0) IF(NUM_A.LT.1) ALL TYPEK(-4,INFO,NI,13,0) IF(LFOAL.LE.0.) ALL TYPEK(-4,INFO,NI,14,0) IF(LSPAN.LE.0.) ALL TYPEK(-4,INFO,NI,15,0) 6-14

213 IF(FLOWMAX.LT.0.) ALL TYPEK(-4,INFO,NI,16,0) IF(DH.LT.0.) ALL TYPEK(-4,INFO,NI,17,0) IF(LU1.LT.10) ALL TYPEK(-4,INFO,NI,18,0) IF(NPOINT1.LT.2) ALL TYPEK(-4,INFO,NI,19,0) PERFORM ANY REQUIRED ALULATIONS TO SET THE INITIAL VALUES OF THE OUTPUTS HERE OUT(1)=XIN(1) OUT(2)=0. OUT(3)=0. OUT(4)=0. OUT(5)=0. OUT(6)=0. OUT(7)=0. OUT(8)=0. RETURN TO THE ALLING PROGRAM RETURN *** ITS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED IF(INFO(1).NE.IUNIT) THEN RESET THE UNIT NUMBER IUNIT=INFO(1) ITYPE=INFO(2) READ IN THE VALUES OF THE PARAMETERS IN SEQUENTIAL ORDER NUM_S=JFIX(PAR(1)+0.1) XNS=DBLE(NUM_S) NUM_P=JFIX(PAR(2)+0.1) XNP=DBLE(NUM_P) AAL=PAR(3) AAW=PAR(4) ON=PAR(5) FRTAN=PAR(6) FRUL=PAR(7) PF=PAR(8) LU=JFIX(PAR(9)+0.1) NPOINT=JFIX(PAR(10)+0.1) GTEST=PAR(11) NUM_M=JFIX(PAR(12)+0.1) NUM_A=JFIX(PAR(13)+0.1) LFOAL=PAR(14) LSPAN=PAR(15) FLOWMAX=PAR(16) DH=PAR(17) LU1=JFIX(PAR(18)+0.1) NPOINT1=JFIX(PAR(19)+0.1)

214 RETRIEVE THE URRENT VALUES OF THE INPUTS TO THIS MODEL FROM THE XIN ARRAY IN SEQUENTIAL ORDER TIN=XIN(1) FLW=XIN(2) TAMB=XIN(3) BEAM=XIN(4) THETA=XIN(5) TMAX=XIN(6) TANGLE=XIN(7) IPUMP=INT(XIN(8)+0.1) TWANT=XIN(9) HEK THE INPUTS FOR PROBLEMS IF(FLW.LT.0.) ALL TYPEK(-3,INFO,2,0,0) IF(BEAM.LT.0.) ALL TYPEK(-3,INFO,4,0,0) IF(TWANT.GT.TMAX) ALL TYPEK(-3,INFO,9,0,0) IF(ERRORFOUND()) RETURN PERFORM ALL THE ALULATION HERE FOR THIS MODEL. STORE THE INLET FLOW RATE AS AN OUTPUT TO DO MASS FLOW HEKING DETERMINE THE VISOSITY OF PROPYLENE GLYOL REGARDING TO THE INLET TEMPERATURE OF FLUID W(1)=TIN NW(1) = NPOINT1 ALL DYNAMIDATA(LU1,1,NW,1,W,Z,INFO,*5) ALL LINKK('TYPE 236','DYNAMIDATA',1,99) 5 IF(ERRORFOUND()) RETURN 1 MUF=Z(1)/1000 S=PI*DH*DH/4 FLOWMIN=4000.*S*MUF/DH*3600. IDNMIN=3.6*300.0 FIND THE LENGTH OF REEIVER TUBE WITH NO SUN RAY DUE TO ENDLOSS SANGLE=DASIN(AAW/LSPAN) OTANGLE=(90.-TANGLE)*RDONV IF(SANGLE.GT.OTANGLE) THEN HSHADOW=AAW-LSPAN*DSIN(OTANGLE) ELSE HSHADOW= 0. LENDLOSS=LFOAL*DTAN(THETA*RDONV) IF (LENDLOSS.GT.NUM_M*AAL)THEN AAP_TOTAL = 0 ELSE AAP_TOTAL=NUM_A*(NUM_M*AAL-LENDLOSS)*(AAW-HSHADOW) 6-16

215 AAP=AAL*AAW DETERMINE INIDENE ANGLE MODIFIER FROM ALL TO DATA IF((BEAM.GT.0.).AND.(THETA.LE.90.)) THEN X(1)=THETA NX(1)=NPOINT ALL DYNAMIDATA(LU,1,NX,1,X,Y,INFO,*10) ALL LINKK('TYPE 236','DYNAMIDATA',1,99) 10 IF(ERRORFOUND()) RETURN 1 XKAT=Y(1) IF(XKAT.LT.0.) XKAT=0. ELSE XKAT=0. ********************START RIGHT MODE***************** SKIP TO THE RIGHT MODE IF((IPUMP.GT. 0).AND.(TWANT.GT.TIN)) GO TO 120 ONSTANT FLOW MODEL OF A SOLAR OLLETOR HEK TO SEE IF THERE IS FLOW FLW=XIN(2) IF(FLW.GT.0.) THEN ALULATE F'UL FTEST=FRUL/GTEST/PF/ON IF(FTEST.GE.1.) THEN FPUL=FRUL ELSE FPUL=-GTEST*PF*DLOG(1.-FRUL/GTEST/PF/ON) DETERMINE MODIFIERS FOR OFF-TEST FLOW RATE AND OLLETORS IN SERIES RTEST=GTEST*PF*(1.-DEXP(-FPUL/GTEST/PF)) R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP)/PF)) 1 /RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK ALULATE USEFUL ENERGY GAIN AND OLLETOR TEMP QU=R1*R2*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) TAL=QU/FLW/PF+TIN TOUT=DMIN1(TAL,TMAX) QDUMP=FLW*PF*(TAL-TOUT) ELSE TAL=TAMB TOUT=DMIN1(TAL,TMAX) QU=0. QDUMP=0. FLW= ONTINUE WHEN SOLAR BEAM IS GREATER THAN THE MINIMUM INTENSITY, THE PTS STARTS TO TRAK 6-17

216 IF(BEAM.GE. IDNMIN) THEN GUESS A RATIO FOR THE START OF THE ITERATIONS RATIO=1.0 IOUNT=1 ALULATE THE USEFUL ENERGY GAIN JUST TO SEE IF ITS POSITIVE QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) MIDPAR = FRTAN*XKAT*BEAM MIDPAR2 = FRUL/ON*(TIN-TAMB) IF(BEAM.LE.0.) QU=0. SET AN INITIAL FLOW RATE FOR THE ALULATION FLW=FLOWMAX IF THERE IS USEFUL ENERGY GAIN, ALULATE THE MAXIMUM FLOW, IF NOT, TURN THE PUMPS OFF AND GET OUT MIDPAR = QU IF(QU.GT.0) THEN 122 IF(GTEST.GT.0.) THEN FTEST=FRUL/GTEST/PF/ON IF(FTEST.GE.1.) THEN FPUL=FRUL ELSE FPUL=-GTEST*PF*DLOG(1.-FRUL/GTEST/PF/ON) DETERMINE MODIFIERS FOR OFF-TEST FLOW RATE AND OLLETORS IN SERIES RTEST=GTEST*PF*(1.-DEXP(-FPUL/GTEST/PF)) R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP)/PF)) 1 /RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK RATIO=R1*R2 ELSE RATIO=1.0 QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) FLW=QU/PF/(TWANT-TIN) TOUT=TWANT HEK TO SEE IF THE FLOW RATE HAS ONVERGED IF(GTEST.GT.0.) THEN R1A=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP) 1 /PF))/RTEST IF((DABS(R1A-R1).GT.0.01).AND.(IOUNT.LT.100)) THEN IOUNT=IOUNT+1 GOTO 122 MIDPAR = FLW 6-18

