基于优化VMD-SVM参数的铣削颤振识别  

Milling Chatter Recognition Based on Optimized VMD-SVM Parameters

作　　者：贾广飞[1] 姚海洋 JIA Guangfei;YAO Haiyang(School of Mechanical Engineering,Hebei University of Science and Technology,Shijiazhuang 050018,China)

机构地区：[1]河北科技大学机械工程学院,石家庄050018

出　　处：《组合机床与自动化加工技术》2023年第6期149-153,共5页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：河北省自然科学基金项目(E2020208052)。

摘　　要：铣削颤振是影响加工质量和切削效率的重要因素之一,为了有效抑制颤振发生,需要对铣削过程进行颤振识别。针对颤振识别中特征提取和识别准确率问题,提出基于麻雀搜索算法(SSA)优化变分模态分解(VMD)和支持向量机(SVM)关键参数的铣削颤振识别方法。采用SSA对VMD的关键参数以平均包络熵最小原则进行寻优;使用优化后的VMD分解铣削颤振信号,根据相关系数法筛选敏感的本征模态函数(IMF)分量重构原信号;提取重构信号的小波包能量熵构造颤振特征向量矩阵并输入经SSA优化的SVM模型中进行模型训练和铣削颤振识别。结果表明,提出的经SSA优化的VMD-SVM模型与未优化的VMD-SVM模型相比具有更高的识别准确率。Milling chatter is one of the important factors affecting machining quality and cutting efficiency.In order to effectively suppress the occurrence of chatter,it is necessary to identify the chatter in the milling process.Aiming at the problem of feature extraction and recognition accuracy in chatter recognition,a milling chatter recognition method based on sparrow search algorithm(SSA)to optimize the key parameters of variational mode decomposition(VMD)and support vector machine(SVM)is proposed.SSA is used to optimize the key parameters of VMD based on the principle of minimum average envelope entropy;The optimized VMD is used to decompose the milling chatter signal,and the sensitive intrinsic mode function(IMF)components are selected according to the correlation coefficient method to reconstruct the original signal;The energy entropy of the wavelet packet of the reconstructed signal is extracted to construct the chatter feature vector matrix and input into the SSA optimized SVM model for model training and milling chatter recognition.The results show that the proposed SSA optimized VMD-SVM model has higher recognition accuracy than the non optimized VMD-SVM model.

关 键 词：铣削颤振 变分模态分解 麻雀搜索算法 平均包络熵 支持向量机 

分 类 号：TH161[机械工程—机械制造及自动化] TG54[金属学及工艺—金属切削加工及机床]