E & internal energy generation rate (W)

Transcription

1 Seventh International IBPSA Conferene Rio de Janeiro, Brazil August 13-15, 2001 DOMUS 1.0: A BRAZILIAN PC PROGRAM FOR BUILDING SIMULATION Nathan Mendes, Riardo C.L.F. de Oliveira and Gerson H. dos Santos Pontifial Catholi University of Paraná PUCPR/CCET Thermal Systems Laboratory Rua Imaulada Coneição, 1155 Curitiba PR, Brazil nmendes@et.pupr.br ABSTRACT The software DOMUS has been developed to model oupled heat and moisture transfer in multi-zone buildings. DOMUS has been oneived to be a very user-friendly software so that to stimulating a larger number of users in Brazil to use building simulation software and meet national energy onservation program goals. Besides this aspet, as the weather in Brazil permits the analysis of building arhiteture, thermal omfort passive strategies an be easily analyzed. DOMUS models predit temperature profiles within multi-layer walls for any time step and the temperature and relative humidity for eah zone. In the artile, we will show the DOMUS 1.0 models and interfae and some apabilities. INTRODUÇÃO We desribe the DOMUS software for simulation of multi-zone buildings and analysis of both thermal omfort and energy use. The software DOMUS has been developed to model oupled heat and moisture transfer in buildings, in order to analyze the performane of typial Brazilian rooms when subjeted to any kind of limate onditions. DOMUS has been built in C++Builder whih is an OOP language to be a fast and preise easy-to-use software. The program runs in the Windows 95, 98 and NT operating systems. The user interfae onsists of a series of windows in whih the user an enter the relevant input data and review the results. The user an move between other appliations when the DOMUS simulations are running and several DOMUS projets an be open and running at the same time. DOMUS projets with all their input information and results may be saved and reopened. DOMUS development onsisted of a CACSD environment reation for the analysis of thermal building performane. The program has a simplified CAD interfae so that the user an easily edit the building geometry. The alulation of ondution heat transfer through multi-layer walls is done by using the finite differene approah and, for eah zone, it was employed a apaitive lumped model to alulate the temperature and relative humidity. DOMUS models predit temperature profiles within multi-layer walls for any time step and the temperature and relative humidity for eah zone. Input files ontaining hourly data provide information on the onditions at the interior and exterior of the building, a library of material properties is also available. The onvetion oeffiients at the exterior of the wall are alulated hourly from wind veloity and diretion data. DOMUS has been oneived to be a very userfriendly software so that to stimulating a larger number of users in Brazil to use building simulation software and meet Brazilian energy onservation program goals. Besides this aspet, as the weather in Brazil permits the usage of passive arhiteture, DOMUS will have a module, whih will allow the user to simulate, in an easy way, passive or lowenergy strategies to reah thermal omfort onditions. In the artile, we will show the DOMUS 1.0 models and interfae and some apabilities. MATHEMATICAL MODEL The present work uses a dynami model for analysis of the hygrothermal behavior of a room without HVAC system. Therefore, a lumped formulation for temperature as well as for water vapor is adopted. Eq. 1 desribes the energy balane, where the room is submitted to loads of ondution, onvetion, shortwave solar radiation, inter-surfae long-wave radiation and infiltration. dt E& int t + E& g = ρv (1) dt E & t energy flow that rosses the room (W) E & internal energy generation rate (W) g ρ density (kg/m 3 ) speifi heat of (J/kg-K) V room volume (m 3 ) T int room temperature ( o C)

