The Study of Multi-objective Genetic Algorithm applied to Computer Aided Building Opening Design

• Main Authors: Bo-Syun Chueh, 闕伯勳
• Other Authors: Hong-Ching Shih
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/3tpcc7

碩士 === 銘傳大學 === 媒體空間設計研究所碩士班 === 94 === The decision process of design is usually evaluated by different factors at the same time, this is called multi-objective optimization problem. When facing a multi-objective optimization problem, most of the designers seldom consider all the designing factors to do a systematic analysis during the designing process. It’s difficult to plan a complete solution due to omitting some messages, This research is trying to use Multi-objectives Genetic Algorithm which integrating Multi-objectives Optimization and Genetic Algorithm to build a Computer Aided Design system to test and verify if the algorithm can process the architectonic issues. By building a natural daylight planning system, the designing process needs to consider different factors which include allocation, natural daylight, and energy control during the designing process. The system can assist architect designers for planning and be an aided design tool. This research discusses when designing an individual residence, designers have to build a system which considering the regulation, natural light and design factors including size of the space, allocation of the space, location and size of the window. The research uses the experiment that a housewife’s activity needs in the house. The convergence of evolution trend proves the feasibility that Multi-objectives genetic algorithm can be applied to computer aided design. The process also considering furniture size, after manually adjust the solution, the system can match users’ needs. It shows the research can be reference.