Study On Elite Genetic Types Based On Genetic Algorithm For Building Physical Performance Optimization

Under the rapid pace of China's modernization,with the increase of building energy consumption,the demand for building energy-saving and emission reduction is also increasing.Due to the complexity of architectural design,traditional design methods can no longer meet the current status of multi-demand of contemporary architecture,and its qualitative analysis and discrete evolutionary process cannot accurately meet the optimal requirements for building physical performance,such as building energy consumption.Based on the parametric design,this paper combines genetic algorithms as the support of evolutionary algorithms,and combines professional computer software such as building modeling,physical performance simulation,and data statistical analysis.With the iSIGHT's optimization calculation framework as a platform,a building optimization design platform driven by physical performance based on genetic algorithm is built,the design method of the physical performance optimization driven by the original scheme for the indefinite target is completed.In the early stage of optimization design,in order to reduce the number of design variables in the scheme,leave only the variables that are more relevant and more sensitive to the optimization goal,also in order to save the optimization cost,orthogonal tables is used for orthogonal test analysis,and more relevant design variables is determined.The research object of this paper is the office building located in Wuhan City,Hubei Province.Applying the above-mentioned optimization design platform,the single-objective optimization analysis with thermal energy minimization as the research object is first performed.As for the result of the completed optimization of the design variables values,for the objective function,take a solution that is superior in performance as an excellent solution set,by comparing the commonalities of various design variables between solution sets,we sum up the elite gene types that really make the physical performance of buildings outstanding,and the corresponding design plan is gained.At the same time,a multi-objective optimization analysis was conducted with the objective functions of the minimization of thermal energy consumption throughout thebuilding,minimization of thermal discomfort within the building,and maximization of daylight factor throughout the year,the elite gene types and the design plan were also gained.Finally,based on the above-mentioned gene types and the optimization of architectural physical properties,the design strategies for different volumes and different forms of office buildings in hot summer and cold winter regions were summarized,providing designers with appropriate guidance and assistance.Based on the optimization of architectural physics performance,this paper summarizes a set of effective design techniques based on optimized design platform and the cooperation of multiple disciplines.The design platform does not provide a single optimal solution.Instead,it integrates the common range of elastic and inclusive design variables in the excellent solution set.In the actual architectural design case,a value range of design variables that has latitude and can ensure excellent architectural physical performance often has more reference value and operation space.This kind of optimization design method that focuses on the good solution set rather than the optimal solution also provides a new direction for the study of architectural optimization design.