2 E & t, on the energy onservation equation, The term inludes loads for building envelope (ondution), fenestration (ondution and solar radiation) and openings (ventilation and infiltration). The ondution heat flux - Q (t) - that rosses the room ontrol surfae is alulated by the Newton s law for ooling, Q(t) = ha T [ (t) T (t)] & (2) n int where h represents the onvetion heat transfer oeffiient, A, the heat transfer area, and T n (t) the envelope internal surfae temperature. This temperature is alulated by an energy balane, in an elemental volume, using the Fourier s law as it is presented below: 2 T T = λ 2 t x ρ (3) Thus, the temperature T shown in Eq. 3, is the temperature for a ontrol volume within the building envelope, alulated as a funtion of the following thermophysial onstants: density (?), speifi heat () and thermal ondutivy (?). On the external side of the room, the walls, eiling, doors and windows are exposed to solar radiation and to onvetion heat transfer. This way, the external boundary ondition (x=0) of Eq. 3 an be mathematially expressed as: T ( Text Tx 0 ) + αqr λ = hext = x x= 0 where: ext ( T T ) h onvetion heat transfer (W/m2) ext x=0 α q r absorbed solar radiation (W/m2) λ thermal ondutivity (W/m-K) (4) On the internal side (x=l), we have inluded the inter-surfae long-wave radiation as: T λ = h ( T Tx= L ) + x (5) ) int int x = L 4 4 σf f εθ(t sur T x = L where: f f shape fator. ε emissivity. σ Stefan-Boltzmann onstant ( W/m 2 -K 4 ) T sur temperature of internal surfaes of surrounding walls (K). T = The temperature x L of Eq. 5 is equivalent to a temperature of the n-th node of the wall the temperature needed to alulate Q & (t). For the floor, we have adopted the imposedtemperature boundary ondition, equaling T x= 0 to the ground temperature at a depth of 2m. On the other hand, for the eiling, long-wave radiation losses were onsidered (R lw ) so that Eq. 4 has assumed the following form: T λ = hext = x x= 0 where the term emissivity. ( Text Tx 0 ) + αqr ( ε) eilrlw ( ε ) eil represents the eiling The infiltration loads formulation was taken from ASHRAE (1993). The solar radiation (diret and refleted) ame from models presented by Szokolay (1993) and ASHRAE (1993). In terms of water-vapor balane, it was onsidered ventilation, infiltration and internal generation from equipment and people breath so that the lumped formulation beomes: (& & )( ) m inf where: m& inf m& vent Wext Wint m + m W W + m& + m& vent ext int b ger = ρ mass flow by infiltration (kg/s) mass flow by ventilation (kg/s) V dw dt external humidity ratio (kg water/kg dry ) internal humidity ratio (kg water/kg dry ) b water vapor flow from the breath of oupants (kg/s) m& ger ρ V internal water-vapor generation rate (kg/s) density (kg dry /s) room volume (m3) The water-vapor mass flow from the people breath is alulated as it is shown in ASHRAE (1993), whih int

3 takes into aount the room temperature and humidity ratio and physial ativity as well. The eletri oil-heating system that is optional an be desribed as: ρ V dt dt = P () t ha[ T () t T () t ] A where P(t) is the thermal power generated by the heater, ρ, the density,, the speifi heat, V, the heater oil volume, T A (t), the room temperature, T (t), the heater temperature, t, time, h, onvetion heat transfer oeffiient and A the heat exhange area. DOMUS INTERFACE Figure 1 - Adjaent walls with total symmetry In the ase shown in Fig. 1, the geometri treatment is simple as the 1 st floor eiling just plays either a role as a eiling or as a floor. The software interfae is modularly divided in the parts that are desribed below. Building Desription Module This module is responsible by the definition of the building geometry and dimensions. A building an have several zones and eah zone an have several walls and eah wall an have several layers and eah layer shall be defined by its physial harateristis. The user an interat with building envelope and openings and after defining the geometry, a text file will be automatially reated so that the user an verify all the inputs. Inter-Zone Relationship As a building an have several zones, an important aspet is how to geometrially oneive the interzone relationship in order to have a generi ode. This implementation stage is one of the most omplex and that takes longer to get ready. As the software allows the user to freely deide what he wants to, all the geometri onsisteny has to be transferred into the ode. The theoreti basis for the solution of these omplexities is the omputational geometry diretly turned to arithmeti and logis operations between 2-D flat surfaes. Figure 2 - Adjaent walls with no symmetry. In Fig. 2, we show the ase in whih the first floor eiling may be either onneted to other 2 zones or exposed to outdoor onditions. In this ase of adjaeny with no symmetry, DOMUS distinguishes whih surfae area is external or internal so that it gives a speial mathematial treatment for the boundary onditions. We show below some examples that require a speial treatment. Figure 3 - Partially adjaent walls

