Between past and future: Daylight simulation and analysis for today

Tags: simulation, International Conference, The Netherlands, daylight simulation, daylight, A. Koutamanis, building regulations, lighting fixtures, building performance, digital building models, performance analysis, Eindhoven, daylight performance, A. Koutamanis Delft University of Technology Berlageweg, Building Structures, digital model, radiosity calculation method, simulation environment, geometric modelling, calculation method, surface reflectance, daylight conditions, calculation, Dutch Government Buildings Agency, modelling, Royal Library
Content: Adaptables2006, TU/e, InterNational Conference On Adaptable Building Structures 4-329 Eindhoven [The Netherlands] 03-05 July 2006 Between past and future: daylight simulation and analysis for today A.M.J. Post, A. Koutamanis Delft University of Technology Berlageweg 1, 1016 BP Delft, The Netherlands [email protected] KEYWORDS simulation; building performance; daylight; integration 1 Introduction In the past few decades, the building industry has been incorporating the efficiency delivered by the personal computer. Especially in the last decade, workflow in architectural and engineering firms has become digitized, leading to higher efficiency and transparent data management. However, other aspects of the electronic revolution have yet to live up to their promise (Koutamanis, 2000). These include issues relating to building performance that can be analysed in 3D virtual building prototypes. Comprehensive 3D digital building models (collaborative or not) are generally considered as the Holy Grail in the field of collaborative prototyping (Yeomans et al., 2006). At the same time, however, the implementation of 3D digital building models is scarce, due to technical, operational and process limitations. We propose that existing technologies are generally sufficient for developing highperformance partial solutions, i.e. solutions focused on specific design aspects. This can support the simultaneous development of working methods for less traditional applications of CAAD (De Groot and Paule, 2002). One such aspect is daylighting. Most building professionals will agree that the daylight performance of buildings is a relevant and crucial issue.(Leslie, 2003) At the same time, both building regulations and practice have not changed significantly during the last few decades. By using a recent Case Study, this paper investigates the possibilities of the combination of design computing and daylight performance analysis. In general there are two main methods for the analysis of daylight performance: physical scale models and full-size digital models. However, with the advance of computational speed, the acceptance of Digital Design tools, the flexibility of geometric, surface and sky models and its potential reliability (estimated at ±10%) computer-based daylight simulation is the rising star (Mardaljevic, 2000). Research has shown that measurement in scale models, long considered to be accurate, can lead to large errors (60-200%) (Nair et al., 1997). 2 Simulating daylight It is widely acknowledged that there are currently two tools (and underlying approaches) that qualify for daylight simulation: Radiance and Lightscape. Although a case can be made for using Lightscape within its particular strengths (Ng and Chan, 2003), we consider Radiance to be superior in daylight
Adaptables2006, TU/e, International Conference On Adaptable Building Structures 4-330 Eindhoven The Netherlands 03-05 July 2006 modelling and far superior in handling complex geometric models. Because of its extended backward raytracing calculation method, Radiance requires no specific modelling strategy, while Lightscape poses severe geometric restrictions originating in its radiosity calculation method. Furthermore, the reliability of Radiance simulation has been proved by several validation studies (Mardaljevic, 2000). Nevertheless, the usability of Radiance suffers from serious drawbacks, especially in the geometric modelling, texturing and surface smoothing capabilities. To cope with this, we have used an in-house translator from the modelling/animation system 3ds Max to Radiance and additions to the Radiance core so as to provide a flexible, reliable and usable simulation environment. Aspects of geometry, surface reflectance and transmittance, and sky modelling need to be defined with great care (Lam et al., 1997). When handled with due care, Radiance provides the best of both worlds: visually compelling imagery and reliable lighting calculation (Ward and Shakespeare, 1998). 