Integrated Heat Air & Moisture Modeling and control

Integrated Heat Air & Moisture Modeling and control van Schijndel, A.W.M. Published in: Proceedings of the 7th International Conference System Simulation in Buildings (SSB 2006), Liege, December 2006

Published: 01/01/2007

Document Version Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication: • A submitted manuscript is the author’s version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher’s website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

Citation for published version (APA): Schijndel, van, A. W. M. (2007). Integrated Heat Air & Moisture Modeling and control. In Proceedings of the 7th International Conference System Simulation in Buildings (SSB 2006), Liege, December 2006. (pp. 1-18). Technische Universiteit Eindhoven.

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ? Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Download date: 07. Oct. 2016

P18, Page 1

Integrated Heat Air & Moisture Modeling and Control A.W.M. (Jos) van Schijndel Eindhoven University of Technology, Netherlands email: [email protected] ABSTRACT The paper presents a recently developed Heat Air & Moisture Laboratory in SimuLink. The simulation laboratory facilitates the integration of the following models: (1) a whole building model; (2) Heating Venting and Air-Conditioning and primary systems; (3) 2D indoor airflow, 3D Heat Air & Moisture transport in constructions and 2D airflow/driving rain. Preliminary results of a new verification study are provided in more detail in this work. Furthermore, a case study, comprehending the impact of Heating Venting and Air-Conditioning control set point strategies on the indoor climate of a monumental church and it's organ, shows how the Heat Air & Moisture Laboratory might be used for design oriented users. It is concluded that the presented simulation environment is capable of modeling and solving a large range of complex integrated Heat Air & Moisture problems and maybe quite useful for design. 1. INTRODUCTION It is widely accepted that simulation can have a major impact on the design and evaluation of building and systems performances. Also, the modeling and simulation of whole building Heat, Air & Moisture (further-on called HAM) responses in relation with human comfort, energy and durability are relevant. The International Energy Agency (IEA) has confirmed this relevance by organizing an Annex project (IEA Annex 41 2005) that will focus on a holistic approach of HAM transfer between the outside, the enclosure, the indoor air and the Heating, Ventilation and Air-Conditioning (HVAC) systems. In this research project the Heat, Air and Moisture Laboratory (HAMLab) is one of the participating simulation tools. An important objective of the mentioned Annex 41 is technology transfer, i.e. the Annex 41 results should be accessible to everybody outside the scientific community. If we consider HAMLab as a technology, the next questions arise: How can we improve the accessibility of the HAMLab modeling technology to designers, educators, practitioners and building engineers? What are the benefits and drawbacks? The aim of this paper is to answer these key questions. The outline is as follows: Section 2 provides a short review on HAM models. Section 3 presents the modeling and simulation environment HAMLab. Detailed information on the physics of the models and validation data can be found in the references. Preliminary results of new verification studies are provided in more detail. Section 4 shows a guideline to improve the accessibility of HAMLab. Section 5 presents the application of this guideline by a case study. Finally, in section 6, the key questions will be revisited and discussed. 2. REVIEW OF HAM MODELING TOOLS This Section summarizes recent HAM models. The aim of this Section is to identify common HAM related modeling problems. The work in this paper is related to: 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 2 (1) IEA Annex 41, (2) Building energy programs, (3) Modeling and simulation techniques; (4) Matlab/SimuLink related tools. For each of the four items, a summary of involved HAM models is presented below. First, currently, 15 different tools are used in IEA Annex 41 common exercises about simulating the dynamic interaction between the indoor climate of a room and the HAM response of the enclosure. All tools model the indoor air and the enclosure. Seven HAM models are stand-alone simulation tools and have promising capabilities for simulating HVAC systems: Bsim, IBPT, IDA-ICE, TRNSYS, EnergyPlus, PowerDomus and HAMLab. The latter is presented in this paper. A complete overview will be published with the final reports, expected in 2008. Second, the energy related software tools at the Energy Tools website U.S. Department of Energy U.S (2006) has been used for several comparison studies. A recent overview is provided by Crawley et al. (2005). Furthermore, Schwab and Simonson (2004) used the same website for a study on programs that might simulate whole building HAM transfer. The criteria were based on: moisture storage in building materials, calculate indoor climate, moisture exchange in HVAC system, access to source code. Three tools met all criteria: Bsim, TRNSYS and EnergyPlus. These tools are already mentioned. Third, Gough (1999) reviews tools with the focus on new techniques for building and HVAC system modeling. In this study four new simulation techniques were investigated. The equationbased method is most relevant for this work. The other three techniques: modal, stochastic and neural networks are not based on physical parameters and are therefore less suitable for solving design problems. The next three equation based