Genetik Algoritmalar Kullanılarak Güneş Işınımı ve Gölgeye Göre Optimal Yüksek Yapı Form Önerileri Üretilmesi

Abstract

Yapılarda ısıtma ve soğutma giderleri ve enerji tüketimi büyük bir yük oluşturmaktadır. Bunun yanında yüksek yapılar uzun ve büyük gölgeler yarattığı için çevrelerindeki yapıların da direkt gün ışığı aldığı saatleri kısıtlamakta ve bazen tamamen engellemektedirler. Yüksek yapılar bu iki özellik ile şehrin enerji tüketimini arttırmakta ve yaşam konforunu azaltmaktadır. Bu çalışmanın amacı yüksek yapıların konsept tasarım aşamasında evrimsel yaklaşımlar kullanılarak hedeflenen toplam yapı alanını çok fazla azaltmadan belirli kriterleri optimum düzeyde sağlamak ve binanın kendisi ile çevresinde enerji tüketimini azaltmak ve yaşam kalitesini arttırmaktır. Tasarımcının genetik algoritmaları kullanarak bu koşulları sağlayan farklı çözüm önerilerini konsept aşamasında görebilmesi ve müdahale edebilmesi önemli bir özelliktir. Genetik algoritmalar ile kentsel doku ve özelliklerine odaklanan çalışmalar yanında yapının kendisine ve yapı kabuğuna odaklanan çalışmalar da bulunmaktadır. Bu çalışma yapının kütlesine ve kentsel dokuya etkisine odaklanmaktadır. Bu önerileri test edebilmek için örnek bir mevcut yüksek yapı ve onun evrimsel algoritmalarla üretilmiş farklı alternatifleri karşılaştırılmıştır. Örnek yapı olarak Chicago’daki Willis Tower (eski ve bilinen adı Sears Tower) seçilmiştir. Çalışmanın yöntemi ilk aşamada Sears Tower’ın mevcut tasarımının toplam kat alanını, belirli bir periyotta aldığı güneş ışınımını, yakın çevredeki binalara yaptığı gölgeyi ve binanın farklı bölgelerinden gökyüzü görünümünü hesaplamaktır. İkinci aşamada ise bu dört uygunluk kriterine göre üretilen alternatifler arasından seçimler yapmaktır. Bu analizleri yapabilmek için Rhinoceros 3D programı ve Grasshopper eklentisinin alt eklentileri kullanılmıştır. Elk eklentisi Chicago ile ilgili şehir verileri ve modelleme için, Ladybug eklentisi güneş ve gölge analizleri için, Wallacei eklentisi de genetik algoritmalar ile simülasyon ve analizler yapmak için kullanılmıştır. Çalışmanın bu aşamasında malzemeler ve yapı elemanları kapsam dışıdır, analizler kütle çalışmaları üzerinden yapılmaktadır.

Keywords

• Genetik Algoritmalar
• Yüksek Yapılar
• Optimizasyon
• Güneş Işınımı
• Gölge

Generating Optimal High-Rise Building Suggestions According to Solar Radiation and Shade Using Genetic Algorithms