3 Case Study: Expansion of RoyaL LIBRARY, The Hague In the spring of 2003 the board of directors of The Royal Library in The Hague was concerned with the combination of daylight and artificial light in a planned addition, which would accommodate a variety of functions, ranging from circulation spaces to an exposition area for delicate objects. The owner of the building (the Dutch Government Buildings Agency) commissioned the development a digital model for simulation. The main aspect of the simulation would be the visual quality of the spaces involved under varying lighting conditions. Acquiring the required data for meaningful simulation was a task of varying difficulty. As is common in architectural design, no 3D digital model was available during the design process. Based on AutoCAD drawings from the architect's office, the construction of the 3D model proved fairly straightforward (Figure 1). Figure 1. Exporting geometry, calculation setup and viewpoints from 3ds Max Color and (diffuse) reflectance were based on colour description in the RAL-Digital 3.0 software from the `Deutsches Institut fur Gutesichering und Kennzeichnung E.V'. For a number of general lighting fixtures, the luminance data were available as photometric webs in the IES format. The remaining light sources, including a custom designed lightwall, were individually defined (Figure 2). Between past and future: daylight simulation for today: A.M.J. Post, A. Koutamanis
Adaptables2006, TU/e, International Conference On Adaptable Building Structures 4-331 Eindhoven The Netherlands 03-05 July 2006 Figure 2. IES-defined lighting fixtures A number of simulations were conducted with Radiance, resulting in series of images depicting the variation in visual experience during the day. To account for the non-linearity and accommodation functions of the eye, specific corrections were made to the datasets using pcond.exe, a utility that comes with Radiance (Figure 3). Figure 3. Varying daylight conditions combined with artificial light (eye-corrected) Other simulated design variations included the introduction of sun screens on several window glazing in the south and west facades (Figure 4). Figure 4. Sun blocking foils on first floor Furthermore, the central colour-varying lightwall provided great changes in atmosphere, reflected in the colour-bleeding effects on nearby surfaces (Figure 5). Figure 5. Custom lighting with changing colours The process of 3D modelling, simulation and visualization were not performed in isolation but in conjunction with the ongoing design and decision-taking. The (intermediate) results of each stage were fed back to the key actors of these processes, allowing architects and consultants to make Between past and future: daylight simulation for today: A.M.J. Post, A. Koutamanis
Adaptables2006, TU/e, International Conference On Adaptable Building Structures 4-332 Eindhoven The Netherlands 03-05 July 2006 constructive use of the images produced (and their implied reliability) in designing and communication. 4 Evaluation of case Following several discussions about the relevance of trying to predict the future, a common quantative subject emerged: the illuminance levels (in lux) to be expected under daylight conditions. Digital simulation was responsible for arriving at this, because it permitted the transparent combination of artistic and engineering elements, i.e. attractive photorealistic imagery and objective numerical data. Analytical representations of the simulation results (Figure 6) are quite reliable in computational terms. However, the project definition did not include any description of the necessary geometric, reflection or lighting conditions for the evaluation of the analysis. With respect to the relevance of combining pictures with numbers, the project would have benefiteed from additional constraints, which would have allowed a higher level of performance. Figure 6. Daylight Factor distribution on floor levels of Royal Library As in most cases, light simulation stimulates the formulation of relevant and interesting questions that can be answered in a practical and verifiable way. This does not imply revolutionary changes in the design process but a rationalization and optimization of design procedures. Architects do not necessarily need to be instructed in matters of light. Transparent feedback and the ability to create alternatives and variations in an efficient and reliable manner frequently suffice (Glaser et al., 2004). 