크리깅모델을 이용한 2 층 열차 차체의 최적설계

Metamodels such as response surface model (RSM) and kriging model are used to meet weight minimization of a double-deck train car-body. Weight minimization problem of the double-deck train car-body is required to decide 36 design variables of thicknesses for large aluminum extruded panel while satisfying stress constraints. Therefore, we attempt to reduce the number of design variables by using RSM as a screening process based on orthogonal array design. Global sensitivity and Pareto plot of responses with respect to design variables are employed as screening criterion. After screening design variables, maximum entropy sampling is employed to select additional sample points. We construct kriging model for the mass and maximum stress of a double-deck train car-body with the reduced design variables. Optimization based on kriging model is performed and the merits of this design methodology are discussed.