Abstract

Heating and cooling costs and energy consumption in buildings constitute a great burden. Especially the energy spent for cooling in summer is more than the energy spent for heating in winter. In high-rise buildings, these expenditures increase in order to reach thermal comfort. Since the buildings are high, they are exposed to harsher climatic conditions. In addition, because high-rise buildings create long and large shadows, they limit and sometimes completely prevent the hours of direct sunlight for the surrounding buildings. With these two features, high-rise buildings increase the energy consumption of the city and reduce the comfort of life. High-rise buildings also try to create as much area as possible, as they have a specific cost and purpose. The aim of this study is to provide certain criteria at the optimum level without reducing the targeted total building area too much by using evolutionary approaches in the concept design phase of high-rise buildings, to reduce energy consumption in the building itself and its surroundings, and to increase the quality of life. It is an important feature that the designer can see and intervene in the concept phase of different solution proposals that meet these conditions by using genetic algorithms. Numerous studies have been conducted on form and mass in architectural design using evolutionary approaches and genetic algorithms, which are a subset of this approach. With the development of computers and therefore computational possibilities, the number and level of detail of these studies have increased and the computation time has shortened. In this study, Wallacei plugin, which is an evolutionary approach plugin using NSGA-II / Non-Dominated Sorting Genetic Algorithm II, was used. As plug-ins related to solar and energy analysis such as Ladybug have developed, this type of data has begun to be included in the studies. In addition to the studies focusing on the urban texture and features with genetic algorithms, there are also studies focusing on the building itself and the building envelope. This study focuses on the mass of the building and its effect on the urban fabric. In order to test these suggestions, a sample existing high-rise building and its different alternatives produced by evolutionary algorithms were compared. The Willis Tower (formerly Sears Tower) in Chicago was chosen as an example building. The simple but effective mass form of the structure provides advantages and diversity in analysis and calculations. It is located in a climatic region like Chicago where all four seasons are experienced, and its effects on its surroundings can be easily observed. The method of the study is to calculate the total floor area of the current design of Sears Tower, the solar radiation it receives in a certain period, the shadow it casts on the nearby buildings and the sky view from different parts of the building. The second step is to make choices among the alternatives produced according to these four eligibility criteria. Rhinoceros 3D program and sub-plugins of Grasshopper plugin were used to make these analyzes. The Elk plugin was used for city data and modeling related to Chicago, the Ladybug plugin was used for sun and shadow analysis, and the Wallacei plugin was used for simulation and analysis with genetic algorithms. At this stage of the study, materials and structural elements are out of scope, analyzes are made over mass studies.

Keywords

• Genetic Algorithms
• High-Rise Buildings
• Optimization
• Solar Radiation
• Shade

References

1. Bäck, T. (1996). Evolutionary Algorithms in Theory and Practice. Oxford University Press.
2. Choi, J., Nguyen, P. C. T. and Makki, M. (2020). The design of social and cultural orientated urban tissues through evolutionary processes. International Journal of Architectural Computing, SAGE Publications, pp. 1–29.
3. Deb, K., Pratap, A., Agarwal, S. and Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, vol. 6, no. 2, pp. 182-197.
4. Holland, J. H. (1975). Adaptation in natural and artificial systems: An introductory analysis with applications to biology, control, and artificial intelligence. U Michigan Press.
5. Koenig, R., Yufan, M., Knecht, K., Aichinger, A., & Konieva, K. (2020). Integrating urban analysis, generative design, and evolutionary optimization for solving urban design problems. Environment and Planning B: Urban Analytics and City Science, 47(6): 997–1013.
6. Makki, M. and Showkatbakhsh, M. (2018). Control of Morphological Variation Through Population Based Fitness Criteria. Learning, Adapting and Prototyping, Proceedings of the 23rd CAADRIA Conference, Beijing, China, The Association for Computer-Aided Architectural Design Research in Asia (CAADRIA), Hong Kong, vol. 1, pp. 153–162.
7. Makki, M., Showkatbakhsh, M., Tabony, A. and Weinstock, M. (2018). Evolutionary Algorithms for Generating Urban Morphology: Variations and Multiple Objectives. International Journal of Architectural Computing, vol. 17, no. 1, pp. 5–35.
8. Miao, Y., Koenig, R., & Knecht, K. (2020). The Development of Optimization Methods in Generative Urban Design: A Review. SimAUD: Symposium on Simulation for Architecture & Urban Design, 247–254, Vienna, Austria.

9. Mitchell, M. (1998). An Introduction to Genetic Algorithms. MIT Press.
10. Navarro, D., Makki, M. and Bermejo, A. (2018). Urban-Tissue Optimization through Evolutionary Computation. Mathematics, vol. 6, no. Special Issue: Evolutionary Computation, pp. 1–16.
11. Petrov, M. and Walker, J. (2020). Optioneering Methods for Optimization - Methods of exploring primary and secondary performance criteria in urban design. Proceedings of the 38th eCAADe Conference - Volume 1, TU Berlin, Berlin, Germany, 16-18 September 2020, pp. 29-36.
12. Randall, M. Kordrostami, T., Makki, M. (2020). The Taikoo Shing Superblock: Addressing urban stresses through sequential evolutionary simulations. D. Holzer, W. Nakapan, A. Globa, I. Koh (eds.), Proceedings of the 25th CAADRIA Conference - Volume 1, Online, pp. 415-424.
13. Showkatbakhsh, M. and Makki, M. (2020). Application of Homeostatic Principles within Evolutionary Design Processes: Adaptive Urban Tissues. Journal of Computational Design and Engineering, Oxford, vol. 7, no. 1, pp 1-17.