217 IF THE ALULATED FLOW IS GREATER THAN THE MAXIMUM FLOW, RUN AT THE MAXIMUM FLOW IF(FLW.GT.FLOWMAX) THEN FLW=FLOWMAX IF(GTEST.GT.0.) THEN R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP) 1 /PF))/RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK RATIO=R1*R2 ELSE RATIO=1. QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) TAL=QU/FLW/PF+TIN IF(IPUMP.GT.1) THEN TOUT=DMIN1(TAL,TWANT) QDUMP=FLW*PF*(TAL-TOUT) ELSE QDUMP=0. MIDPAR2 = TAL ELSE IF (FLW.LT.FLOWMIN) THEN FLW=FLOWMIN IF(GTEST.GT.0.) THEN R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP) 1 /PF))/RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK RATIO=R1*R2 ELSE RATIO=1. QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) TAL=QU/FLW/PF+TIN IF(IPUMP.GT.1) THEN TOUT=DMIN1(TAL,TWANT) QDUMP=FLW*PF*(TAL-TOUT) ELSE QDUMP=0. MIDPAR2 = TAL ELSE QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) FLW=QU/PF/(TWANT-TIN) TAL=QU/FLW/PF+TIN IF(IPUMP.GT.1) THEN TOUT=DMIN1(TAL,TWANT) QDUMP=FLW*PF*(TAL-TOUT) ELSE QDUMP=

218 MIDPAR2 = TAL ELSE FLW=FLOWMIN IF(GTEST.GT.0.) THEN R1=XNS*FLW/XNP*PF/AAP*(1.-DEXP(-FPUL*AAP/XNS/(FLW/XNP) 1 /PF))/RTEST XK=R1*AAP*FRUL/(FLW/XNP)/PF/XNS/ON R2=(1.-(1.-XK)**NUM_S)/XNS/XK RATIO=R1*R2 ELSE RATIO=1. QU=RATIO*AAP_TOTAL*(FRTAN*XKAT*BEAM-FRUL/ON*(TIN-TAMB)) TOUT=QU/FLW/PF+TIN FLW=0. QU=0. QDUMP=0. TAL=TAMB+FRTAN*XKAT*BEAM*ON/FRUL TOUT=DMIN1(TAL,TMAX) ELSE FLW=0. QU=0. QDUMP=0. TOUT=TAMB SET THE OUTPUTS FROM THIS MODEL IN SEQUENTIAL ORDER AND GET OUT OUT(1)=TOUT OUT(2)=FLW OUT(3)=QU OUT(4)=QDUMP OUT(1)=S OUT(2)=MUF OUT(3)=DH OUT(4)=FLOWMIN OUT(5)=TAL OUT(6)=LENDLOSS OUT(7)=HSHADOW OUT(8)=AAP_TOTAL OUT(5)=Z(1) OUT(6)=MIDPAR OUT(7)=MIDPAR2 OUT(8)=AAP_TOTAL EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON 6-20

219 RETURN 1 END Appedix 6.9 ode of type 237: main control of solar cooling for constant-outlettemperature control SUBROUTINE TYPE237 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE237 Author: Ming Qu Editor: Date: TRNSYS 7.5 last modified: oct 2007 NOTE: This controller can only be used with Solver 0 (Successive substitution) This controller is main integrated controller for IW solar cooling system with constant outlet temperature control. It is used to control the outlet temperature of the solar field by varying the flow rate. It also controls the switch between by-pass and chiller and the switch between hot-water chiller and direct fired chiller. Description: there is no solar pump control signal from this component, since the control of solar pump is determined in the PTS component. *** *** Model Parameters *** No. of oscillations - [1;+Inf] High limit cut-out of circulating fluid temperature [-Inf;+Inf] *** *** Model Inputs *** Flowrate of solar loop kg/hr [0;+Inf] ooling load kj/hr [-Inf;+Inf] HTF temperature at the inlet of the chiller [-Inf;+Inf] Upper dead band of inlet temperature [-Inf;+Inf] Lower dead band of inlet temperature [-Inf;+Inf] Input control function of inlet temperature - [0;1] hilled water supply temperature of HW chiller in the last timestep [-Inf;+Inf] chilled water setpoint of chiller [-Inf;+Inf] *** *** Model Outputs *** Output control signal of HT ctrl - [0.0;1.0] ounter of stick in temperature control - [0.0;1.0] Osillation number of the temperature control - [0.0;1.0] Lastcall output of temperature control - [0.0;1.0] Output control of bypass - [0.0;1.0] Output control of chiller - [0.0;1.0] Output control of fired chiller - [0.0;1.0] 6-21

220 Output control of cooling load - [0.0;1.0] Ratio of HTF over bypass - [-Inf;+Inf] Ration of HW - [-Inf;+Inf] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE237!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=2,NI=8,NO=10,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE LOAL VARIABLE DELARATIONS HARATER*12 INTStr HARATER*160 WarnMsg INTEGER NSTK,!No of_oscillations 1 INT_1, 1 NTOLD_1, 1 IOS_1!Osillation number DOUBLE PREISION 1 TMAX,!max temperature at inlet of the chiller 1 LAST_1, 1 LOAD,!cooling load 1 T_chillerin,!HTF_temperature_at_the_inlet_of_the_chiller 1 UP_T,!Upper_dead_band_of_inlet_temperature 1 LOW_T,!Lower_dead_band_of_inlet_temperature 1 LOAD_TRL,!LOAD ONTROL SIGNAL 1 HILLER_TRL,!HW HILLER ONTROL 1 T_HL_PREV_O,!OUTLET TEMPERATURE OF HW FROM THE HILLER 1 IF_1,!ONTROL FUTION INPUTS 6-22

221 1 T_HW_SETPOINT,!HILLED WATER SETPOINT 1 R_BYPASS,!RATIO OF HTF OVER THE BYPASS 1 R_LDIVERTER,!RATIO OF HW OVER THE HW HILLER 1 FR_sr,!FLOW RATE OF THE SOLAR OLLETION LOOP 1 SOLAR_TRL!SOLAR OLLEITON LOOP IS RUNING DATA STATEMENTS DATA YHEK/'MF1','PW1','TE1','TE1','TE1','F1','TE1','TE1'/ DATA OHEK/'F1','DM1','DM1','F1','F1','F1','F1','F1', 1 'F1','F1'/ TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN PERFORM LAST ALL MANIPULATIONS IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN 1 INT_1=JFIX(OUT(2)*DELT/(TFINAL-TIME0)*100.d0) IF(INT_1.GE.10) THEN WRITE (INTStr,*) INT_1 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) RETURN FROM THIS MODEL AS NO "AFTER-ONVERGENE" MANIPULATIONS ARE REQUIRED IF(INFO(13).GT.0) RETURN PERFORM FIRST ALL MANIPULATIONS IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required!set THE PROPER NUMBER OF PARAMETERS AORDING TO THE MODE NSTK=JFIX(PAR(1)+0.01) 6-23

222 IF(NSTK.LT.0) ALL TYPEK(4,INFO,0,1,0)!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN!set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values NSTK = JFIX(PAR(1)+0.01) TMAX= PAR(2) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) ITYPE=INFO(2)!reread the parameter values NSTK=JFIX(PAR(1)+0.01) TMAX= PAR(2) *** PERFORM ALL THE ALULATION HERE FOR THIS MODEL. *** PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of outlet temperature of solar collector control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of outlet temperature of solar collector control LEAR STIK OUNTER ON FIRST ALL OF T-S IF(INFO(7).EQ.0) THEN OUT(3)=0.d0 6-24

