半球形件冲压工艺的自适应NSGA-Ⅱ算法多目标优化  

Multi-objective optimization on stamping process for hemispherical part based on adaptive NSGA-Ⅱalgorithm

作　　者：范桂杰[1] Fan Guijie(College of Intelligent Manufacturing,Shandong Polytechnic,Jinan 250104,China)

机构地区：[1]山东职业学院智能制造学院,山东济南250104

出　　处：《锻压技术》2021年第11期143-149,共7页Forging & Stamping Technology

基　　金：山东省高等学校青创人才引育计划(2019152)。

摘　　要：为了减小半球形件的冲压能耗和提高冲压件质量,提出了基于自适应NSGA-Ⅱ算法的冲压工艺多目标优化方法。介绍了半球形件冲压工艺,建立了半球形件的冲压能耗和多目标优化模型。以冲压速度、压边力为优化参数,使用正交实验法设计了2因素4水平正交实验。基于Dynaform软件获得了不同实验条件下的能耗和冲压件起皱情况,使用二阶响应面拟合了目标参数与工艺参数的回归模型。在NSGA-Ⅱ算法中引入了交叉概率和变异概率随个体适应度的自适应变化方法,从而给出了自适应NSGA-Ⅱ算法对多目标优化模型的求解方法。经验证,自适应NSGA-Ⅱ算法搜索的Pareto前沿解优于传统NSGA-Ⅱ算法的Pareto前沿解。选择一个优化点进行生产验证,经自适应NSGA-Ⅱ算法优化后,半球形冲压件表面无失稳起皱现象,实际能耗均值与优化值误差仅为0.44%,说明了所建模型的精确性和优化方法的有效性。In order to reduce the stamping energy consumption of hemispherical parts and improve the quality of stamping parts,a multi-objective optimization method of stamping process based on adaptive NSGA-Ⅱalgorithm was proposed.Then,the stamping process of hemispherical parts was introduced,and the stamping energy consumption and multi-objective optimization model of hemispherical parts were established.Taking the stamping speed and blank holder force as the optimization parameters,a two-factor and four-level orthogonal experiment was designed by the orthogonal experiment method.Furthermore,the energy consumption and wrinkling situation of stamping parts under different experimental conditions were obtained by software Dynaform ynaform,and the regression model between target parameters and process parameters was fitted by the second-order response surface.Finally,the adaptive variation method of crossover probability and mutation probability with individual fitness was introduced into NSGA-Ⅱalgorithm,and the solution method of adaptive NSGA-Ⅱalgorithm for the multi-objective optimization model was given.It is verified that the Pareto frontier solution searched by the adaptive NSGA-Ⅱalgorithm is better than that of the traditional NSGA-Ⅱalgorithm,and an optimization point is selected for production verification.After optimization by adaptive NSGA-Ⅱalgorithm,there is no instability and wrinkling phenomenons on the surface of hemispherical parts,and the error between the average value of actual energy consumption and the optimization value is only 0.44%,which shows the accuracy of the established model and the effectiveness of the optimization method.

关 键 词：半球形件 自适应NSGA-Ⅱ算法 响应面模型 冲压速度 压边力 

分 类 号：TG386[金属学及工艺—金属压力加工]