基于SFS-SVM的V形件弯曲工艺参数优化研究  

Research on Parameter Optimization of V Type Bending Parts Based on SFS-SVM Model

作　　者：徐承亮[1] 曹志勇[2] 王大军[3] 胡吉全[4] XU Chengliang;CAO Zhiyong;WANG Dajun;HU Jiquan(Industry College,Guangzhou Vocational College of Technology and Business,Guangzhou Guangdong 511442,China;School of Materials Science and Engineering,Hubei University,Wuhan Hubei 430062,China;School of Automation,Chongqing University of Posts and Telecommunications,Chongqing 400065,China;School of Logistics Engineering,Wuhan University of Technology,Wuhan Hubei 430072,China)

机构地区：[1]广州科技贸易职业学院产业学院,广东广州511442 [2]湖北大学材料科学与工程学院,湖北武汉430062 [3]重庆邮电大学自动化学院,重庆400065 [4]武汉理工大学物流工程学院,湖北武汉430072

出　　处：《机床与液压》2022年第2期162-166,共5页Machine Tool & Hydraulics

基　　金：广东省普通高校特色创新项目(自然科学类)(2018GKTSCX053)。

摘　　要：影响V形工件弯曲回弹的工件尺寸、力学性能、负载条件和材料各向异性等众多因素相互耦合,表现出高度复杂的非线性,从而导致回弹预测结果的不确定性。把板料回弹后的张开角和圆角半径作为目标优化函数,将支持向量机模型部署到顺序向前筛选算法中以高效筛选出最优的特征变量参数子集,从而提高弯曲回弹模型预测结果的精度与可靠性。有限元分析和实验结果的对比验证了算法模型的可行性。There are many factors influencing the springback of V shape bending parts, such as workpiece size, mechanical properties and loading conditions, which interact with each other, representing highly complicated non-linear.The bending angle and radius were taken as two objective functions.A SVM model was constructed to deploy to the sequential forward select algorithm.Thus an optimal feature subset was selected with high efficiency.Comparative experimental results with FEM and actual experiment validate the feasibility of the proposed model.

关 键 词：V形工件 支持向量机 顺序向前筛选算法 弯曲回弹 

分 类 号：TG386.31[金属学及工艺—金属压力加工]