4 In Fig. 3 is shown the ase where the walls are partially adjaent. Atually, this is the most omplex ase. Only a part of eah wall has an adjaeny (region R) to another zone. The program exeutes an intersetion operation (operation AND ) between both walls. The results will be a region defined for a retangle. If openings are added within the adjaeny area, they should be onsidered in the intersetion operation as well, whih inreases the algorithm omplexity. Configuration Module In this module, it is allowed to the user onfigure all the parameters diretly related to the simulation, whih are ited below: Building Envelope Editor The wall layers are edited in a partiular window (Fig. 6), where it is shown all the available materials, geographial orientation (wall azimuth and inlination), soil properties, painting, et. Interfae The edition interfae (Fig. 4) is simple, allowing the user to build a onstrution without muh of speifi knowledge. There are two visualization panels, one showing the whole building and a seond one showing just the urrent wall in the edition module where the user an hange its dimensions and physial properties. By liking on that wall, the user an open the building element desription window (Figure 6). Figure 6 Building envelope desription window. Weather The external building limate is hosen in the weather onfiguration window (Fig. 7). To make easier, the program load all the existing weather files in a speified diretory and list all of them to the user. Figure 4 Building edition general sheme. For the 3-D visualization panel, it is provided tools to rotate, translate and hange building sale and olors. A building an be loaded from text files saved in a speial DOMUS format. In Fig. 5, it is shown the main sreen layout. Figure 7 Weather Configuration window The DOMUS hourly weather files provide dry bulb temperature, relative humidity, diret and diffuse solar radiation and wind speed and diretion. Figure 5 - Main sreen layout Presently, the program reads just weather files in the DOMUS format (.dom). Afterwards, it will be added options for several weather file formats suh as TRY (Test Referene Year), TMY (Typial

5 Meteorologial Year) and WYEC (Weather Year for Energy Calulations). The internal and external onvetion heat transfer oeffiients are also onfigured on the weather window. In the ase, the user wants the oeffiients to be alulated in time of exeution, he just has to lik in the option. DOMUS has orrelations for those oeffiients obtained by Allandari and Hammond (Clarke, 1985). Internal Gains Figure 8 shows the window where the user an onfigure the internal energy gains of eah zone. The number of gains is limitless and an be from different sort suh as people, equipment, lights, vapor generation et. Figure 9 HVAC systems window. Simulation General parameters Figure 10 shows the general parameters onfiguration window. Figure 8 Internal gains As the user enters the gains, a figure is displayed to show the internal power for eah hour. Beyond that, in this window, ventilation and infiltration an be onfigured for eah partiular zone. HVAC Systems In this window, the user will be able to speify HVAC equipment for eah zone. The user will also be able to define a ontroller for eah equipment in eah zone. Figure 10 - general parameters onfiguration window Output Reports Before exeuting the simulation, the user has to hoose all the types of reports that the program will have to generate. However, the higher the number of options hosen slower will be the simulation. The program provides a great variety of report options suh as hourly reports for temperature, relative humidity, thermal omfort level (PMV and PPD), energy onsumption, thermal loads, energy osts, and monthly general statistis. The report options an be seen in Fig

6 Figure 12 Building sample Fig. 13 shows the internal and external temperature and relative humidity. Figure 6 - Report onfiguration window It is possible to the user to monitor the temperature and/or the ontroller output with different simulation time steps. It is important to remember that for very small time steps, the program will generate very large files. RESULTS Figure 13 - Internal and external temperature and relative humidity. Fig. 14 presents the Fanger`s omfort indexes (PMV and PPD) along the month of July in the ity of Rio de Janeiro, whih shows a good predit mean vote with a low perentage of dissatisfied people. We have hosen one-month period from July 1st to July 31st to simulate the building shown in Figure 11. We have seleted the Rio de Janeiro weather file with a time step of 5 minutes. Figure 14 - Report of PMV and PPD