223 FR_sr = XIN(1) LOAD = XIN(2) T_chillerin = XIN(3) UP_T = XIN(4) LOW_T = XIN(5) IF_1= XIN(6) T_HL_PREV_O = XIN(7) T_HW_SETPOINT = XIN(8) SOLAR LOOP ONTROL IF (FR_sr.GT.0.) THEN SOLAR_TRL = 1. ELSE SOLAR_TRL = 0. LOADS ONTROL IF (LOAD.LE.0.) THEN LOAD_TRL = 0. ELSEIF (LOAD.GT.0) THEN LOAD_TRL = 1. ******************************** ONTROL HT at inlet of chiller ******************************** TEMPERATURE ONTROL NTOLD_1=OUT(4)! control signal for the last time step LAST_1=OUT(1) IOS_1=JFIX(OUT(3)+0.01) IF(IOS_1.EQ.NSTK) OUT(2)=OUT(2)+1.d0 IF(IOS_1.GE. NSTK) GO TO 15 IF(INFO(7).EQ.0) THEN IF (XIN(6).GT. 0.5) THEN NTOLD_1 = 1 OUT(4)=DBLE(NTOLD_1) ELSE NTOLD_1 = 0 OUT(4)=DBLE(NTOLD_1) IF (NTOLD_1.GT.0.5) GOTO

224 !OUTPUT WAS 0 LAST ALL OUT(1)=0.d0 IF (T_chillerin.GT.UP_T) OUT(1)=1.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 GO TO 15!OUTPUT WAS 1 LAST ALL 10 OUT(1)=1.d0 IF (T_chillerin.LT.LOW_T) OUT(1)=0.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 15 IF((ABS(LAST_1-OUT(1)).LT.1.E-06).AND.(IOS_1.NE.NSTK)) THEN OUT(3)=OUT(3) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(3)=OUT(3)+1.d0 IF THE OUTLET TEMPERATURE OF HILLER IS HIGHER THATN 14 IF(T_HL_PREV_O.LT.13.1) THEN HILLER_TRL = OUT(1)*LOAD_TRL*SOLAR_TRL ELSE HILLER_TRL = 0.0 BYPASS ONTROL BASED ON THE HILLED WATER OUTLET TEMPERATURE IF(HILLER_TRL.GE. 0.5) THEN IF(T_HL_PREV_O.GE.(T_HW_SETPOINT+4.0)) THEN R_BYPASS=0.0 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+3.0)) THEN R_BYPASS=0.2 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+2.0)) THEN R_BYPASS=0.4 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT-1.0)) THEN R_BYPASS=0.65 ELSE R_BYPASS=1.0 ELSE R_BYPASS=1.0 OUTPUT DELARATION OUT(5)= SOLAR_TRL*(1-HILLER_TRL) OUT(6)= HILLER_TRL OUT(7)= LOAD_TRL - HILLER_TRL OUT(8)= LOAD_TRL OUT(9)= R_BYPASS OUT(10)= HILLER_TRL

225 SET THE STORAGE ARRAY AT THE END OF THIS ITERATION IF NEESSARY NITEMS= STORED(1)= ALL SET_STORAGE_VARS(STORED,NITEMS,INFO) REPORT ANY PROBLEMS THAT HAVE BEEN FOUND USING ALLS LIKE THIS: ALL MESSAGES(-1,'put your message here','message',iunit,itype) ALL MESSAGES(-1,'put your message here','warning',iunit,itype) ALL MESSAGES(-1,'put your message here','severe',iunit,itype) ALL MESSAGES(-1,'put your message here','fatal',iunit,itype) EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END Appedix 6.10 ode of type 243: main control of solar cooling with storage tank for shifting energy by constant-outlet-temperature SUBROUTINE TYPE243 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE243 Author: Ming Qu Editor: Date: TRNSYS 7.5 last modified: Oct 2007 NOTE: This controller can only be used with Solver 0 (Successive substitution) *** *** Model Parameters *** No. of oscillations - [1;+Inf] High limit cut-out of circulating fluid temperature [-Inf;+Inf] the hot water rated flow of absorption chiller kg/hr [0;+Inf] the target temeprature of solar collection loop [0;+Inf] *** *** Model Inputs *** Flowrate of solar loop kg/hr [0;+Inf] ooling load kj/hr [-Inf;+Inf] HTF temperature at the inlet of the chiller [-Inf;+Inf] Upper dead band of inlet temperature [-Inf;+Inf] 6-27

226 Lower dead band of inlet temperature [-Inf;+Inf] Input control function ofht - [0;1] hilled water supply temperature of HW chiller in the last timestep [-Inf;+Inf] chilled water setpoint of chiller [-Inf;+Inf] the temperature at the outlet2 of the storage tank [0;+Inf] *** *** Model Outputs *** Output control signal ofht ctrl - [0.0;1.0] ounter of stick in HT control - [0.0;1.0] Osillation number of the HT control - [0.0;1.0] Lastcall output of HT control - [0.0;1.0] Output control of hw chiller - [0.0;1.0] Output control of fired chiller - [0.0;1.0] Output control of cooling load - [0.0;1.0] RA2 - [0;1.0] RB2 - [0;1.0] R2 - [0;1.0] Flow rate of the HTF discharged from the storage tank kg/hr [0;+Inf] Output control of charge the storage tank - [0;1.0] Output control of discharge the storage tank - [0;1.0] output for debug - [-Inf;+Inf] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE243!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=4,NI=9,NO=14,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE 6-28

227 LOAL VARIABLE DELARATIONS HARATER*12 INTStr HARATER*160 WarnMsg INTEGER NSTK,!No of_oscillations 1 INT_1, 1 NTOLD_1, 1 IOS_1!Osillation number DOUBLE PREISION 1 TMAX,!max temperature at inlet of the chiller 1 LAST_1, 1 LOAD,!cooling load 1 T_chillerin,!HTF_temperature_at_the_inlet_of_the_chiller 1 UP_T,!Upper_dead_band_of_inlet_temperature 1 LOW_T,!Lower_dead_band_of_inlet_temperature 1 LOAD_TRL,!LOAD ONTROL SIGNAL 1 HILLER_TRL,!HW HILLER ONTROL 1 T_HL_PREV_O,!OUTLET TEMPERATURE OF HW FROM THE HILLER 1 IF_1,!ONTROL FUTION INPUTS 1 T_HW_SETPOINT,!HILLED WATER SETPOINT 1 R_BYPASS,!RATIO OF HTF OVER THE BYPASS 1 R_LDIVERTER,!RATIO OF HW OVER THE HW HILLER 1 FR_SR,!FLOW RATE OF THE SOLAR OLLETION LOOP 1 SOLAR_TRL,!SOLAR OLLEITON LOOP IS RUNING 1 FR_R_HILLER,!THE RATED FLOW RATE OF THE HOT WATER REQUIRED BY HILLER 1 T_SR_TARGET,!THE TARGET TEMPERATURE OF THE SOLAR OLLETOR 1 T_S3,!THE TEMPERATURE AT OUTLET 2 OF THE STORAGE TANK 1 RA2,!FLOW RATIO OF STREAM2 OF THE BYPASS DIVERTER 1 RB2,!FLOW RATIO OF STREAM2 OF THE STORAGE DIVERTER 1 R2,!FLOW RATIO OF STREAM2 OF THE HILLER DIVERTER 1 FR_DISHARGE,!FLOW RATE OF THE HTF FROM THE STORAGE TANK 1 FR_HILLER,!FLOW RATE OF HTF TROUGH THE HILLER 1 HARGE_TRL,!HARGING TRL 1 DISHARGE_TRL!DISHARGING TRL DATA STATEMENTS DATA YHEK/'MF1','PW1','TE1','TE1','TE1','F1','TE1','TE1', 1 'TE1'/ DATA OHEK/'F1','DM1','DM1','F1','F1','F1','F1','F1', 1 'F1','F1','MF1','F1','F1','DM1'/ TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN PERFORM LAST ALL MANIPULATIONS 6-29

