Reliability-based topology optimization using successive standard response surface method

Abstract

Standard response surface method (SRSM) is used for generating limit state function since the chosen random variables are independent each other. Also, successive SRSM is implemented for obtaining an accurate limit state function with much fewer experimental points than traditional response surface method (RSM). Reliability-based topology optimization (RBTO) based on bi-directional evolutionary structural optimization (BESO) using RSM and successive SRSM was performed, and the results were compared with each other. From the result of carrying out the BESO using the successive SRSM, the following conclusions were obtained. (1) It was verified that both RSM and SRSM based on BESO can be effectively applied to RBTO. (2) That volume of RBTO is 5-9% larger than that of DTO as expected. (3) The computing cost of SRSM in topology optimization is reduced to 51.8% comparing to RSM.