method techniques of Gough (1999) are already included in one or more tools of the previous Section: Neutral Model Format and IDA solver are included in IDA-ICE and TRNSYS. The other 11 mentioned equation based method techniques of Gough (1999) are simulation environments with limited HAM models for building simulation. Fourth, Riederer (2005) provides a recent overview of Matlab/SimuLink based tools for building and HVAC simulation. The tools who are most related with the work in this paper are discussed now. SIMBAD provides HVAC models and related utilities to perform dynamic simulation of HVAC plants and controllers. The toolbox focuses on thermal processes with limited capabilities for moisture transport simulation. In addition to this work, this paper presents how models that include moisture transport, can be simulated in SimuLink. The International Building Physics Toolbox (IBPT) is constructed for the thermal system analysis in building physics. The tool capabilities also include 1D HAM transport in building constructions and multi-zonal HAM calculations (Sasic Kalagasidis 2004). All models including the 1D HAM transport in building constructions are implemented using the standard block library of SimuLink. The developers notice the possibility to couple to other codes / procedures for 2D and 3D HAM calculations. In addition to this work, this paper shows how this can be done using FemLab. All HAM tools mentioned, face at least one limitation that cannot be solved by the tool itself. Either a problem occurs at the integration of HVAC systems models into whole building models, or a problem occurs at the integration of 2D and 3D geometry based models (for example airflow and HAM response of constructions) into whole building models. The first problem is caused by the difference in time constants between HVAC components and controllers (order of seconds) and building response (order of hours). This can cause inaccurate results and long simulation duration times (Gouda et al. 2003 and Felsman et al. 2002). The second problem is caused by 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 3 the lack of lumped parameter tools that include internal 2D, 3D finite element method (FEM) capabilities (Sahlin et al. 2004). In order to be able to solve all above mentioned problems in a single simulation environment, this environment should be able to integrate: (1) a whole building modeling tool for the simulation of the indoor climate and energy amounts; (2) a high spatial-resolution tool, for the simulation of HAM responses of building constructions and internal/external airflow, for example a 2D and 3D FEM PDE solver; (3) a high time-resolution tool, for the accurate simulation of HVAC systems and controllers, for example a high quality ODE solver. These three tools were developed and implemented using MatLab/SimuLink/FemLab in the socalled Heat, Air & Moisture Laboratory (HAMLab) It will be presented in the following Section with the focus on already published and new preliminary verification and validation studies. 3. HAMLAB The best way to describe HAMLab is a collection of HAM models and modeling facilities running in the Matlab/SimuLink/FemLab simulation environment. 3.1 Whole building modeling tool The whole building model originates from ELAN, a computer model for building energy design (de Wit and Driessen 1988). The ELAN model, together with an analog hygric model has been implemented in a Building Physics Toolbox in MatLab (Schijndel and de Wit 1999). A major recent improvement is the development of a model in SimuLink (de Wit 2006). The implemented numerical model consists of a continuous part, solved with a variable time step and a discrete part, solved with a time step of one hour. For the HVAC system and the room response of indoor climatic variations, a continuous model is used. For the external climate a discrete model is used. The main advantages of this numeric hybrid approach are: (1) the dynamics of the building systems where small time scales play an important role (for example on/off switching) are accurately simulated; (2) the model becomes time efficient as the discrete part uses 1-hour time steps; (3) the moisture (vapor) transport model is also included. With this feature, the (de-) humidification of HVAC systems can also be simulated. In (van Schijndel 2003c), the heat transport part of HAMBase is verified with a standard test method ASHRAE (2001). In (Schellen 2002), the HAMBase model has been subjected to a validation study. The measured and simulated air temperature and Rh of several church are compared. Although the final results of the IEA Annex 41 become available for the public in 2008, preliminary results of both a verification and a validation exercise are presented already below. The verification exercise comprehends the dynamic interaction of the indoor climate of a room and the HAM response of the enclosure The geometry is identical to the one in the ASHRAE standard test (ASHRAE 2001). Further test conditions are as follows: (1) there is only one 1D construction type with linear properties; (2) the exposure is completely isothermal; (3) the outside relative humidity (RH) is 30%; (4) there are no windows; (5) the internal gains equal 500g/hour; (6) the ventilation rate is 0.5 ach. The exercise seeks to provide the analytical 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 4 solution as well as the simulation results of the model. The model solution is obtained by applying HAMBase. The exact solution is obtained by integration of a FemLab model of the construction into a first order model of the room modeled in SimuLink. The FemLab model block is based on the following PDE: ∂w = ∇ ⋅ (D w (w)∇w) ∂t

