Computational Approach in Investigating Surface and Site Radiation in the Early Phase of Designing Two-Story Wooden House in Orio District, Kitakyushu, Japan

Abstract

Along with the rise of awareness in implementing environmentally-conscious urban design, the classical process in approaching urban environmental analysis should be shifted to performance-based considerations. In architecture and urban design, the tools developed recently hold promise to assist the decision-making process in the early phase of designing. Thus, the further performance of the intended architecture project, particularly related to energy consumption and its environmental impact, can be nearly accurately predicted. Although many studies contain the computer design process for analyzing urban energy performance, the Multi-Objective Optimization (MOO) process in the sub-tropical climate of Japan has not been widely used. This research investigates hypothetically the surface and site radiation in the early phase of designing a two-level wooden house in Orio District, Kitakyushu, Japan, through parametric and generative algorithms. The research was designed to map the best possible design solution of free-form loft-twisted structures through iterating design variables related to design elements such as building orientation, base radius, twisting and scaling factor, and the angle of the roof slope. The optimization targets the minimum quantity of solar radiation both for the site and the building surface affected by the geometry. The simulation uses an EPW file of Shimonoseki as the input weather file and is scheduled for fitting the period in the extreme hot week during the summertime. The findings are successful to produce a design with better performance compared to the benchmark model even though with insignificant improvement.

Keywords

• multi-objective optimization
• parametric design
• urban microclimate