Multi-objective optimization of building's life cycle performance in early design stages

SUSTAINABLE BUILT ENVIRONMENT D-A-CH CONFERENCE 2019

生态环境科学

Yu, H.^1 ; Yang, W.^1 ; Li, Q.^1

School of Architecture, Tianjin University, Tianjin

300072, China^1

关键词: Building performance;    Early design stages;    Life cycle inventory datum;    Life-cycle environmental impact;    Life-cycle performance;    Operating energy consumption;    Optimal solution sets;    Parametric approach;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/323/1/012116/pdf
DOI  :  10.1088/1755-1315/323/1/012116

学科分类：环境科学（综合）

来源: IOP

【 摘 要 】

The early design stages of buildings have the highest potential for optimizing energy efficiency. From the aspect of architectural design, the architects can hardly make the best decisions of the complexed variables of geometries and materials. The choice and weighting of different optimization objectives also have no reference while considering the impacts from the whole view of the life cycle. In this paper, a simulation-based parametric approach through the use of multi-objective optimization method is framed. The extraction of geometry variables from the original design and the principles of explaining optimization results obtained from different objective combinations are discussed. This process is carried out in Rhino and Grasshopper. To fully consider the aesthetic and building performance step by step in early design stages, the geometry and material parameters are tested simultaneously. The life cycle inventory data of materials and energy resources are imported by programming, and local data is partly used. Life cycle environmental impacts (PED and GWP), life cycle cost, and operating energy consumption are used as optimization objectives. A case study is verified on a project of a small campus museum building in northern China. The results indicate that the combination of geometry and material values in the optimal solution set is various and needs to be artificially selected according to actual needs. The optimization objectives should be considered comprehensively; an incomplete set may result in poorly behavior of the unselected objectives.