(1)

where w is moisture content, t is time and Dw is diffusivity. In figure 1, the exact solution and the model solutions are shown.

Figure 1. Verification results: The indoor air RH of a room including moisture buffering of the air and one construction These preliminary results show that the exact solution obtained using FemLab/SimuLink and the HAMBase models solutions are satisfactory. The three HAMBase models represent one continuous model (HAMBase-SimuLink) and two discrete models HAMBase and HAMBaseR. The latter is a (R)esearch version, including a even more accurate modeling of the constructions. More details can be found in (Schijndel 2004).

7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 5 The subject of the validation exercise was to simulate two real test rooms, which are located at the outdoor testing site of the Fraunhofer-Institute of building physics in Holzkirchen. During the winter (2005-2006) tests were carried out with the aim to compare the measurements with the models developed within the Annex 41. In the reference room a standard common used gypsum plaster with a latex paint is used. The walls and the ceiling of the test room are fully coated with aluminium foil. The next figures present the measured and simulated Rh (figure 2) and heating power (figure 3) in the reference room.

Figure 2. The measured and simulated Rh in the reference room

Figure 3. The measured and simulated heating power in the reference room 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 6 The results show a very good correlation between simulation and measurement for the reference room. Similar results were obtained with the test room. 3.2 High spatial-resolution tool HAM responses of building constructions and internal/external airflow are modeled and simulated with FemLab (now called COMSOL). This is an environment for modeling scientific and engineering applications based on partial differential equations (PDE). Two FemLab models are developed as part of the benchmark study, concerning 1D HAM transport in constructions (Hagentoft et al. 2002). One FemLab model shows an excellent agreement with the solution of the analytical case. A second FemLab model, simulating the insulated roof case, shows a good consensus with the other solutions. In (van Schijndel 2003a) a validation of the simulated water absorption in an initially dry brick cylinder (length 24 mm) is presented. All sides except the bottom are sealed. One side is submerged in water at t=0 sec. The measured 1D water absorption process of several brick materials is successfully simulated. In (van Schijndel 2003a) a validation of a combined heat and moisture transport for a 2D geometry is presented. The problem is based on a 2D-problem presented in (Kunzel et al. 1996): a brick specimen, initially dry and with two of the flanks sealed, is placed with its lower surface about 1 cm under water. The simulated moisture uptake profiles agree well with the measurements of (Kunzel et al. 1996). In (van Schijndel 2003a) a CFD model in FemLab for indoor airflow is successfully verified with the results of (Sinha 2000). The problem is modeled by 2D incompressible flow using the Boussinesq approximation with constant properties for the Reynolds (Re) and Grasshof (Gr) numbers. The verification study included several combinations of Re and Gr . In (van Schijndel 2005a) a k-epsilon model in FemLab concerning outdoor airflow (wind) is verified by another exercise of the mentioned IEA Annex 41 exercise, on boundary conditions. The preliminary results are in good agreement with the results obtained by other models (Fluent) 3.3 High time-resolution tool HVAC systems and controllers are modeled and simulated using SimuLink. This is a platform for the simulation of dynamic systems. In (van Schijndel 2003c), results of ODE based models of a heat pump, an energy roof and a thermal energy storage (TES) are compared with measurements. In (van Schijndel 2005b), a room of a famous museum in the Netherlands is simulated, using HAMLab. The integrated HAM model in SimuLink, consisting of the room, HVAC system and controller was validated by measurements. The model was successfully used to design a new HVAC controller strategy. New preliminary validation results are provided by Neuhaus and Schellen (2006). A comparison of the indoor climate for a usual thermostatic heating and a hygrostatic heating was gained for measurements on a test site. The results were compared with predicting simulation results from a simulation model. Figure 4 shows the results of a humidistatically controlled room.


