基于AMGA与IPSO算法的42CrMo超声滚挤压工艺参数优化
Optimization of 42CrMo ultrasonic rolling extrusion process parameters based on AMGA and IPSO algorithm
石青松 1徐红玉 1王晓强 1付浩然1
作者信息
- 1. 河南科技大学机电工程学院,河南洛阳 471003
- 折叠
摘要
为确定超声滚挤压轴承套圈工艺参数的最优解集,以42CrMo钢为研究对象,表面粗糙度、残余压应力和硬度为表层性能评价指标,设计正交试验.基于试验数据,采用多元回归法建立评价指标的数学预测模型,进行方差分析.对超声滚挤压工艺参数分别采用存档微遗传算法(AMGA)和改进粒子群(IPSO)算法进行多目标优化,对优化后的Pareto前沿图与计算效率进行对比分析,结果表明:在Pareto前沿图中,AMGA最优迭代2000次优于IPSO算法最优迭代3600次;得到超声滚挤压工艺参数最优解集:转速[250,355]r·min-1、进给速度[13,24]mm·min-1、振幅[16,22]µm、静压力[488,650]N;表层性能评价指标最优解集:表面粗糙度[0.398,0.501]μm、残余压应力[823,986]MPa、硬度[713,742]HV.通过试验验证了算法优化的可靠性和精确性.
Abstract
To determine the optimal solution set of process parameters of ultrasonic rolling extrusion bearing ring,the orthogonal tests were designed taking 42CrMo steel as the research object,surface roughness,residual compressive stress and hardness as the surface per-formance evaluation indicators.Based on the test data,the multiple regression method was used to establish the mathematical prediction model for evaluation indicators and the variance analysis was carried out.The archived microgenetic algorithm(AMGA)and improved parti-cle swarm optimization(IPSO)algorithm were used to optimize the parameters of ultrasonic rolling extrusion,and the Pareto frontier diagrams and computational efficiency were compared and analyzed.The results show that in the Pareto frontier diagrams,the optimal iteration 2000 times of AMGA is better than the optimal iteration 3600 times of IPSO algorithm.The optimal solution set of ultrasonic rolling extrusion process param-eters is obtained:rotation speed of[250,355]r·min-1,feed speed of[13,24]mm·min-1,amplitude of[16,22]μm,static pressure of[488,650]N.The optimal solution set of surface performance evaluation indexes:surface roughness of[0.398,0.501]μm,residual compressive stress of[823,986]MPa,hardness of[713,742]HV.The reliability and accuracy of algorithm optimization were verified by tests.
关键词
超声滚挤压/多元回归法/方差分析/AMGA/IPSO算法Key words
ultrasonic rolling extrusion/multiple regression method/variance analysis/AMGA/IPSO algorithm引用本文复制引用
基金项目
国家重点研发计划(2022YFC2805702)
出版年
2024
NETL
NSTL
万方数据