基于粒子群算法的冷挤压模具结构参数优化设计

Optimal Design of Structural Parameters of Cold Extrusion Dies Based on Particle Swarm Optimization

针对石油射孔弹挤压件,对一次挤压成形和先预挤压再终成形两种工艺进行了对比,确定了合理的工艺方案,模具结构采用预应力挤压组合凹模。采用拉丁超立方方法,分别对预挤压和终成形的组合凹模的径向直径比和过盈系数进行了抽样,应用有限元方法对抽样所得的不同结构进行数值模拟,将所得模拟结果作为响应,建立等效应力与组合凹模结构参数的Kriging模型。以降低凹模内壁最大等效应力为优化目标,分别对预挤压和终成形两个阶段的组合凹模结构参数进行了优化分析。应用Kriging模型结合粒子群算法,在结构参数的变量空间内进行全局寻优,得到了各组合凹模最佳的径向直径比和过盈系数,确定出优化的组合凹模结构尺寸,为该挤压组合凹模的设计提供了定量的优化依据。

Aiming at the petroleum perforating charge shell extrusion part,the once extrusion forming process was contrasted with the pre-extrusion and final-forming process,and the reasonable process was decided using the pre-stressed combined die.The radial diameter ratio and interference coefficient of the combined die in pre-extrusion and final-forming stage were sampled.The different die structures through sampling were simulated by finite element method,and the Kriging model of effective stress and structure parameters of combined die was established.The structure parameters of combined die in pre-extrusion and final-forming stage were optimized with the objective of decreasing maximum effective stress on the surface of die cavity.The optimum radial diameter ratio and interference coefficient are obtained and the optimum structure parameters of combined die are decided by using particle swarm optimization with Kriging model for global optimization in the variable space.It has provided a basis for the extrusion combined die optimum design.