P18, Page 7

Figure 4 Measured (red) and simulated (blue) of a hygrostatic controlled indoor climate (top: temperature, bottom: Relative humidity) 4. GUIDELINE This Section provides a step-by-step guideline. Each step requires more knowledge of the building and systems details as well as the modeling environment. After each step, the simulation results should be evaluated. If the results are not satisfactory due to oversimplification of the model(s) go the next step. Step 0: Selection of a similar building. The first step neither requires Matlab nor specific knowledge of modeling parameters. At the website of HAMLab there is a web base application containing already simulated results of several building types (HAMcases). For each building type graphical output is presented including time series and statistical information of the temperature, relative humidity (Rh), thermal comfort, heating, cooling and (de)humidification in each zone for a specific period (year, month week). A building should be selected that is most similar to the design. There is small change that the design and the selected building from the database are almost the same. In this case no further simulation is required. Step 1: Editing/simulating the HAMBase model This step requires software package Matlab and basic knowledge of HAM parameters of the building. After downloading HAMLab (public domain), the selected building of step 0 can easily 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 8 be simulated by typing the name of building type at the Matlab prompt. The same results as presented at the website occur. The input file is a text file that can be edited to match the design. The model parameters are explained in the same file. The HAMBase model facilitates default systems and controllers. If it is necessary to model more details of the systems and controllers go step 2. Step 2: Including detailed systems and controllers This step also requires: SimuLink, included in MatLab, some experience with this tool and basic knowledge of systems and controllers. It is very easy to export the HAMBase model to a SimuLink block. The input/output structure of this block is as follows: for each zone the input consists of a heat and moisture source and the output consists of the air temperature, comfort temperature and Rh. The exported SimuLink model simulates a free-floating situation of the building, where all systems and controller settings of the original MatLab model are ignored. Detailed systems and controllers models can be added in SimuLink in order to simulate a controlled situation of the building. Note that also models mentioned in Section 3.3 can be used in this stage. If it is necessary to model more details of the constructions or airflow go step 3. Step 3: Including detailed HAM/Airflow models This step requires: Beside the Matlab software, a separate toolbox FemLab (=COMSOL), experience with this tool and specific knowledge of PDE modeling. To integrate HAM/Airflow models into SimuLink, the next sub-steps are required: (a) Select a FemLab model that is most suitable for the problem. Inspiring models can be found at the FemLab and HAMLab websites; (b) adapt the model, start simulations and evaluate the simulation results. If satisfactory, (c) export the FemLab model to SimuLink using the standard export facility or using S-Functions as shown in (van Schijndel 2003d). Step 4: Evaluation If the final model is ready, the simulation results should be evaluated and if necessary the modeling parameters and/or the models themselves should be modified. 5. CASE STUDY We revisit the work of (van Schijndel et al. 2003b) and apply the guideline. In the Walloon Church in Delft a monumental church organ is present which has been restored in the spring of 2000. To prevent causing damage to the organ again, the indoor climate has to meet certain requirements. The main objectives were: (I) Development of a integrated model for simulating the indoor climate, the detailed moisture distributions of wood and the HVAC system. (II) Evaluation of the current set point operation strategy of the HVAC system of the Walloon church. (III) Development and evaluation of new strategies including RH control. Step 0: Selection of a similar building. Select a church from the HAMcases website and download the HAMBase input mfile. Step 1: Editing/simulating the HAMBase model Edit the input mfile, using the details of the geometry, material properties and boundary conditions of the church model as provided by (Schellen 2002). The hourly simulated air 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 9 temperature and relative humidity of one month (December 2000) inside the church is presented in figure 5, together with measured values. We have to include a more detailed controller and a detailed moisture transport model, so we proceed with the next step.

