The adjoint variable method (AVM) is a method for solving inverse problems, and is used for sensitivity analysis. In this study, the effective usage of inverse analysis based on AVM is considered for the purpose of its use as an approach to design. This paper proposes a method of inverse analysis to determine the optimal arrangement of partition walls in a station building to prevent inflow of wind in winter and to promote heat removal in summer. The analysis was applied to the design of Nobeoka station building, part of a redevelopment project being handled by Office of Kumiko Inui. Usage of inverse analysis based on AVM is proposed as a design approach for determining the optimal placement of partition walls in a semi-outdoor space. In the analysis, surface panels with an aperture ratio property are installed at candidate positions for the partition walls. Forward analysis (CFD) and inverse analysis (AVM) are alternated to compute the optimal aperture ratio. From the resulting aperture ratio, it was determined whether a partition walls was necessary at the candidate position.
The adjoint variable method (AVM) is a method for solving inverse problems, and is used for sensitivity analysis. This paper reports on a study considering the effective usage of AVM, and examines how AVM could be applied to achieve optimal arrangement of trees for the purpose of reducing wind velocity around a station building. The following conclusions were obtained; Preferential arrangement plan of trees around Nobeoka station for the purpose of wind velocity reduction is calculated by AVM. The results showed that in the case of 10 trees with optimal placement, wind velocity would be reduced by 18% compared to wind velocity without trees. In contrast, in the case of 44 trees with all placement, wind velocity would be reduced by 29%. Even trees number is less than 1/4, the effect of 60 ~ 70% was obtained. These results confirmed the validity of the proposed method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.