Transient Analysis of Integrated Shiraz Hybrid Solar Thermal Power Plant Iman Niknia 1, Mahmood Yaghoubi 1, 2

Transcription

1 Transient Analysis of Integrated Shiraz Hybrid Solar Thermal Power Plant Iman Niknia 1, Mahmood Yaghoubi 1, 2 1 School of Mechanical Engineering, Shiraz University, Shiraz, Iran 1, 2 Shiraz University, Academy of Science, Tehran, Iran imanniknia@hotmail.com, yaghoubi@shirazu.ac.ir Abstract Shiraz solar thermal power plant is designed for 250 kw power supply during available sun radiation. It is decided to promote the field of collectors by installing a large parabolic collector and combining the system with a kw hybrid boiler. For the new integrated configuration, thermodynamic analysis is required for engineering design and evaluating thermal performance. For the new system, transient simulation is performed under different working conditions. In the plant, each component is simulated transiently, by considering initial condition and capacity rate of the component as well as all the connecting pipes and instruments. Results of the simulation for thermal performance are compared with field experimental measurements for several periods. Taking into account the thermodynamic concepts and the results of numerical and experimental analysis, the best operation strategies are selected for optimum performance and control philosophy based on the new integrated collector. Keywords: Transient Simulation, Parabolic Concentrator, Experimental Analysis, Solar Thermal Power Plant 1. Introduction The increasing rate of energy demand all around the world and the crises of environmental pollution is one of the major challenges that man has to deal with to develop new energy source for a sustainable development. Green house effects and global warming has made it crucial to devise new green technologies and expand the current rate of renewable energy generation. Being available in vast areas around the world, solar energy is one of the most important renewable energy sources to comply with of the world s need to energy. Due to high prices and low efficiencies encountered for developing solar thermal power plants, selecting efficient working philosophies and improving the working condition of current power plants is very important. Another problem faced for using solar thermal power plants is the unreliability encountered due to variation in environmental conditions such as cloud and wind effects which results in oscillation in the plant performance. In order to resolve the reliability problem, different methods are proposed such as combining the solar thermal power plant with another plant or adding an auxiliary boiler. Shiraz solar thermal power plant with the initial design for 250 kw has been installed in the city of Shiraz. This plant consists of two cycles: an oil cycle and a Rankin steam cycle. In order to make this plant more reliable and to increase its capacity, a new collector is designed, accompanied by an auxiliary boiler, these elements are integrated into the initial power plant. For a detail study of the overall system and evaluating the working philosophy, a fully transient simulation of the system is needed. As a result of calculations, a report of the transient performance of different components can be obtained to optimize the set points defined in the system control philosophy. Thermal simulation models are strong tools which have developed recently for several thermal systems. Stuetzle et al. (2004) with a semi transient modeling investigated the set points for 30 MW SEGS VI [1]. Garcia- Barberena et al. (2009) evaluated the effect of operational strategies on the performance of a solar thermal power plant using SimulCET [2]. Garcia et al. (2009) performed a transient simulation for Nevada solar one power plant using Dinacet. Yao et al. (2009) performed a transient simulation on the pioneer 1MW solar thermal central receiver system in China and studied the system performance under different working conditions. These codes are made for special purposes and they are not available for the present specific simulation. ST: Solar Thermal Application 1715

2 Nomenclature A area m 2 F R collector heat removal factor h enthalpy I t incident solar radiation kj/s.m 2 m mass flow rate kg/s Q u useful energy gain kj/s T temperature collector overall loss factor W/m 2 K World Renewable Energy Congress XI Greek symbols ΔT temperature difference K (τα) n normal transmittance absorptance η efficiency Subscripts i entering e exiting U L env environment es isentropic s steam Therefore in this study a new code is developed which is specially designed for the Shiraz solar thermal power plant (SSTPP). The code developed is unique in the sense that it has various capabilities to comply with elements used in SSTPP and has the best approach toward finding highest performance of SSTPP. 2. Methodology The 250 kw design of Shiraz solar thermal power plant consist of an oil cycle and a steam cycle (Cycles B and C in Fig. 1). A C B Figure 1- Process flow diagram of the new designed system In the new designed system a 100 meter collector is integrated into the system (inserting cycle A of the Fig. 1, to the previous system). Considering the environmental conditions such as wind speed design condition, a computer code is developed for an evacuated tube of parabolic trough concentrating collector based on Eq. (1) [4]. Q u=a C [F R (τα) n I t -F R U L ΔT] (1) The performance of the collector field is highly dependent on the environmental condition such as dust and wind and typical conditions presented in Fig. 2. These parameters are inserted through fouling factors in the collector performance relations. In the new kw design, an auxiliary boiler is also integrated into the system. The collectors field is designed to generate 300 kw power and the remaining power to reach the kw nominal capacity would be provided by auxiliary boiler ST: Solar Thermal Application 1716