Figure 5. Validation of the HAMBase model. The measured and simulated air temperature and relative humidity of one month (December 2000) are compared. Step 2: Including detailed systems and controllers After exporting the HAMBase model to SimuLink, a church building model block, with the input/output structure for the church model (containing 2 zones: church and attic), appears in SimuLink , see figure 6.

Figure 6. The church model in SimuLink. With Q = heating power [W], g = humidification [kg/s], T comfort = comfort temperature [oC], T air = air temperature [oC], RH air = air relative humidity [-]


P18, Page 10 The HVAC system is modeled by standard blocks of SimuLink as shown in figure 7.

Figure 7. The HVAC system and controller model in SimuLink The set point of the air temperature is generated by a pulse block with properties: Period: 1 week, start time 04.00 o'clock Sunday, duration: 12 hours, lower value: 10 oC higher value 20 oC. The input of the PID controller consists of the set point minus the actual air temperature. The settings of the PID controller are: P = 107, I=D=0, so in this case it acts like a proportional controller. The output of the controller is limited between 0 en 90 kW. In order to include a detailed moisture transport model, we proceed with step 3. Step 3: Including a detailed HAM model Although FemLab is well equipped for solving complex building physics problems (van Schijndel 2003a), the FemLab model in this application is quit simple. The emphasis of this study lies not on the complexity of the individual models but on the complexity of the combination of models. The moisture transport is assumed to be 1D, dominated by vapor transport. Therefore model (1) is considered as a good ‘starting’ PDE model for the simulation of dynamic moisture transport in wood. How (1) can be modeled in FemLab is explained in (van Schijndel et al. 2003b). This FemLab model is exported to SimuLink using the standard export facility. In figure 8, the input/output structure is shown.

Figure 8. The moisture transport model in SimuLink. The moisture exchange rate in [kg/m2s], w surface = moisture content near the wood surface [kg/m3] and w mean = mean moisture content of the wood [kg/m3]. In figure 9 the results of a validation study by Schellen (2002) is presented. 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 11

Figure 9. Simulation and measurement of moisture profiles in case of drying of a cylinder of wood (diameter 25 mm) by a step in relative humidity from 85% to 35%. The connecting of all models, shown in figure 10, finalizes the complete model

Figure 10. The complete model in SimuLink Step 4: Evaluation The model shown in figure 10 is used to study the moisture content near the surface and the drying rate (defined as the rate of change of moisture content near the surface) for two cases: a) 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 12 no heating and b) full heating capacity, i.e. no limitations in air temperature or relative humidity changing rate. Figure 11 shows the simulated indoor air temperature, relative humidity and moisture content of the wood near the surface for the two cases. The simulation period is one month. The church is heated four times a month.

Figure 11. The simulated indoor air temperature, the relative humidity and the moisture content of the wood near the surface. The simulation period is December 2000. From figure 11 the difference in peak drying rate for the two cases, no heating and full heating capacity, is very clear: The peak drying rate in the case of full heating is of order 100 times bigger than in the case of no heating. These peaks can cause drying induced stresses and have to be minimized to prevent possible damaging of the wood. In (van Schijndel et al. 2003b) two more advanced control strategies were simulated. The first type is a limited indoor air temperature changing rate. The second type is a limited indoor air relative humidity changing rate. Recommendations from international literature suggest that a changing rate of 2 oC/hour will preserve the interior of churches The study shows that a limitation of indoor air temperature changing rate of 2 oC/hour can reduce the peak drying rates by a factor 20. A limitation of the relative humidity changing rate of 2 %/hour can reduce the peak drying rates by a factor 50. The latter has the disadvantage that the heating time is not constant. The 2 oC/hour control strategy was implemented in the real church. Evaluation after a year showed that the operation strategy functioned satisfactory.


