基于粒子群算法的油箱底壳拉延工艺参数优化  

Optimization of Drawing Process Parameters of Oil Tank Bottom Shell Based on Particle Swarm Optimization

作　　者：檀生辉 吴勇 王东 蒋立伟 伍旭东 吴鹏 Tan Sheng-hui;Wu Yong;Wang Dong;Jiang Li-wei;Wu Xu-dong;Wu Peng(Wuhu Research Institute,Xidian University,An Hui Wuhu 241002)

机构地区：[1]西安电子科技大学芜湖研究院,安徽芜湖241000

出　　处：《内燃机与配件》2022年第19期56-59,共4页Internal Combustion Engine & Parts

摘　　要：针对钣金产品在成型过程中易产生拉裂和褶皱等缺陷。本文以钣金油箱底壳为例,将压边力、拉延速度、模具间隙、摩擦系数作为试验因素,最大减薄率和最大增厚率作为评价指标。将拉丁超立方抽样得出20组试验数据做对数变换后建立线性回归方程。以最大增厚率的对数值不大于2.5为约束,最大减薄率最小为目标建立有约束的线性方程。用自适应权重粒子群算法对线性方程组进行最小值寻优。得到最大减薄率为21.24%,最大增厚率为9.47%。Aiming at the defects of sheet metal stamping products in drawing forming process.In this paper,the sheet metal tank bottom was taken as an example.Four process parameters including blank holder force,drawing speed,die clearance and friction coefficient were taken as test factors,and the maximum thinning rate and maximum thickening rate were taken as evaluation indexes.The linear regression equation was established after logarithmic transformation of 20 groups of experimental data obtained from Latin Hypercube Sampling.A constrained linear equation was established with the logarithm of the maximum thickening rate not more than 2.5 as the constraint and the minimum of the maximum thinning rate as the objective.An adaptive weighted particle swarm optimization algorithm was used to minimize linear equations.The maximum thinning rate is 21.24%,and the maximum thickening rate is 9.47%.

关 键 词：油箱底壳 拉丁拉立方抽样 回归分析 固定权重 权重线性递减 自适应权重 工艺参数优化 

分 类 号：TH137.86[机械工程—机械制造及自动化]