STRUCTURAL OPTIMIZATION BASED ON RESPONSE SURFACE METHODOLOGY

Using response surface methodology (RSM), the responses of asystem can be expressed in the explicit form of the designvariables. For fitting response surfaces, the central symmetric designand thequasi-simplex design are developed according to central composite design andsimplex design, respectively, in an attempt to cut down the computationalcost involved in repetitive structural analyses with the accuracy beingguaranteed. For a common optimization problem, the RSM canformulate it into an optimal model linearly or nonlinearly constrained. Theoptimization toolbox of Matlab can be employed to find the optima. Thecrashworthiness design shows that the application of the proposedexperimental design methods and optimal model can achieve good accuracy andefficiency.