P18, Page 13 6. DISCUSSION Accessibility of the HAMLab models for design The case study shows that in order to evaluate peak drying rates in monumental objects (i.e. church organs) for several control strategies of the heating systems, it is essential to use an similar integrated HAM model as presented in this paper. Furthermore, the provided guideline and the application of it in the form of a case study, might improve the accessibility of the HAMLab models for design. Limitations, drawbacks and benefits It may be concluded that the simulation environment HAMLab is capable of solving a large range of integrated HAM problems. Furthermore, it is promising in solving common modeling problems of other HAM models such as problems caused by the difference in time constants between HVAC components and building response and problems caused by the lack of lumped parameter tools that include internal 2D, 3D FEM capabilities. However, there are some limitations: (1) some specific solvers such as time-dependent k-epsilon turbulence solvers are not available yet. This means that, for example, time dependent 3D airflow around buildings cannot be solved; (2) although it is possible to construct a full 3D integrated HAM model of the indoor air and all constructions in FemLab, the simulation duration time would probably be far too long to be of any practical use; (3) the modeling of radiation is just recently available in FemLab and is not included in this research. Possible drawbacks are: (1) the software package MatLab itself and basic knowledge of Matlab are required to use the models; (2) at this moment HAMLab is research tool. Therefore it lacks facilities for design-oriented users such as user-friendly interfaces and user guides. (3) although state-of-art solvers are present, the simulation of FEM based integrated models can easily become very computation time consuming. The main benefits of HAMLab are: (1) it takes advantage of the facilities of the well maintained Matlab/SimuLink and FemLab simulation environment such as the state-of-art ODE/PDE solvers, controllers library, graphical capabilities etc; (2) all presented HAMLab models in this paper are public domain; (3) although not explicitly shown in this paper, compared to other HAM models, it is relative easy to integrate new models that are based on ODEs and/or PDEs. (4) the simulation facilitates open source modeling and if desired, models can be compiled into stand-alone applications. Current projects At this moment we are working on: (1) implementation of web based models for designers using the Matlab web server toolbox, to meet drawbacks 1 and 2; (2) speed up of computation duration times by the use of system identification modeling approximations of complex models, to meet drawback 3. (3) development of a full 2D integrated HAM model for indoor air, building constructions and outdoor wind profiles & rain distributions. This project is a pilot study for a full 3D model within the framework of the IEA Annex 41. (4) An international version of the HAMcases website.