3 integrated into the system. The boiler is also to keep the output power steady and reliable when the absorbed irradiance is low due to technical or environmental deficiencies. Figure 2- a typical illustration with dust particles settlement both on the mirror and the absorber tube In this paper one of the control philosophies proposed for the system is considered. With parametric study, control set points are evaluated for the optimum performance of the system. In order to study the system performance, a computer code is prepared and a transient simulation is performed on the entire system. In the first stage of simulation, the code is used to study the performance of the old system to compare the results with the collected data from the power plant. For the transient modelling, attempt is made to take into account the parameters such as wind effect and heat capacities of all components. The thermal programming is modelled in the computer code similar to the approach developed by Schwarzbözl [5]. With a lumped capacity method using an energy balance for a mass with capacity C and initial temperature T0, which is heated by a capacity flow rate Ċ, the capacity of the system is inserted into the prepared transient computer code. For the insulated connecting pipes, their heat capacities are also considered and the pipes are modelled using plug flow model [5]. The pipes are divided into many segments and the environmental loss is evaluated by summing the losses from every single segment by Eq.(2). UA T i T env (2) Pumps in this simulation are single speed with fixed flow rates. Effects of pumps heat generation on the system fluid temperature increase are neglected. Working fluid characteristics are temperature dependent which is considered through the entire simulation. For example the VP1 thermal oil is used in the new collector loop. Typical variation of its property with temperature is demonstrated in the following equation. ρ= T( C) T 2 ( C) T 3 ( C) kg/m 3 (3) As it can be seen in the process flow diagram (Fig. 1), oil cycle consists of 3 shell and tube heat exchangers: A pre heater, a boiler and a super heater heat exchanger. Feedback tanks are modeled as mixing tanks and the heat recovery tank is not modeled in the current simulation. For modeling the turbine an isentropic efficiency for the turbine is assumed. Considering the efficiency of the turbine and the design outlet temperature, the enthalpy of the outlet fluid of the turbine would be obtained using Eq. (4). It is assumed that the quality of the fluid at the outlet of the turbine is 1 and there is no moisture content in the fluid flowing out of the turbine. η turbine =(h i -h e )/(h i -h es ) (4) 3. Results To evaluate the validity of the code, initially comparison is made between the simulation and experimental measurements from the current working plant without the new collector. For the validation process, the radiation data are obtained from a Pyranometer and inserted into the prepared code. Fig. 3-a shows the beam radiation data for 22 th of June 2009 which is used in the validation process. Due to the presence of cloud, wind and dust, the plant performance is quit variable therefore some data such as radiation and wind speed are inserted manually at different simulation times and dust effect is modelled through fouling factors [6]. For the simulation day sky was hazy and wind average speed was 4 m/s. For the simulation wind effect is used to calculate heat transfer ST: Solar Thermal Application 1717

4 coefficients. Fig 3-b shows a sample of scattered clouds in the sky for the simulation day. Figure 4 shows a schematic of the computer simulation performed for the same day. Beam radiation w/m^ :00 10:50 11:40 12:30 13:20 14:10 15:00 15:50 a- Radiation data b- Scattered clouds in the sky Figure 3- Weather condition, 22 th of June 2009 Figure 4- A schematic of the computer simulation for validation From the results of modeling, the temperatures of inlet and outlet of oil from the collectors' field versus time for the simulation day for the model of Fig. 4 are illustrated in Fig. 5. This comparison shows acceptable agreement between experiment and modelling calculations :0011:3012:0012:30 13:0013:3014:0014:3015:0015:3016:0016:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30 16:00 16:30 Experimental results Modeling results Experimental results Modeling results a- Collector's field inlet oil temperature b- Collector's field outlet oil temperature Figure 5- variation of oil temperatures for 22 June 2009 Generated steam flow rate and steam temperature versus time are also compared for the same day and results are reported in Fig 6. Between hours 13:00 to 14:00 and 15:30 to 16:30 two jumps between experiment and modeling are encountered. These jumps are encountered because oil valves for heat exchanger loop is opened manually which is not predefined in the control philosophy. This is done to keep the heat exchangers hot and reduce thermal shocks caused by sudden flow of hot oil. Results of validation of the simulation process show good agreement with experimental data, therefore we can proceed to perform parametric study of the new system modelling shown in Fig. 1. For the rest ST: Solar Thermal Application 1718