228 IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN 1 INT_1=JFIX(OUT(2)*DELT/(TFINAL-TIME0)*100.d0) IF(INT_1.GE.10) THEN WRITE (INTStr,*) INT_1 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) RETURN FROM THIS MODEL AS NO "AFTER-ONVERGENE" MANIPULATIONS ARE REQUIRED IF(INFO(13).GT.0) RETURN PERFORM FIRST ALL MANIPULATIONS IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required!set THE PROPER NUMBER OF PARAMETERS AORDING TO THE MODE NSTK=JFIX(PAR(1)+0.01) IF(NSTK.LT.0) ALL TYPEK(4,INFO,0,1,0)!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN!set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values NSTK = JFIX(PAR(1)+0.01) TMAX= PAR(2) FR_R_HILLER = PAR(3) T_SR_TARGET = PAR(4) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN

229 THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) ITYPE=INFO(2)!reread the parameter values NSTK=JFIX(PAR(1)+0.01) TMAX= PAR(2) FR_R_HILLER = PAR(3) T_SR_TARGET = PAR(4) *** PERFORM ALL THE ALULATION HERE FOR THIS MODEL. *** PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of outlet temperature of solar collector control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of outlet temperature of solar collector control LEAR STIK OUNTER ON FIRST ALL OF T-S IF(INFO(7).EQ.0) THEN OUT(3)=0.d0 FR_SR = XIN(1) LOAD = XIN(2) T_chillerin = XIN(3) UP_T = XIN(4) LOW_T = XIN(5) IF_1= XIN(6) T_HL_PREV_O = XIN(7) T_HW_SETPOINT = XIN(8) T_S3=XIN(9) SOLAR LOOP ONTROL IF (FR_SR.GT.0.) THEN SOLAR_TRL = 1. ELSE SOLAR_TRL = 0. LOADS ONTROL 6-31

230 IF (LOAD.LE.0.) THEN LOAD_TRL = 0. ELSEIF (LOAD.GT.0) THEN LOAD_TRL = 1. ONTROL OF TOUT OF ST & SR NTOLD_1=OUT(4)! control signal for the last time step LAST_1=OUT(1) IOS_1=JFIX(OUT(3)+0.01) IF(IOS_1.EQ.NSTK) OUT(2)=OUT(2)+1.d0 IF(IOS_1.GE. NSTK) GO TO 15 IF(INFO(7).EQ.0) THEN IF (XIN(6).GT. 0.5) THEN NTOLD_1 = 1 OUT(4)=DBLE(NTOLD_1) ELSE NTOLD_1 = 0 OUT(4)=DBLE(NTOLD_1) IF (NTOLD_1.GT.0.5) GOTO 10!OUTPUT WAS 0 LAST ALL OUT(1)=0.d0 IF((T_chillerin.GT.UP_T).OR. (T_S3.GT.UP_T)) OUT(1)=1.d0 GO TO 15!OUTPUT WAS 1 LAST ALL 10 OUT(1)=1.d0 IF ((T_chillerin.LT.LOW_T).AND. (T_S3.LT.LOW_T)) THEN OUT(1)=0.d0 15 IF((ABS(LAST_1-OUT(1)).LT.1.E-06).AND.(IOS_1.NE.NSTK)) THEN OUT(3)=OUT(3) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(3)=OUT(3)+1.d0 HW HILLER ONTROL IF((LOAD_TRL.GT.0.5).AND.(T_HL_PREV_O.LT.13.1)) THEN IF (OUT(1).GT. 0.5) THEN 6-32

231 ELSE HILLER_TRL =1.0 HILLER_TRL =0.0 ELSE HILLER_TRL =0.0 BYPASS ONTROL BASED ON THE HILLED WATER OUTLET TEMPERATURE IF(HILLER_TRL.GE. 0.5) THEN IF(T_HL_PREV_O.GE.(T_HW_SETPOINT+4.0)) THEN R_BYPASS=0.0 FR_HILLER=FR_R_HILLER ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+3.0)) THEN R_BYPASS=0.2 FR_HILLER=FR_R_HILLER*0.8 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+2.0)) THEN R_BYPASS=0.4 FR_HILLER=FR_R_HILLER*0.6 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT-1.0)) THEN R_BYPASS=0.65 FR_HILLER=FR_R_HILLER*0.35 ELSE R_BYPASS=1.0 FR_HILLER=0. ELSE R_BYPASS=1.0 FR_HILLER=0. DIVERTER ONTROLS FR_DISHARGE=0. HARGE_TRL=0.0 DISHARGE_TRL=0.0 RA2=0. RB2=0. R2=0. IF(HILLER_TRL.GT. 0.5) THEN IF(FR_SR.GE.FR_HILLER) THEN RA2=1. RB2=FR_HILLER/FR_SR R2=1. HARGE_TRL=1.0 ELSE IF(T_chillerin.GE. LOW_T)THEN IF(T_S3.GE. LOW_T)THEN RA2=1. RB2=1. FR_DISHARGE=FR_HILLER-FR_SR IF(FR_DISHARGE.GT. roh_htf*v_st/delt) 6-33

232 1 FR_DISHARGE=roh_HTF*V_ST/DELT R2=FR_SR/FR_HILLER DISHARGE_TRL=FR_DISHARGE/FR_R_HILLER ELSE RA2=1. RB2=1. R2=1. ELSE IF(T_S3.GE. LOW_T)THEN R2=0.0 FR_DISHARGE=FR_HILLER IF(FR_DISHARGE.GT. roh_htf*v_st/delt) 1 FR_DISHARGE=roh_HTF*V_ST/DELT DISHARGE_TRL=FR_DISHARGE/FR_R_HILLER OUTPUT DELARATION OUT(5)= HILLER_TRL OUT(6)= LOAD_TRL*(1-HILLER_TRL) OUT(7)= LOAD_TRL OUT(8)= RA2 OUT(9)= RB2 OUT(10)= R2 OUT(11)= FR_DISHARGE OUT(12)= HARGE_TRL OUT(13)= DISHARGE_TRL OUT(14)= FR_HILLER EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END Appedix 6.11 ode of type 242: main control of solar cooling with storage tank control for preheat controlled by constant-outlet-temperature SUBROUTINE TYPE242 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE242 Author: Ming Qu Editor: Date: TRNSYS 7.5 last modified: Oct

