Future climate adaptive building shells 'Optimizing energy and comfort by inverse modeling'.

Transcription

1 Boer, B. d., Bakker, L., Oeffelen, E. C. M. v., Loonen, R. C. G. M., Costola, D., & Hensen, J. L. M., Future climate adaptive building shells 'Optimizing energy and comfort by inverse modeling'. Proceedings of the 8th Energy Forum on Solar Building Skins, 6-7 December, Bressanone, Italy, pp Future climate adaptive building shells optimizing energy and comfort by inverse modelling Bart de Boer, Leo Bakker, Linda van Oeffelen1 Roel Loonen, Daniel Cóstola, Jan Hensen2 1 2 TNO, Delft, Netherlands, b.deboer@tno.nl TU/e, Eindhoven, Netherlands, r.c.g.m.loonen@tue.nl Abstract The traditional design process of building shells leads even today to substantially static systems. Building shell properties, like insulation value, thermal mass and window area, are kept constant throughout the year for the majority of buildings. As a result, buildings do not perform optimally and large energy consuming HVAC installations are used to compensate for the poor performance of the building itself. The assumption of the FACET project (Dutch acronym for: Adaptive façade technology for increased comfort and lower energy use in the future ) is that buildings with climate adaptive thermal and daylight properties will have a much better energy performance while maintaining a high comfort standard. The development of climate adaptive buildings shells (CABS), based on theoretically ideal adaptive properties, enables maximizing indoor comfort and minimizing energy use for heating, cooling, ventilation and lighting. This enables the realisation of nearly zero energy, or even energy producing buildings in the near future. 1. Optimizing energy and comfort by an inverse approach To explore the potential for climate adaptive building shells, the question is not just How do dynamic building shells perform using dynamic simulations? Instead, a step forward is made by asking the question: What would be the ideal dynamic properties of a building shell to get the desired indoor climate at variable outdoor climate conditions?. This so-called inverse approach provides a set of theoretical ideal building shell performance parameters, within a fairly realistic bandwidth, for different climate conditions at various time cycles (seasons, day-night, hours). The idea of an inverse approach can be translated as turning around the order input => model/simulation => output as depicted in the figure below. Modelling Simulating Inverse approach Figure 1: Schematised representation of the inverse approach 15

2 In contrast to the normal simulation work, the input at the inverse simulation approach is an unknown variable (=?) with a desired known output (=!). In this case, dynamic instead of static building properties (such as U value, solar transmittance, ventilation rate, etc.) are needed. The 'ideal shell' is simulated as a black box using dynamic simulation programs. By defining the desired output, the inverse simulations will calculate which building shell properties (within an acceptable range of values) are needed to stay within the defined comfort zone at the lowest energy use. Materials and technologies to reach this ideal behaviour can be future-oriented but they are also in part already available, in either low or high technical solutions, such as smart glazing, variable vacuum insulation, insulating shading elements, etc. Further technology and/or system development is desired to fully meet the requirements. In this paper we firstly present the simulation-based methodology for the inverse approach. After that, we discuss the set-up and findings of our user acceptance experiments with respect to dynamic façades. The paper finishes with providing insights into the potential of CABS and the future outlook of our project. 2. Defining the ideal dynamic thermal and visual properties of the building envelope The building envelope is the place where indoor and outdoor environments meet each other, and as such it forms the interface for multiple physical interactions. In order to get well-informed building performance data, it is important to evaluate these impacts in an integrated way. In earlier publications, however, we reported on results of the inverse approach for CABS in respect to the thermal and visual aspects separately [1,2]. The use of this initial approach was mainly imposed by a lack of capabilities in existing building performance simulation software. With currently existing tools, it is not possible to predict visual, thermal and energy performance while considering controllable building shell properties under the influence of optimization-based operational strategies [3]. To overcome this limitation, we developed a co-simulation strategy to fulfill all the requirements of the integrated inverse approach. Figure 2: Overview of the simulation strategy This new approach makes use of a virtual test facility for performance evaluation of advanced building control strategies [4], and uses the building controls virtual test bed (BCVTB) [5] to facilitate exchange of data during simulation runtime. The basic principles of the simulation strategy are illustrated in Figure 2. ESP-r is used as simulator for the thermal domain, and Radiance for the (day)light domain. Sequences of building shell behavior are automatically generated and evaluated on the basis of a multi-objective optimization algorithm. On the one hand, the algorithm aims to minimize the energy demand for space conditioning, while sustaining high levels of satisfaction with the thermal environment. At the same time, the algorithm 16

