Aiming at the safety problem and lightweight design requirements of train-set emergency battery box

the optimization analysis was carried out by integrating multiple optimization objectives. The thickness of the main components was taken as the design variables

the total mass of the emergency battery box and the minimum stress under harsh conditions were taken as the optimization objectives

and the first-order natural frequency was taken as the constraint function. The sample data were obtained by Box-Behnken design method. A low-order polynomial response surface model was established based on the sample data

and multi-objective optimization was carried out using the third-generation non-dominated sorting genetic algorithm (NSGA-Ⅲ). The results show that compared with the single response surface method or genetic algorithm

the combination of response surface method and genetic algorithm adopted in this paper makes the optimized parameters more reasonable

lightweight and safety are guaranteed.