SECTION OPTIMIZATION OF RIGID STEEL FRAMES USING RESPONSE SURFACE MODELS

Abstract

The increase of combination number according to the increase of variables is a serious problem in using section optimization for steel structures. Response Surface Method (RSM) which uses approximate functions made by sampling data was proofed as one of effective optimization methods. This paper shows features of the proposed method that combined with genetic algorithm by applying it to allowable stress design of a middle rise steel structure and changing the setting of RSM which are a response surface algorithm and a space filler algorithm. In addition, the feature of the optimized structures is analyzed by nonlinear dynamic response analysis.