반응 표면법을 이용한 운전석 에어백 및 구속 시스템 최적화

A practical optimization approach for improving for crashworthiness of impact structure of FMVSSS 208 sled test model is introduced. Due to the non-linear of crash simulation and the instability of explicit finite element analysis itself, even small changes in model parameters may cause significant changes in the result. So, global-searching ability of optimization algorithm is more important In this paper the robustness of two methods(Genetic algorithm vs. Stochastic Response Surface Method) are studied and compared with the results of deterministic optimization methods in view of running time. These approaches showed promise in implementing impact structure optimization. Especially, the SRSM(Stochastic Response Surface Method), an extension to general response surface method was developed as a part of this work for reducing the number of simulation required for estimation of function.