233 NOTE: This controller can only be used with Solver 0 (Successive substitution) *** *** Model Parameters *** No. of oscillations - [1;+Inf] High limit cut-out of circulating fluid temperature [-Inf;+Inf] the hot water rated flow of absorption chiller kg/hr [0;+Inf] the target temeprature of solar collection loop [0;+Inf] the density of the HTF kg/m^3 [0;+Inf] the volume of the storage tank m^3 [0;+Inf] *** *** Model Inputs *** Flowrate of solar loop kg/hr [0;+Inf] ooling load kj/hr [-Inf;+Inf] HTF temperature at the inlet of the chiller [-Inf;+Inf] Upper dead band of inlet temperature [-Inf;+Inf] Lower dead band of inlet temperature [-Inf;+Inf] Input control function of HW chiller - [0;1] hilled water supply temperature of HW chiller in the last timestep [-Inf;+Inf] chilled water setpoint of chiller [-Inf;+Inf] the temperature at the outlet2 of the storage tank [0;+Inf] the inlet temperature of solar field [-Inf;+Inf] the temperature at the bottom of the storage tank [-Inf;+Inf] the average temperature of the heat capacity tank [-Inf;+Inf] *** *** Model Outputs *** Output control signal of HWchiller ctrl - [0.0;1.0] ounter of stick in HWchiller control - [0.0;1.0] Osillation number of the HWchiller control - [0.0;1.0] Lastcall output of HWchiller control - [0.0;1.0] Output control of bypass - [0.0;1.0] Output control of fired chiller - [0.0;1.0] Output control of cooling load - [0.0;1.0] RA2 - [0.0;1.0] RB2 - [0.0;1.0] output for solar control - [0.0;1.0] output of late charging - [0.0;1.0] output of preheating - [0.0;1.0] RE2 - [0.0;1.0] RF2 - [0.0;1.0] temperature at the bottom of the storage tank [-Inf;+Inf] average temperature of heat capacity tank [-Inf;+Inf] Output control of HW chiller - [0.0;1.0] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) 6-35

234 ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE242!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=6,NI=12,NO=19,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE LOAL VARIABLE DELARATIONS HARATER*12 INTStr HARATER*160 WarnMsg INTEGER NSTK,!No of_oscillations 1 INT_1, 1 NTOLD_1, 1 IOS_1!Osillation number DOUBLE PREISION 1 TMAX,!max temperature at inlet of the chiller 1 LAST_1, 1 LOAD,!cooling load 1 T_chillerin,!HTF_temperature_at_the_inlet_of_the_chiller 1 UP_T,!Upper_dead_band_of_inlet_temperature 1 LOW_T,!Lower_dead_band_of_inlet_temperature 1 LOAD_TRL,!LOAD ONTROL SIGNAL 1 HILLER_TRL,!HW HILLER ONTROL 1 T_HL_PREV_O,!OUTLET TEMPERATURE OF HW FROM THE HILLER 1 IF_1,!ONTROL FUTION INPUTS 1 T_HW_SETPOINT,!HILLED WATER SETPOINT 1 R_BYPASS,!RATIO OF HTF OVER THE BYPASS 1 FR_SR,!FLOW RATE OF THE SOLAR OLLETION LOOP 1 SOLAR_TRL,!SOLAR OLLEITON LOOP IS RUNING 1 FR_R_HILLER,!THE RATED FLOW RATE OF THE HOT WATER REQUIRED BY HILLER 1 T_SR_TARGET,!THE TARGET TEMPERATURE OF THE SOLAR OLLETOR 1 T_S3,!THE TEMPERATURE AT OUTLET 2 OF THE STORAGE TANK 1 RA2,!FLOW RATIO OF STREAM2 OF THE BYPASS DIVERTER 6-36

235 AT THE AT 1 RB2,!FLOW RATIO OF STREAM2 OF THE STORAGE DIVERTER 1 RE2,!FLOW RATIO OF STREAM2 OF THE PREHEAT DIVERTER 1 RF2,!FLOW RATIO OF STREAM2 OF THE LTHARGE DIVERTER 1 T_S4,!THE TEMPERATURE AT INLET 2 OF THE STORAGE TANK 1 T_SR_INLET,!THE INLET TEMPERATURE OF THE SOLAR FIELD 1 LTHARGE_TRL,!THE ONTROL OF LATE AFTERNOON HARGE 1 PREHEAT_TRL,!THE ONTROL OF DISHARING THE STORAGE FOR PERHEAT 1 SOLAR_TRL_LST,!THE ONTROL OF LAST SOLAR ONTROL 1 PREHEAT_TRL_LST,!THE ONTROL OF LAST PREHEATDISHARGE 1 LTHARGE_TRL_LST,! THE ONTROL OF LAST LATEHARGE 1 T_H_AVE,! the average temperature of the heat capacity tank 1 T_S3_LST,! TEMPERATURE AT THE TOP OF THE STORAGE TANK AT LAST TIME STEP 1 T_H_AVE_LST,!AVERAGE TEMEPRATURE OF THE HEAT APATIY TANK LAST TIME STEP 1 roh_htf,!density OF THE HTF 1 V_ST,!THE VOLUME OF THE STORAGE TANK 1 M_H_HTF,!THE MASS IN THE STORAGE TANK 1 M_H_HTF_LST,!THE MASS IN THE STORAGE TANK AT THE LAST TIME 1 T_S4_LST! TEMPERATURE AT THE BOTTOM OF THE STORAGE TANK DATA STATEMENTS DATA YHEK/'MF1','PW1','TE1','TE1','TE1','F1','TE1','TE1', 1 'TE1','TE1','TE1','TE1'/ DATA OHEK/'F1','DM1','DM1','F1','F1','F1','F1','F1','F1' 1,'F1','F1','F1','F1','F1','TE1','TE1','F1','MA1','TE1'/ LAST TIME STEP TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN PERFORM LAST ALL MANIPULATIONS IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN 1 INT_1=JFIX(OUT(2)*DELT/(TFINAL-TIME0)*100.d0) IF(INT_1.GE.10) THEN WRITE (INTStr,*) INT_1 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) 6-37

236 RETURN FROM THIS MODEL AS NO "AFTER-ONVERGENE" MANIPULATIONS ARE REQUIRED IF(INFO(13).GT.0) RETURN PERFORM FIRST ALL MANIPULATIONS IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required!set THE PROPER NUMBER OF PARAMETERS AORDING TO THE MODE NSTK=JFIX(PAR(1)+0.01) IF(NSTK.LT.0) ALL TYPEK(4,INFO,0,1,0)!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN!set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values NSTK = JFIX(PAR(1)+0.01) TMAX= PAR(2) FR_R_HILLER = PAR(3) T_SR_TARGET = PAR(4) roh_htf = PAR(5) V_ST = PAR(6) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) 6-38

237 ITYPE=INFO(2)!reread the parameter values NSTK=JFIX(PAR(1)+0.01) TMAX= PAR(2) FR_R_HILLER = PAR(3) T_SR_TARGET = PAR(4) roh_htf = PAR(5) V_ST = PAR(6) *** PERFORM ALL THE ALULATION HERE FOR THIS MODEL. *** PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of outlet temperature of solar collector control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of outlet temperature of solar collector control LEAR STIK OUNTER ON FIRST ALL OF T-S RETRIEVE THE STORED VALUE FROM THE OUT ARRAY IF(INFO(7).EQ.0) THEN OUT(3) = 0.d0 SOLAR_TRL_LST=OUT(10) PREHEAT_TRL_LST = OUT(12) LTHARGE_TRL_LST = OUT(11) T_S3_LST = OUT(15) M_H_HTF_LST = OUT(18) T_H_AVE_LST = OUT(16) T_S4_LST = OUT(19) FR_SR = XIN(1) LOAD = XIN(2) T_chillerin = XIN(3) UP_T = XIN(4) LOW_T = XIN(5) IF_1= XIN(6) T_HL_PREV_O = XIN(7) T_HW_SETPOINT = XIN(8) T_S3=XIN(9) T_SR_INLET = XIN(10) T_S4 = XIN(11) T_H_AVE = XIN(12) SOLAR LOOP ONTROL IF (FR_SR.GT.0.) THEN SOLAR_TRL =