P18, Page 14 REFERENCES ASHRAE 2001. Standard method of test for the evaluation of building energy analysis computer programs, standard 140-2001. Clarke J.A. 2001. Domain integration in building simulation, Building and Environment 33, pp. 303-308 Crawley, D.B., Hand, J.W. Kummert, M. & Griffith B. 2005. Contrasting the capabilities of building performance simulation programs, 9TH IBPSA Conference Montreal pp 231- 238 Felsmann C., Knabe G., Werdin H. 2002. Simulation of domestic heating systems by integration of TRNSYS in a MATLAB/SIMULINK model, Proc. 6TH Int. Conference System Simulation in Buildings Liege, December 16-18 Gouda M.M., Underwood C.P., Danaher S. 2003. Modelling the robustness properties of HVAC plant under feedback control, Building Serv. Eng. Res. Techn. 24,4 pp. 271-280 Gough M.A. 1999. A review of new techniques in building energy and environmental modelling, BRE Contract No. BREA-42. Garston, Watford, UK: Building Research Establishment. Hagentoft C.E. and 13 other authors. 2002. HAMSTAD – Final Report,, Report R-02:8, Chalmers Univ. of Tech., Sweden IEA Annex 41 2005. http://www.ecbcs.org/annexes/annex41.htm Kunzel H.M., 1996. IEA Annex 24 HAMTIE, Final Report – Task 1 Neuhaus, E. & Schellen, H.L., 2006, Onderzoek hygrostatisch geregeld stoken GEOhuis (in Dutch), BPS-report 06.02.K, Eindhoven University of Technology. Rieder P. MatLab/SimuLink for building and HVAC simulation – state of art. 9TH Int. IBPSA Conference Montreal 2005, pp1019-1026 Sasic Kalagasidis, HAM-Tools; An Integrated Simulation Tool for Heat, Air and Moisture Transfer Analyses in Building Physics, PhD thesis, Chalmers University of Technology 2004. Sahlin P. Eriksson L., Grozman P., Johnson H., Shapovalov A. Vuolle M. 2004. Whole-building simulation with symbolic DAE equations and general purpose solvers. Building and Environment 39, pp. 949-958. Schellen, H.L. 2002, Heating Monumental Churches, Indoor Climate and Preservation of Cultural heritage; PhD thesis, Eindhoven University of Technology. Schijndel, A.W.M. van & Wit, M.H. de, 1999, A building physics toolbox in MatLab, 7TH Symposium on Building Physics in the Nordic Countries Goteborg, pp81-88 Schijndel A.W.M. van 2003a. Modeling and solving building physics problems with FemLab, Building and Environment 38, pp 319-327 Schijndel A.W.M. van, Schellen, H.L., Neilen D. 2003b. Optimal setpoint operation of the climate control of a monumental church, 2ND international conf. on research in building physics, Leuven, Sept. 14-18 2003, pp. 777-784 Schijndel A.W.M. van, Wit M.H. de 2003c. Advanced simulation of building systems and control with SimuLink, 8TH IBPSA Conference Eindhoven. pp. 1185-1192 Schijndel, A.W.M. van 2003d. Integrated building physics simulations with FemLab/SimuLink /Matlab, 8TH Int. IBPSA Conference, Eindhoven, pp. 1177-1184 Schijndel A.W.M. van 2004. Advanced use of HAMLab on CE1, Annex 41 preliminary paper TUE Oct 2004 Paper A41-T1-NL-04-5, available in 2008. 7th International Conference on System Simulation in Buildings, Liège, December 11-13, 2006

P18, Page 15 Schijndel A.W.M. van 2005a. Application of HAMLab on Subtask 3, CE 1, Annex 41 preliminary paper TUE Oct 2004 Paper A41-T1-NL-05-1, available in 2008. Schijndel A.W.M. van, Schellen H.L. 2005b. Application of an integrated indoor climate & HVAC model for the indoor climate design of a museum, submitted for the 7TH symposium on building physics in the Nordic Countries, 13-15 June 2005, Iceland Schwab M. and Simonson C.J. 2004. Review of building energy simulation tools that include moisture storage in building materials and HVAC systems, Annex 41 preliminary paper A41-T2-C-04-6, available in 2008. Sinha S.L. 2000. Numerical simulation of two-dimensional room air flow with and without buoyancy, Energy and Buildings 32 pp121-129 U.S. Department of Energy, 2005 http://www.eere.energy.gov/ Wit M.H. de ,H.H. Driessen 1988, ELAN A Computer Model for Building Energy Design. Building and Environment, Vol.23,4, pp. 285-289 Wit, M.H. de, 2006, HAMBase, Heat, Air and Moisture Model for Building and Systems Evaluation, Bouwstenen 100, ISBN 90-6814-601-7 Eindhoven University of Technology SIMULATION TOOLS BSIM 2006. Energyplus2006. FemLab 2006. Fluent 2006. HAMLab 2006. HAMcases 2006. IBPT 2006. IDA ICE 2006. Matlab 2006. Power Domus 2006. SIMBAD 2006. TRNSYS 2006.

