基于SPEA2SDE算法的超声滚挤压42CrMo表层性能工艺参数优化

Optimization of surface performance process parameters for ultrasonic rolling extruded 42CrMo based on SPEA2SDE algorithm

为提高金属材料的表层性能,实现超声滚挤压工艺参数的优化控制,以42CrMo为研究对象,以主轴转速、进给速度、振幅和静压力为主要工艺参数进行正交试验。基于方差分析构造F值进行检验,探究工艺参数对各表层性能的贡献率。基于试验数据构建了工艺参数与材料表层性能之间的预测模型,并验证了该模型的准确性。基于SPEA2SDE算法对预测模型进行了多目标优化,分析了迭代次数对遗传算法中Pareto最优解集的影响。结果表明:对表面粗糙度、残余应力的贡献率为振幅>静压力>主轴转速>进给速度,对硬度的贡献率为振幅>主轴转速>进给速度>静压力;通过试验值与预测值的对比,表面粗糙度、表面残余应力和硬度的平均误差百分比均在6%以下;SPEA2SDE算法可获得综合性能最优的加工参数组合,用于指导实际的生产加工。

To improve the surface performance of metal materials and realize the optimal control of ultrasonic rolling extrusion process parameters,42CrMo was taken as the research object,and the orthogonal tests were carried out with spindle speed,feed speed,amplitude and static pressure as the main process parameters.Constructing F-values based on variance analysis was tested,and the contribution rate of process parameters to the performance of each surface layer was explored.A predictive model between process parameters and material surface properties was constructed based on test data,and the accuracy of the model was verified.The multi-objective optimization was performed on the prediction model based on the SPEA2SDE algorithm,and the influence of iterations on the Pareto optimal solution set in the genetic algorithm was analyzed.The results show that the contribution rate to surface roughness and residual stress is amplitude>static pressure>spindle speed>feed speed,while the contribution rate to hardness is amplitude>spindle speed>feed speed>static pressure.By comparing the test values with the predicted values,the average error percentage of surface roughness,surface residual stress and hardness are all below 6%.Processing parameters combination under the optimal comprehensive performance can be obtained by using SPEA2SDE algorithm,which is used to guide the actual production.