238 ELSE SOLAR_TRL = 0. LOADS ONTROL IF (LOAD.LE.0.) THEN LOAD_TRL = 0. ELSEIF (LOAD.GT.0) THEN LOAD_TRL = 1. HW HILLER ONTROL NTOLD_1=OUT(4)! control signal for the last time step LAST_1=OUT(1) IOS_1=JFIX(OUT(3)+0.01) IF(IOS_1.EQ.NSTK) OUT(2)=OUT(2)+1.d0 IF(IOS_1.GE. NSTK) GO TO 15 IF(INFO(7).EQ.0) THEN IF (XIN(6).GT. 0.5) THEN NTOLD_1 = 1 OUT(4)=DBLE(NTOLD_1) ELSE NTOLD_1 = 0 OUT(4)=DBLE(NTOLD_1) IF (NTOLD_1.GT.0.5) GOTO 10!OUTPUT WAS 0 LAST ALL OUT(1)=0.d0 IF (T_chillerin.GT.UP_T) OUT(1)=1.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 GO TO 15!OUTPUT WAS 1 LAST ALL 10 OUT(1)=1.d0 IF (T_chillerin.LT.LOW_T) OUT(1)=0.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 15 IF((ABS(LAST_1-OUT(1)).LT.1.E-06).AND.(IOS_1.NE.NSTK)) THEN OUT(3)=OUT(3) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(3)=OUT(3)+1.d0 IF THE OUTLET TEMPERATURE OF HILLER IS HIGHER THATN 14 IF(T_HL_PREV_O.LT.13.1) THEN 6-40

239 ELSE HILLER_TRL = OUT(1)*LOAD_TRL*SOLAR_TRL HILLER_TRL = 0.0 BYPASS ONTROL BASED ON THE HILLED WATER OUTLET TEMPERATURE IF(HILLER_TRL.GE. 0.5) THEN IF(T_HL_PREV_O.GE.(T_HW_SETPOINT+4.0)) THEN R_BYPASS=0.0 FR_HILLER=FR_R_HILLER ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+3.0)) THEN R_BYPASS=0.2 FR_HILLER=FR_R_HILLER*0.8 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+2.0)) THEN R_BYPASS=0.4 FR_HILLER=FR_R_HILLER*0.6 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT-1.0)) THEN R_BYPASS=0.65 FR_HILLER=FR_R_HILLER*0.35 ELSE R_BYPASS=1.0 FR_HILLER=0. ELSE R_BYPASS=1.0 FR_HILLER=0. INITAL THE VALUE OF THE ONTROLS LTHARGE_TRL=0.0 PREHEAT_TRL=0.0 RA2=0. RB2=0. RE2=0. RF2=0. DIVERTER ONTROLS FR_DISHARGE=0. HARGE_TRL=0.0 DISHARGE_TRL=0.0 RA2=0. RB2=0. IF(LOAD_TRL.GT.0.5) THEN IF(HILLER_TRL.GT. 0.5) THEN IF(FR_SR.GE.FR_HILLER) THEN RA2=FR_HILLER/FR_SR RB2=1. RE2=1. RF2=0. ELSE RA2=

240 RB2=1. RE2=1. RF2=0. LATE HARGE IF((SOLAR_TRL_LST.GT. 0.5).AND. (SOLAR_TRL.LT. 0.5)) THEN IF (T_H_AVE_LST-3.GT. T_S3_LST) THEN RA2=1. RB2=0. RE2=0. RF2=1. LTHARGE_TRL=1.0 PREHEAT DISHARGE IF((SOLAR_TRL_LST.LT. 0.5).AND. (SOLAR_TRL.GT. 0.5)) THEN IF (T_H_AVE_LST.LT. T_S4_LST-3) THEN RA2=1. RB2=1. RE2=0. RF2=0. PREHEAT_TRL=1.0 M_H_HTF = roh_htf*v_st -DELT*FR_SR ELSE M_H_HTF = 0 IF((PREHEAT_TRL_LST.GT. 0.5).AND.(SOLAR_TRL.GT. 0.5)) THEN IF ((DELT*FR_SR).LT. M_H_HTF_LST) THEN RA2=1. RB2=1. RE2=0. RF2=0. PREHEAT_TRL=1.0 M_H_HTF = M_H_HTF - DELT*FR_SR ELSE M_H_HTF = 0 OUTPUT DELARATION 6-42

241 OUT(5)= SOLAR_TRL*(1-HILLER_TRL)*(1-PREHEAT_TRL) OUT(6)= LOAD_TRL - HILLER_TRL OUT(7)= LOAD_TRL OUT(8)= RA2 OUT(9)= RB2 OUT(10)= SOLAR_TRL OUT(11)= LTHARGE_TRL OUT(12)= PREHEAT_TRL OUT(13)= RE2 OUT(14)= RF2 OUT(15)= T_S3 OUT(16)= T_H_AVE OUT(17)= HILLER_TRL OUT(18)= M_H_HTF OUT(19)= T_S EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END Appedix 6.12 code of type 245: ontrol of solar heating with auxiliary heater controlled by constant-outlet-temperature SUBROUTINE TYPE245 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE245 Author: Ming Qu Editor: Date: TRNSYS 7.5 last modified: Jun 2007 NOTE: This controller can only be used with Solver 0 (Successive substitution) *** *** Model Parameters *** setpoint of the temperature of heater [0;+Inf] *** *** Model Inputs *** Flowrate of solar loop kg/hr [0;+Inf] Heating load kj/hr [-Inf;+Inf] Inlet temperature of the heat capacity tank [0;+Inf] *** *** Model Outputs *** Output control of heater - [0.0;1.0] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) 6-43

242 ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE245!SET THE ORRET TYPE NUMBER HERE TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=1,NI=3,NO=2,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE PARAMETERS INPUTS HARATER*12 INTStr HARATER*160 WarnMsg DOUBLE PREISION 1 FR_sr,!FLOW RATE OF THE SOLAR OLLETION LOOP\ 1 HLOAD,!HEATING LOAD 1 HLOAD_TRL,!HEATING LOAD ONTROL 1 SOLAR_TRL,!SOLAR OLLEITON LOOP ONTROL 1 HEATER_TRL,!AUXILARY HEATER ONTROL 1 SOLAR_TRL_LST,!SOLAR PUMP ONTROL AT THE LAST TIMESTEP 1 T_HT_IN,! THE INLET TMEPERATURE OF THE HEAT APAITY TANK 1 HEATER_TRL_LST,!HEATER ONTROL AT THE LAST TIMESTEP 1 TSP!THE SETPOINT OF THE HEATER DATA STATEMENTS DATA YHEK/'MF1','PW1','TE1'/ DATA OHEK/'F','F'/ TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() 6-44

243 SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN DO ALL THE VERY LAST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN PERFORM ANY 'AFTER-ITERATION' MANIPULATIONS THAT ARE REQUIRED HERE e.g. save variables to storage array for the next timestep IF (INFO(13).GT.0) THEN NITEMS=0 STORED(1)=... (if NITEMS > 0) ALL SET_STORAGE_VARS(STORED,NITEMS,INFO) RETURN DO ALL THE VERY FIRST ALL OF THE SIMULATION MANIPULATIONS HERE IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required SET THE REQUIRED NUMBER OF INPUTS, PARAMETERS AND DERIVATIVES THAT THE USER SHOULD SUPPLY IN THE INPUT FILE IN SOME ASES, THE NUMBER OF VARIABLES MAY DEPEND ON THE VALUE OF PARAMETERS TO THIS MODEL... NIN=NI NPAR=NP NDER=ND!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN 6-45

