WOA-SVM算法在钛合金端铣刀具磨损预测的研究  

Research on Tool Wear Prediction of Titanium Alloy End Milling Based on WOA-SVM Algorithm

作　　者：梁柱[1] 宋小春[2] LIANG Zhu;SONG Xiaochun(College of Intelligent Manufacturing,Guangdong Innovative Technical College,Dongguan Guangdong 523960,China;School of Mechanical and Automotive Engineering,South China University of Technology,Guangzhou Guangdong 510641,China)

机构地区：[1]广东创新科技职业学院智能制造学院,广东东莞523960 [2]华南理工大学机械与汽车工程学院,广东广州510641

出　　处：《机床与液压》2022年第15期166-174,共9页Machine Tool & Hydraulics

基　　金：2021年度广东省教育厅科研平台项目(KYXM2021064)。

摘　　要：针对钛合金加工中刀具磨损状态的准确识别问题,建立了基于支持向量机(SVM)和鲸鱼优化算法(WOA)的钛合金刀具磨损预测模型。将SVM和WOA相结合,提出了一种新的WOA-SVM模型,用于钛合金立铣刀刀具磨损的精确估计。通过提取切削力的信号特征作为监测特征,利用邻域保持嵌入(NPE)对监测特征实现降维,提高了WOA-SVM模型的建模效率。实验结果表明:在保证预测精度的前提下,NPE的使用使WOA-SVM的建模时间减少了90%以上;与PSO-SVM和GSA-SVM等常用方法相比,WOA-SVM具有较高的预测精度,建模时间减少了30%以上;所建模型能有效预测钛合金加工刀具的磨损状态。Aiming at the problem of accurate recognition of tool wear state in titanium alloy processing,a wear prediction model for titanium alloy tool based on support vector machine(SVM)and whale optimization algorithm(WOA)was established.A new WOASVM model was proposed by combining SVM and WOA for accurate estimation of titanium alloy end milling tool wear.By extracting the signal feature of cutting force as monitoring feature,the dimension reduction of monitoring feature was realized by using neighborhood preserving embedding(NPE),by which the modeling efficiency of WOA-SVM model was improved.The experimental results show that using NPE reduces the modeling time of WOA-SVM by more than 90%on the premise of ensuring the prediction accuracy.Com⁃pared with the common methods such as PSO-SVM and GSA-SVM,WOA-SVM has higher prediction accuracy and the modeling time is decreased by more than 30%.The model can be used to effectively predict the wear state of titanium alloy cutting tools.

关 键 词：刀具磨损估计 邻域保持嵌入(NPE) 支持向量机(SVM) 钛合金 鲸鱼优化算法(WOA) 

分 类 号：F416.41[经济管理—产业经济] TG54[金属学及工艺—金属切削加工及机床]