5 of the analysis the radiation data are calculated using Daneshyar method [7] and the fouling caused by dust and system performance are neglected and the system at full performance is modelled. Steam mass flow rate kg/s :00 13:30 14:00 14:30 15:00 15:30 16:00 Experimental results Modeling results a- Steam flow rate b- Steam temperature :00 12:30 13:00 13:30 14:00 14:30 15:00 15:30 16:00 Experimental results Modeling results Figure 6- Comparison of steam production between simulation and experiment During analysis, comparison is also made for the oil temperature rise flowing in a section of a pipe with 77 meters length, installed at the outlet of the collector's field as shown in Fig. 7. It can be seen that during the first few minutes of start up the heat capacity causes a delay in the system temperature rise and after a few minutes due to high radiation this effect diminishes and becomes negligible. This effect results in a delay of system response to sudden variation in the environment. The time delay 550 should be studied in order to define an efficient control philosophy for the auxiliary boiler. The control philosophy should be designed in a manner to reduce the effect of 450 system instability as much as possible. 09:00 09:30 10:00 10:30 11:00 11:30 12:00 Beginning of the pipe End of the pipe Figure 7- variation of oil temperature flowing in a pipe at different location For the entire day of the simulation, sum of the energy of the generated steam from Eq. (5) for 3 different control philosophies are determined and compared in Table 1. Sum of the Energy of the Generated Steam = h s *dm s 4 Since the maximum oil outlet temperature of the field is fixed due to technical limitation to 558 K, oil temperature entering the field is compared. Table 1- Sum of the energy of the Generated Steam Oil temperature entering the field Sum of the energy of the Generated Steam 513 K E+07 kj 498 K E+07 kj 483 K E+07 kj With increasing the inlet temperature to the field as a controlling variable with fixed radiation, mass flow rate through the heat exchangers has to be reduced. This causes reduction in heat transfer to the steam. It should also be noted that the minimum temperature of the oil entering the field is controlled by heat exchangers heat transfer capabilities and in this simulation with oil flow rate of 14 kg/s, minimum temperature that can be achieved is no less than 463 K. Next the effect of new collector integration is investigated. The process flow diagram in Fig. 1 is analysed for oil inlet temperature of 483 K leaving other parameters fixed (loops A and B and C are modelled). Since the new designed heat exchanger has limited capacity, it can only transfer limited fraction of the oil energy to the steam. This makes the returning oil to the new collector to be still higher than the design condition. To solve this problem another heat exchanger should be considered in order to heat the oil in the main loop before entering the heat exchanger E203. In this simulation ST: Solar Thermal Application 1719

6 this excess energy is absorbed by a heat sink and only one heat exchanger for the new loop is modelled. Performing transient simulation for the integrated system results are presented in Table 2. It shows the effect of collector integration on the system energy absorption (oil inlet temperature to the field is fixed to 483 K). These results demonstrate the effect of new collector in improving the overall system thermal performance. Table 2-Effect of collector integration method on generated steam for 22 th of June 2009 Method of integrating new collector No integration Integration with one heat exchanger Sum of the energy of E+07 kj E 07 kj the generated steam The effect of the new loop on the steam temperature can be seen in Fig. 6-a. This figure shows noticeable temperature rise for the generated steam. A parametric study on the capacity of the loop s heat exchanger is also performed and the results are presented in Fig. 6-b. It shows the trend of the maximum heat transfer rate of the heat exchanger versus heat exchanger s designed over all heat transfer coefficient. The trend shows that selecting a heat exchanger with overall heat transfer coefficient more than 15 kj/s K is redundant and hardly improves the system performance. Steam temperature K Maximum heat transfer rate kj/s Overall heat transfer coeficient kj/s K Inlet Outlet a- Inlet and outlet steam temperatures b-maximum heat transfer rate versus over all heat transfer coefficient Figure 6- Thermal performance of the integrated system Conclusions Many different studies should be performed in order to design and optimize the performance of a solar thermal power plant (STPP). It is acknowledged that in the design process of a STPP the main goal of the design is to increase the thermal quality or mass flow rate of the generated steam in order to increase power generation of the system. Two major methods can achieve this goal: 1- Increasing the temperature of the outlet steam 2- Increasing the mass flow rate of the outlet steam. In this paper it is shown that the capacity of such systems can be increased with an external loop without changing the main system configuration or design. The proposed design is modelled and simulated with a computer code which proves to be useful in the analysis and improvement of thermal performance of such systems. 5. References [1] Th. Stuetzle, N. Blair, J.W. Mitchell, W.A. Beckman (2004), Automatic control of a 30 MW SEGS VI parabolic trough plant, Solar Energy 76, pp [2] J. García-Barberena, P. Garcia, M. Sanchez, M.J. Blanco, C. Lasheras, A. Padrós, J. Arraiza (2009), Analysis of the influence of the operational strategies in plant performance using SIMULCET, simulation software for parabolic trough power plants, Solar paces 2009, Berlin, Germany. [3] Zh. Yao, Zh. Wang, Zh. Lu, Xiudong Wei (2009), Modeling and simulation of the pioneer 1MW solar thermal central receiver system in China, Renewable Energy 34, pp [4] J.A.Duffie, W. A. Beckman (1991),Solar engineering of thermal processes, John Wiley & Sons, [5] P. Schwarzbözl, D. Zentrum, für Luft und Raumfahrt e.v. (2006), A TRNSYS model library for solar thermal electric components (STEC), Reference manual release 3.0, D Köln, Germany, November [6] K. Azizian, M. Yaghoubi, R.Hesami, S. Mirhadi (2010), Shiraz pilot solar thermal power plant design, construction, installation, commissioning procedure, 7th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics, Antalya, Turkey, July 2010 [7] M. Daneshyar (1978), Solar radiation statistics for Iran, Solar energy 21, pp ST: Solar Thermal Application 1720