244 !set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values TSP = PAR(1) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) ITYPE=INFO(2)!reread the parameter values TSP = PAR(1) *** PERFORM ALL THE ALULATION HERE FOR THIS MODEL. *** PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of outlet temperature of solar collector control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of outlet temperature of solar collector control LEAR STIK OUNTER ON FIRST ALL OF T-S IF(INFO(7).EQ.0) THEN HEATER_TRL_LST = OUT(1) SOLAR_TRL_LST = OUT(2) FR_sr = XIN(1) HLOAD = XIN(2) T_HT_IN =XIN(3) SOLAR LOOP ONTROL IF (FR_sr.GT.0.) THEN SOLAR_TRL = 1. ELSE SOLAR_TRL = 0. LOADS ONTROL 6-46

245 IF (HLOAD.LE.0.) THEN HLOAD_TRL = 1. ELSEIF (LOAD.GT.0) THEN HLOAD_TRL = 0. HEATER ONTROL HEATER_TRL = 0. IF ((SOLAR_TRL_LST.LT. 0.5).AND. (SOLAR_TRL.GT. 0.5)) THEN IF (T_HT_IN.LT.TSP) HEATER_TRL = 1. IF ((HEATER_TRL_LST.GT. 0.5).AND. (T_HT_IN.LT.TSP)) THEN HEATER_TRL = 1. OUTPUT DELARATION OUT(1)= HEATER_TRL OUT(2)= SOLAR_TRL SET THE STORAGE ARRAY AT THE END OF THIS ITERATION IF NEESSARY NITEMS= STORED(1)= ALL SET_STORAGE_VARS(STORED,NITEMS,INFO) REPORT ANY PROBLEMS THAT HAVE BEEN FOUND USING ALLS LIKE THIS: ALL MESSAGES(-1,'put your message here','message',iunit,itype) ALL MESSAGES(-1,'put your message here','warning',iunit,itype) ALL MESSAGES(-1,'put your message here','severe',iunit,itype) ALL MESSAGES(-1,'put your message here','fatal',iunit,itype) EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END Appedix 6.13 ode of type 244: control of solar cooling with auxiliary heater for preheat SUBROUTINE TYPE244 (TIME,XIN,OUT,T,DTDT,PAR,INFO,INTRL,*) ************************************************************************ Object: Differential Based Solar ontroller; IISiBat Model: TYPE244 Author: Ming Qu Editor: 6-47

246 Date: TRNSYS 7.5 last modified: Jun 2007 NOTE: This controller can only be used with Solver 0 (Successive substitution) *** *** Model Parameters *** No. of oscillations - [1;+Inf] High limit cut-out of circulating fluid temperature [-Inf;+Inf] Temperature set point of the heater [0;+Inf] *** *** Model Inputs *** Flowrate of solar loop kg/hr [0;+Inf] ooling load kj/hr [-Inf;+Inf] HTF temperature at the inlet of the chiller [-Inf;+Inf] Upper dead band of inlet temperature [-Inf;+Inf] Lower dead band of inlet temperature [-Inf;+Inf] Input control function of inlet temperature - [0;1] hilled water supply temperature of HW chiller in the last timestep [-Inf;+Inf] hilled water setpoint of chiller [-Inf;+Inf] Inlet temperature of the heat capacity tank [0;+Inf] *** *** Model Outputs *** Output control signal of HT ctrl - [0.0;1.0] ounter of stick in temperature control - [0.0;1.0] Osillation number of the temperature control - [0.0;1.0] Lastcall output of temperature control - [0.0;1.0] Output control of bypass - [0.0;1.0] Output control of chiller - [0.0;1.0] Output control of fired chiller - [0.0;1.0] Output control of cooling load - [0.0;1.0] Ratio of HTF over bypass - [-Inf;+Inf] Ration of HW - [-Inf;+Inf] Output control of heater - [0.0;1.0] *** *** Model Derivatives *** (omments and routine interface generated by TRNSYS Studio) ************************************************************************ TRNSYS acess functions (allow to acess TIME etc.) USE Trnsysonstants USE TrnsysFunctions REQUIRED BY THE MULTI-DLL VERSION OF TRNSYS!DE$ATTRIBUTES DLLEXPORT :: TYPE244!SET THE ORRET TYPE NUMBER HERE 6-48

247 TRNSYS DELARATIONS DOUBLE PREISION XIN,OUT,TIME,PAR,T,DTDT,TIME0,TFINAL,DELT INTEGER*4 INFO(15),NPMAX,NI,NO,ND,IUNIT,ITYPE,INTRL(4) HARATER*3 OHEK,YHEK USER DELARATIONS - SET THE MAXIMUM NUMBER OF PARAMETERS (NP), INPUTS (NI), OUTPUTS (NOUT), AND DERIVATIVES (ND) THAT MAY BE SUPPLIED FOR THIS TYPE PARAMETER (NP=3,NI=9,NO=12,ND=0,NSTORED=0) REQUIRED TRNSYS DIMENSIONS DIMENSION XIN(NI),OUT(NO),PAR(NP),YHEK(NI),OHEK(NO) INTEGER NITEMS ADD DELARATIONS AND DEFINITIONS FOR THE USER-VARIABLES HERE LOAL VARIABLE DELARATIONS HARATER*12 INTStr HARATER*160 WarnMsg INTEGER NSTK,!No of_oscillations 1 INT_1, 1 NTOLD_1, 1 IOS_1!Osillation number DOUBLE PREISION 1 TMAX,!max temperature at inlet of the chiller 1 LAST_1, 1 LOAD,!cooling load 1 T_chillerin,!HTF_temperature_at_the_inlet_of_the_chiller 1 UP_T,!Upper_dead_band_of_inlet_temperature 1 LOW_T,!Lower_dead_band_of_inlet_temperature 1 LOAD_TRL,!LOAD ONTROL SIGNAL 1 HILLER_TRL,!HW HILLER ONTROL 1 T_HL_PREV_O,!OUTLET TEMPERATURE OF HW FROM THE HILLER 1 IF_1,!ONTROL FUTION INPUTS 1 T_HW_SETPOINT,!HILLED WATER SETPOINT 1 R_BYPASS,!RATIO OF HTF OVER THE BYPASS 1 R_LDIVERTER,!RATIO OF HW OVER THE HW HILLER 1 FR_sr,!FLOW RATE OF THE SOLAR OLLETION LOOP 1 SOLAR_TRL,!SOLAR OLLEITON LOOP ONTROL 1 HEATER_TRL,!AUXILARY HEATER ONTROL 1 SOLAR_TRL_LST,!SOLAR PUMP ONTROL AT THE LAST TIMESTEP 1 T_HT_IN,! THE INLET TMEPERATURE OF THE HEAT APAITY TANK 1 HEATER_TRL_LST,!HEATER ONTROL AT THE LAST TIMESTEP 1 TST_HEATER!THE SET POINT OF SUPPLY TEMPERATURE OF THE HEATER DATA STATEMENTS DATA YHEK/'MF1','PW1','TE1','TE1','TE1','F1','TE1','TE1'/ DATA OHEK/'F1','DM1','DM1','F1','F1','F1','F1','F1', 1 'F1','F1','F','F'/ TRNSYS FUNTIONS TIME0=getSimulationStartTime() TFINAL=getSimulationStopTime() DELT=getSimulationTimeStep() 6-49