7 The graphs presented in Figures 13 and 14 an be saved as a text file. We an also zoom them in/out and selet a different plotting time step. CONCLUSIONS We have desribed the DOMUS building simulation program, whih has a CAD interfae The software was developed for building thermal analysis, by using the finite differene method. DOMUS has been built in C++Builder whih is an OOP language to be a fast and an easy-to-use software. The program runs in the Windows 95, 98 and NT operating systems. The user interfae onsists of a series of windows in whih the user an easily enter the relevant input data and review the results. The user an move between other appliations when the DOMUS simulations are running and several DOMUS projets an be open and running at the same time. DOMUS projets with all their input information and results may be saved and reopened. The program provides a great variety of report options suh as hourly reports for temperature, relative humidity, thermal omfort level (PMV and PPD), energy onsumption, thermal loads, energy osts, and monthly general statistis. For further work, we will inorporate within DOMUS models the UMIDUS program features, inreasing its potential for hygrothermal simulation. REFERENCES ASHRAE Handbook-Fundamentals, Atlanta: ASHRAE, CLARKE, J.A., Energy Simulation in Building Design, Adam Hilger Ltd., USA, CRAWLEY, D. B.; LAWRIE, L. K. e PEDERSEN, C. O.. Energyplus: a new generation building energy simulation program. Proeedings of renewable and advaned energy systems for the 21st entury, Lahaina, Maui, Havaí (EUA), ESRU. ESP-R: a building and plant energy simulation environment, user guide version 9. ESRU publiation University of Strathlyde. Glasow (Reino Unido),1997. FAUCONNIER, R.; GUILLEMARD, P. e GRELAT, A. Algorithmes des simulateurs du omportement thermique des bâtiments BILGA et BILGO. Annales de I ITBPT. N 458, p ,1987. Porous Building Elements"; Building Simulation Conferene - IBPSA 99, p , Japan MENDES N., MENDONÇA K.C, SANCHEZ H.M. e FONSECA C.E.L., Simulação Computaional do Edifíio Sede da TELEPAR: Analise Energétia e Proposta de Retrofit, V Enontro Naional de Conforto Térmio no Ambiente Construído/ENTAC e II Enontro Latino Ameriano de Conforto no Ambiente Construído, Fortaleza-CE, novembro de MENDES N., ARAÚJO H.X. e OLIVEIRA G.H.C., O Problema do Controle de Temperatura em Aqueimento de Edifiações, VIII Enontro Naional de Tenologia do Ambiente Construido (ENTAC 2000), Salvador, 2000a. MENDES N., OLIVEIRA G.H.C e ARAÚJO H.X., O Uso do Matlab/Simulink para Análise de Comportamento Térmio de Ambientes, VIII Enontro Naional de Ciênias Térmias - ENCIT 2000, Porto Alegre, 2000b. MENDES N., Sun, Rain and Moisture Influenes on Condution Cooling Loads, International Conferene on Comfort and Thermal Performane of Buildings (COTEDI-2000), Venezuela, Junho, MENDES N., The Potential of Passive Latent Cooling in Brazil, XVII PLEA - Passive and Low Energy Conferene, England, 2000d. MENDES N., RIDLEY I. and LAMBERTS R., Condution Heat Transfer through Porous Walls in Brazil, 34th National Heat Transfer Conferene, ASME Amerian Soiety of Mehanial Engineers, Pittsburgh, Pennsylvania, August 20 22, 2000e. PRESSMAN, R.S. " Engenharia de Software'; Makron Books do Brasil, 1995; ROUX, J. J. Proposition de modèles simplifiés pour l étude du omportement thermique des bâtiments, thèse de Doteur-Ingénieur, INSA de Lyon, Lyon Frane, WALTON, G. N.. Airflow network model for element-based building flow modelling. ASHRAE Transations. Vol. 95, part 2, p , WINKELMANN, F. C.; BIRDSALL, B. E. e BUHL, W. F.. DOE-2 Supplement Version 2.1, LBL Lawrene Berkeley National Laboratory, EUA, MENDES N., RIDLEY I., LAMBERTS R., PHILPPI P.C. and BUDAG K.," UMIDUS: A PC Program for the Predition of Heat and Moisture Transfer in

8 - 90 -