http://www.bsim.dk/ http://www.eere.energy.gov/ http://www.comsol.com/ http://www.fluent.com/ http://sts.bwk.tue.nl/hamlab/ http://sts.bwk.tue.nl/hamcases http://www.ibpt.org/ http://www.equa.se/ice/intro.html http://www.mathworks.com/ http://www.powerdomus.com/ http://software.cstb.fr/ http://sel.me.wisc.edu/trnsys/


DISCUSSION Paper 18 : Jos van Schijndel • Questions from : Henk Peitsman, TNO Built Environment – The Netherlands How did you taken into account with 1. the natural air infiltration and 2. the effect of the people on sunday in the services ? Answers : 1. We measured the CO2 decay which was used to estimate the infiltration. 2. We estimated the number of people present and calculated the human induced heat and moisture sources. • Questions from : Jean Lebrun, University of Liège – Belgium About validation : 1. Why not using humidity ratio in place of relative humidity (for easier comparison) ? 2. Why not comparing temperature for giving heating power (easier than the contrary) ? Answers : 1 & 2 : The validation data of the IEA-Annex 41 were provided by so-called exercises which could not be influenced by us. As modellers we asked similar questions to the developers of the validation exercise : a) what about the temperature distribution in the room ? b) what about the mass of the building ? c) what about measurement errors ? However, the developers of the exercise have not responded properly yet. The exercise is still running. In 2008 final exercises are expected.

COMMENTS AND QUESTIONS : GENERAL DISCUSSION Questions of Arthur Dexter (Chairman) – University of Oxford – U.K. KEY ISSUES ? • • • •

How detailed do models need to be ? How can we be sure that the model is accurate enough in all of the situations in which it will be used ? How can we measure the quality (completeness?) of the training data we use to calibrate/identify the model ? How do we design simulations that can be run in a reasonable time ?

• Question from Jean Lebrun to Jim Braun on the question of A. Dexter / Key issues : What about using the "simplified models" without any training ? Would it be good enough, after having estimated the parameters on the basis of physical data and, possibly tuning it better thanks to reference simulation ? Answer of Jim Braun Yes, I think it would be good enough. However the time, expertise and effort required to collect the detailed physical characteristics of the building and create a forward model is large. I think it would be more cost effective to train a simple model if the goal is to determine a control strategy. An inverse model also has the potential of being more accurate and could adapt to changes in the building. Comments from Louis Laret on the question of A. Dexter / Key issues : Choice of models and models use I suggest a quite different approach based on problems to be solved and how to find the best answer. From this point of view, models are tools or means to bring information to solve the problem. There are no "absolute" model and there will be never able to address all possible issues – But if different models could be used (simple, detailed…), it should be possible to collect enough information for a good answer. Also the data input is very time consuming, more than time computation. Today, people are isolated with the model they know well, without true comparisons. An attempt to progress in this way, should be to define a common data structure connected with different simulation tools. As soon as enough data are input, the different tools could be used, compared,…. A trial in this direction was the objective of a "90" European project : "COMBINE". This trial should be started again.

Comments from Fritz Schmidt on the question of A. Dexter / Key issues : I feel models are used to give answers. Then a model is adequate if it gives the answer in the desired accuracy. If we have "a detailed model available" which is able to answer a broad range of questions, we may use them also to get simple answers, if this is more efficient than developing a specific model. But never use a model to answer questions which it is not designed for. Comments from Madjid Madjidi on the question of A. Dexter / Key issues : We can attack the energy saving market with very simple models, so we need always very good reasons (proves), if going into more complicated models. Answer : I agree! Comments from John Mitchell on the question of A. Dexter / Key issues : Simple models based on solid physical principles are needed for extrapolation into unknown areas. ASHRAE is embarking developing guides for constructing buildings that use 50 % less energy, than buildings designed to current building standards. The guides will be based on simulations using generic building and HVAC models with systems that have not been extensively experimental by studied. For example, dedicated outdoor air systems are promising but little data on their performance are available. So, we need simple models to help sort through and select promising configurations and equipment, and to make solid estimates of their annual energy use.