248 SET THE VERSION INFORMATION FOR TRNSYS IF(INFO(7).EQ.-2) THEN INFO(12)=16 RETURN PERFORM LAST ALL MANIPULATIONS IF (INFO(8).EQ.-1) THEN IF(ErrorFound()) RETURN 1 INT_1=JFIX(OUT(2)*DELT/(TFINAL-TIME0)*100.d0) IF(INT_1.GE.10) THEN WRITE (INTStr,*) INT_1 WarnMsg='The controller was stuck during '//TRIM(ADJUSTL( & INTStr))//' percent of the simulation timesteps.' ALL MESSAGES(-1,WarnMsg,'WARNING',INFO(1),INFO(2)) RETURN FROM THIS MODEL AS NO "AFTER-ONVERGENE" MANIPULATIONS ARE REQUIRED IF(INFO(13).GT.0) RETURN PERFORM FIRST ALL MANIPULATIONS IF (INFO(7).EQ.-1) THEN!retrieve unit and type number for this component from the INFO array IUNIT=INFO(1) ITYPE=INFO(2)!set some info array variables to tell the trnsys engine how this type is to work INFO(6)=NO!reserve space in the OUT array using INFO(6) INFO(9)=1!this TYPE should be called until convergence is reached INFO(10)=0!no storage spots are required!set THE PROPER NUMBER OF PARAMETERS AORDING TO THE MODE NSTK=JFIX(PAR(1)+0.01) IF(NSTK.LT.0) ALL TYPEK(4,INFO,0,1,0)!ALL THE TYPE HEK SUBROUTINE TO OMPARE WHAT THIS TYPE REQUIRES TO WHAT IS SUPPLIED IN!THE TRNSYS INPUT FILE ALL TYPEK(1,INFO,NI,NP,ND)!ALL THE INPUT-OUTPUT HEK SUBROUTINE TO SET THE ORRET INPUT AND OUTPUT UNITS ALL RHEK(INFO,YHEK,OHEK)!RETURN TO THE ALLING PROGRAM RETURN PERFORM INITIAL TIMESTEP MANIPULATIONS IF (TIME.LT.(TIME0+DELT/2.)) THEN 6-50

249 !set the UNIT number for future calls IUNIT=INFO(1) ITYPE=INFO(2)!read parameter values NSTK = JFIX(PAR(1)+0.01) TMAX= PAR(2) TST_HEATER = PAR(3) HEK TO SEE IF THIS OMPONENT IS BEING ALLED AT THE VERY END OF A TIMESTEP IF(INFO(7).EQ.-2) RETURN THIS IS AN ITERATIVE ALL TO THIS OMPONENT *** RE-READ THE PARAMETERS IF ANOTHER UNIT OF THIS TYPE HAS BEEN ALLED SINE THE LAST TIME THEY WERE READ IN IF(INFO(1).NE.IUNIT) THEN!reset the unit number IUNIT=INFO(1) ITYPE=INFO(2)!reread the parameter values NSTK=JFIX(PAR(1)+0.01) TMAX= PAR(2) TST_HEATER = PAR(3) *** PERFORM ALL THE ALULATION HERE FOR THIS MODEL. *** PERFORM THE ALULATIONS HEK NUMBER OF ITERATIONS OUT(3),A.K.A. IOS_1, IS STIK OUNTER of outlet temperature of solar collector control OUT(2) IS OUNTING TOTAL NUMBER OF TIMES IN SIM. STIK OURS of outlet temperature of solar collector control LEAR STIK OUNTER ON FIRST ALL OF T-S IF(INFO(7).EQ.0) THEN OUT(3)=0.d0 HEATER_TRL_LST = OUT(11) SOLAR_TRL_LST = OUT(12) FR_sr = XIN(1) LOAD = XIN(2) T_chillerin = XIN(3) UP_T = XIN(4) LOW_T = XIN(5) IF_1= XIN(6) T_HL_PREV_O = XIN(7) T_HW_SETPOINT = XIN(8) T_HT_IN =XIN(9) 6-51

250 SOLAR LOOP ONTROL IF (FR_sr.GT.0.) THEN SOLAR_TRL = 1. ELSE SOLAR_TRL = 0. LOADS ONTROL IF (LOAD.LE.0.) THEN LOAD_TRL = 0. ELSEIF (LOAD.GT.0) THEN LOAD_TRL = 1. ******************************** ONTROL HT at inlet of chiller ******************************** TEMPERATURE ONTROL NTOLD_1=OUT(4)! control signal for the last time step LAST_1=OUT(1) IOS_1=JFIX(OUT(3)+0.01) IF(IOS_1.EQ.NSTK) OUT(2)=OUT(2)+1.d0 IF(IOS_1.GE. NSTK) GO TO 15 IF(INFO(7).EQ.0) THEN IF (XIN(6).GT. 0.5) THEN NTOLD_1 = 1 OUT(4)=DBLE(NTOLD_1) ELSE NTOLD_1 = 0 OUT(4)=DBLE(NTOLD_1) IF (NTOLD_1.GT.0.5) GOTO 10!OUTPUT WAS 0 LAST ALL OUT(1)=0.d0 IF (T_chillerin.GT.UP_T) OUT(1)=1.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 GO TO 15!OUTPUT WAS 1 LAST ALL 10 OUT(1)=1.d0 IF (T_chillerin.LT.LOW_T) OUT(1)=0.d0 IF (T_chillerin.GT.175.) OUT(1)=0.d0 6-52

251 15 IF((ABS(LAST_1-OUT(1)).LT.1.E-06).AND.(IOS_1.NE.NSTK)) THEN OUT(3)=OUT(3) ELSE!OUTPUT HAS HANGED STATE SINE LAST ALL OUT(3)=OUT(3)+1.d0 IF THE OUTLET TEMPERATURE OF HILLER IS HIGHER THATN 14 IF(T_HL_PREV_O.LT.13.1) THEN HILLER_TRL = OUT(1)*LOAD_TRL*SOLAR_TRL ELSE HILLER_TRL = 0.0 BYPASS ONTROL BASED ON THE HILLED WATER OUTLET TEMPERATURE IF(HILLER_TRL.GE. 0.5) THEN IF(T_HL_PREV_O.GE.(T_HW_SETPOINT+4.0)) THEN R_BYPASS=0.0 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+3.0)) THEN R_BYPASS=0.2 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT+2.0)) THEN R_BYPASS=0.4 ELSEIF(T_HL_PREV_O.GE.(T_HW_SETPOINT-1.0)) THEN R_BYPASS=0.65 ELSE R_BYPASS=1.0 ELSE R_BYPASS=1.0 HEATER ONTROL HEATER_TRL = 0. IF ((SOLAR_TRL_LST.LT. 0.5).AND. (SOLAR_TRL.GT. 0.5)) THEN IF (T_HT_IN.LT.TST_HEATER) HEATER_TRL = 1. IF ((HEATER_TRL_LST.GT. 0.5).AND. (T_HT_IN.LT.TST_HEATER)) THEN HEATER_TRL = 1. OUTPUT DELARATION OUT(5)= SOLAR_TRL*(1-HILLER_TRL) OUT(6)= HILLER_TRL OUT(7)= LOAD_TRL - HILLER_TRL OUT(8)= LOAD_TRL OUT(9)= R_BYPASS 6-53

252 OUT(10)= HILLER_TRL OUT(11)= HEATER_TRL OUT(12)= SOLAR_TRL SET THE STORAGE ARRAY AT THE END OF THIS ITERATION IF NEESSARY NITEMS= STORED(1)= ALL SET_STORAGE_VARS(STORED,NITEMS,INFO) REPORT ANY PROBLEMS THAT HAVE BEEN FOUND USING ALLS LIKE THIS: ALL MESSAGES(-1,'put your message here','message',iunit,itype) ALL MESSAGES(-1,'put your message here','warning',iunit,itype) ALL MESSAGES(-1,'put your message here','severe',iunit,itype) ALL MESSAGES(-1,'put your message here','fatal',iunit,itype) EVERYTHING IS DONE - RETURN FROM THIS SUBROUTINE AND MOVE ON RETURN 1 END