Improving thermal performance of automatically generated floor plans using a geometric variable sequential optimization procedure

摘要: Abstract This paper presents an approach for the optimization of floor plan designs. These are generated using a hybrid evolutionary approach, which produces alternative designs according to user’s preferences and requirements. Once generated, algorithm is used improve thermal performance each solution. The evaluates possible transformation several design variables in plan, such as orientation reflection, window size, overhang fin wall translation. A geometric variable sequential procedure satisfy strategy. case study single-family house carried out, where two sets, with 12 solutions each, assessed, optimized criteria. results demonstrate that plans may by up 41% single-level 54% two-level When comparing within set, ranks first 33% 29% better than worst design, second set respectively.