3 strives to maximize the use of daylight, while preventing the occurrence of glare and ensuring views outside. Figure 3: FACET approach with (a) simulation, (b) real building Figure 3(a) represents the basic idea of the simulation methodology. Figure 3(b) illustrates how this approach could look like in the real world. The external façade is subdivided in large number (e.g. 5 x 6) of elements. For each simulation time step in the finite prediction horizon of e.g. two days, the thermo physical and optical properties of each façade element are optimized. The optimization procedure is repeated on a receding horizon basis. The end result is a sequence of adaptable façade elements, including window-to-wall ratio, transparency, thermal resistance, albedo values, etc. Figure 4 shows an example of the simulation result for optimal building shell transparency and thermal resistance over time. Figure 4: Simulation result showing optimal transparency and thermal resistance over a one week period 3. User acceptance of dynamic façades Adaptive, dynamic façades imply that movements within the façade will occur. Especially adapting the façade to optimal indoor visual comfort, conditions will lead to movements within the façade during the day. These movements might disturb occupants due to the changes in daylight levels and daylight distribution indoors, changes in the view to outdoors, or just by the movements of the façade elements itself, or simply the noise induced to realize these changes. It is crucial that the changes in the façade do not cause substantial discomfort and thus spoiling the aimed CABS benefits. 17

4 Figure 5: (a) Shading mechanism, (b) façade in the test room plus user interface To gain insight into these complex issues regarding the user acceptance of adapting façades, a limited explorative research with 25 test subjects was executed. The following hypotheses were defined: 1. Changes within the façade will often lead to discomfort and disturbance. 2. Smaller movements will cause less discomfort and are more acceptable than less large movements (the total moved distance within the façade is the same). 3. Logical perceived changes (e.g. avoiding glare) are more acceptable than less logical perceived changes. 4. The possibility to overrule the changes in the façade will have a positive effect on the user acceptance. A test façade (figure 5) was constructed with eight motorized opaque shadings. The façade was controlled with fixed schedules (figure 6): (1) less large movements, (2) smaller movements and (3) a light sensor based automatic control to induce logical perceived changes. All schedules were executed with and without overrule option. Participants were asked to fill in a computerized questionnaire concerning perceived light quality, rate of disturbance, etc. Figure 6: Top position and bottom shading in time schedule 1 and 2 Preliminary results fortunately showed that changes in the façade did not lead to the expected increase in discomfort and disturbance. Furthermore, the large movements are less distracting than smaller movements. Automated control using a light sensor to induce logical perceived changes (compared to fixed schedules) shows decreased perceived glare, and increased distraction due to movements in the façade. If the overrule option is enabled (compared to not enabled) respondents are more content with their view to outdoors (on average 20%) and the perceived light quality (9%). The noise induced by the movements of the shading devices was not very disturbing to the test subjects (probably due to careful design). The results are now processed further statistically and will be introduced during the presentation of this paper. 18

5 4. Energy saving potential and future outlook The integral, inverse, thermal and daylight simulation approach of FACET shows a large energy saving potential, while maintaining a high level of comfort. In comparison to a standard new Dutch office building (EPC value of 1,1), a reduction of the cooling demand with a factor 3 and a heating demand with a factor of 6-10 is possible [2]. The method of superposition for visual optimization with pre-processed Radiance output shows an effective way to enable evolutionary optimizations. The results provide feedback on the defined demands. There are many variants with similar performance output. The method will be extended among other more working positions and light deflecting properties of the façade elements. In addition, full scale tests have been executed to get feedback on the user experience on FACET façades. These results stress the importance of taking into account control aspects during the design process of buildings with dynamic façades. The FACET project will continue until March 2013, and by this time more results can be expected in terms of fully integral inverse modelling results for offices, schools and dwellings. Furthermore proof of concept for different adaptive building shell concepts and possible opportunities and potential for development of new technologies are expected. A big challenge will be to translate the theoretical, desired behaviour into real world CABS concepts. 5. References [1] Boer, B.J. de, Ruijg, G.J., Loonen, R.C.G.M., Trčka,M., Hensen J.L.M., Kornaat W., Climate adaptive building shells for the future - optimization with an inverse modelling approach, ECEEE 2011 [2] Boer, B.J. de, Ruijg, G.J., Bakker L., Kornaat W., Kurvers, S., Alders, N., Raue, A., Hensen J.L.M., Loonen, R.C.G.M., Trcka M. Energy saving potential of climate adaptive building shells - Inverse modelling of optimal thermal and visual behaviour, Proceedings of the International Adaptive Architecture Conference, 3-5 March 2011, London, p [3] Loonen, R.C.G.M., Trcka, M., Costola, D., & Hensen, J.L.M "Performance simulation of climate adaptive building shells - Smart Energy Glass as a case study", Proceedings of the System Simulation in Buildings Conference SSB, 1315 December, Liege, Belgium, p. 19. [4] Hoes, P., Loonen, R.C.G.M., Trcka, M. & Hensen, J.L.M Performance prediction of advanced building controls in the design phase using ESP-r, BCVTB and Matlab. Proceedings of Building Simulation and Optimization 2012, Loughborough, UK. [5] Wetter, M Co-simulation of building energy and control systems with the Building Controls Virtual Test Bed, Journal of Building Performance Simulation, 4(3): Acknowledgements FACET is a Dutch EOS-LT R&D project, subsidised by Agentschap NL. Partners in the project are TNO, TUD, TU/e, CHRI and ECN. For more information please visit: 19