5 Conclusions and recommendations In the total lifecycle of a building, from project idea up to demolition, there are several questions concerning the flexibility of the process decision chain. In conjunction with providing solutions for worst case scenarios in the design stage, we should focus on development of structural methods for optimizing building performance throughout the building's entire lifecycle. We suggest a reconsideration of the processes that define the decision chain: 1. Development of a workable knowledge base of existing (daylight) design solutions in building practice There are many buildings that demonstrate not only understanding and intelligent use of daylight with respect to building quality and performance but also truly innovative approaches and products. This perfromance can be measured and form the basis of future design choices. 2. Creation of specific, adaptable design brief criteria by means of intelligent best practice selection In many run-of-the-mill projects (e.g. housing, office buildings) there is insufficient time and attention for finer points and designers exhibit the unfortunate tendency to revert to stereotypes that may perform poorly. Designs briefs can improve by using the accumulated knowledge of 5.1 Between past and future: daylight simulation for today: A.M.J. Post, A. Koutamanis
Adaptables2006, TU/e, International Conference On Adaptable Building Structures 4-333 Eindhoven The Netherlands 03-05 July 2006 3. Development of understandable design proposal analysis representation of design proposals In order to be able to compare designs we need global measures of the quality of daylight performance. These include visual comfort issues like glare, functional zoning, light level variance and speed, and make use of general indices like the Daylight Factor and Daylight Autonomy ratings. 4. Construct (re)presentation policies that will provide combined access to knowledge base and design proposal analysis Computer Simulation, especially if coupled to extensive collections of well-documented cases and precedents, provides the means for effective, efficient and reliable specification, analysis and synthesis. Moreover, intelligent computer simulation can be an unobtrusive, supportive activity in the background of design activities that both identifies potential pitfalls and allows deferment of the solution on the basis of informed opinions (as opposed to unfounded guesses). 5. Reconsideration of building regulation possibilities Good daylight design requires more accurate specifications, awareness of lifecycle aspects and attention to performance, as well as a closer interest in what research has to offer on the methodical and practical levels. Daylight requirements should be defined in terms of daylight performance, not in rules of thumb or general principles. 6 References DE GROOT, E. & PAULE, B. (2002) DIAL-Europe: New Functionality s for an Integrated Daylighting Design Tool. Timmermans, Harry (Ed.), Sixth Design and Decision Support Systems in Architecture and urban planning - Part one: Architecture Proceedings Avegoor, the Netherlands), 2002. GLASER, D. C., FENG, O., VOUNG, J. & XIAO, L. (2004) Towards an algebra for lighting simulation. Building and Environment. KOUTAMANIS, A. (2000) Digital architectural visualization. Automation in Construction, 9, 347. LAM, K. P., MAHDAVI, A. & PAL, V. (1997) Algorithm and Context: A Case Study of Reliability in Computational Daylight Modeling. CAAD Futures 1997 [Conference Proceedings / ISBN 0-7923-4726-9] Mьnchen (Germany), 4-6 August 1997, pp. 331-344. LESLIE, R. P. (2003) Capturing the daylight dividend in buildings: why and how? Building and Environment. MARDALJEVIC, J. (2000) Daylight Simulation: Validation, Sky Models and Daylight Coefficients. De Montfort University, Leicester, UK. NAIR, G., MCNAIR, D. G. & DITTON, J. (1997) Simple scale models for daylighting design: Analysis of sources of error in illuminance prediction. Lighting Research and Technology an International Journal, 29, 143-150. NG, E. & CHAN, T. Y. (2003) Computational simulation based daylight design for urban sites validation, methodology and legality. Digital Design [21th eCAADe Conference Proceedings / ISBN 0-9541183-1-6] Graz (Austria) 17-20 September 2003, pp. 91-98. WARD, G. & SHAKESPEARE, R. (1998) Rendering with Radiance: The Art and Science of Lighting Visualization. YEOMANS, S. G., BOUCHLAGHEM, N. M. & EL-HAMALAWI, A. (2006) An evaluation of current collaborative prototyping practices within the AEC industry. Automation in Construction, 15, 139. Between past and future: daylight simulation for today: A.M.J. Post, A. Koutamanis

File: between-past-and-future-daylight-simulation-and-analysis-for.pdf
Title: Microsoft Word - Titelpagina volume 1.doc
Author: De Louw
Published: Tue Feb 24 11:18:38 2009
Pages: 5
File size: 1.43 Mb

, pages, 0 Mb

Modern control systems, 7 pages, 0.43 Mb

Volume 1 Issue, 12 pages, 0.91 Mb

, pages, 0 Mb
Copyright © 2018