摘要
为了减小车辆座椅侧板冲压成形回弹量,提出了基于交叉-变异蜂群算法的冲压工艺参数优化方法。分析了冲压工艺参数对车辆座椅侧板回弹量的影响规律,建立了以回弹量最小为目标的冲压优化模型。基于最优拉丁超立方抽样法在四维空间中抽取了120个点,使用Autoform软件对实验结果进行了仿真。使用单隐藏层神经网络拟合工艺参数与回弹量之间的非线性关系,经验证神经网络的拟合精度较高。为了加深人工蜂群算法的局部搜索深度,将交叉和变异思想引入到该算法中,从而给出了基于交叉-变异蜂群算法的参数优化方法。经实验验证,座椅侧板冲压件的减薄率和增厚率满足约束条件,回弹量仅为1.725 mm,说明提出的优化方法可有效减小座椅侧板的冲压回弹量。
In order to reduce springback amount of vehicle seat side panel after stamping,the optimization method of stamping process parameters based on crossover-mutation bee colony algorithm was proposed.Then,the influences of the stamping process parameters on the springback amount of vehicle seat side panel were analyzed,and a stamping optimization model was established by setting the goal of minimizing springback amount.Based on the optimal Latin hypercube sampling method,120 points were extracted in 4-dimensional space,and the experimental results were simulated by software AutoForm.Furthermore,the non-linear relationship between process parameters and springback amount was fit by single hidden layer neutral network,which was clarified that the fitting accuracy of neutral network was very high.In order to deepen the local searching depth of artificial bee colony algorithm,the idea of crossover and mutation was introduced into the algorithm,and a parameter optimization method based on crossover-mutation bee colony algorithm was given.Experimental verification shows that the thinning rate and thickening rate of the stamping piece for seat side panel meet the constraint conditions,and the springback amount is only 1.725 mm,indicating that the proposed optimization method can effectively reduce the springback amount of seat side panel after stamping.
作者
程宝鑫
曹玉苹
Cheng Baoxin;Cao Yuping(Department of Mechanical Engineering,Tianjin Petroleum Vocational and Technical College,Tianjin 301607,China;College of Information and Control Engineering,China University of Petroleum(East China),Qingdao 266580,China)
出处
《锻压技术》
CAS
CSCD
北大核心
2021年第12期79-85,共7页
Forging & Stamping Technology
基金
2018年度天津市高等职业技术教育研究会课题(XVⅢ4006)。
关键词
座椅侧板
交叉-变异蜂群算法
局部搜索深度
BP神经网络
回弹量
seat side panel
crossover-mutation bee colony algorithm
depth of local searching
BP neutral network
springback amount
