基于小样本数据驱动的弹壳打凹-平底成形工艺参数优化决策方法OA北大核心CSTPCD

Aiming at the problems of low quality and short die life during the production commis-sioning,an optimization decision method for cartridge case indenting-heading process parameters driv-en by small sample was proposed.Firstly,the central composite experimental method was used to de-sign experiments,and each experimental scheme was incorporated into the finite element model for numerical simulation.Taking the maximum effective stress of indenting ejector,the maximum effec-tive stress of heading punch and inner fillet at the bottom of the cartridge case as the optimization goals,random forest algorithm was combined to construct the multi-objective optimization model of processing parameters in indenting-heading processes of cartridge cases based on the simulation re-sults.Secondly,the improved multi-objective grey wolf algorithm was applied to optimize the multi-objective optimization model and obtain a pareto solution.The optimal process parameter combination was evaluated and determined using the comprehensive entropy weight-TOPSIS method.Finally,the numerical simulation and processing experiments were carried out with the combination of optimal process parameters.The results show that the simulation results are consistent with the processing experiments,and the inner fillets at the bottom of the cartridge cases are filled fully,and the service life of the die is improved.