基于小波包分析和LS-SVM的钻削刀具状态识别研究

2019-12-11

作者：郝碧君，陈妮，李亮，郭月龙，何宁单位：南京航空航天大学

摘要：为了有效地识别钻削刀具磨损状态，提出一种基于小波包分析和最小二乘支持向量机(LS-SVM)的状态识别方法。通过在线监测钻削加工过程中的钻削轴向力和刀具状态，采用时域分析、频域分析以及小波包分析法对刀具磨损状态的信号进行特征向量提取，建立基于最小二乘支持向量机（LS-SVM）的分类识别模型。通过试验验证了该方法可以提高刀具磨损状态的识别精度。

关键词：刀具状态识别；特征提取；小波包分析；最小二乘支持向量机

中图分类号：TG713.1；TH117.1文献标志码：ADOI：10.3969/j.issn.1000-7008.2019.12.001

Tool Wear Condition Recognition in Drilling Based on

Wavelet Packet Analysis and LSSVM

Hao Bijun，Chen Ni，Li Liang，Guo Yuelong，He Ning

Abstract:In order to identify the wear condition of the drilling tools effectively，this paper presents the condition recognition method based on the wavelet packet analysis and least squares support vector machine(LSSVM).This method monitored the axial drilling forces and tool condition during the drilling process，and used the time domain analysis，frequency domain analysis and wavelet packet analysis method to extract the signal features.The signal features which are extracted could reflect the wear condition of the tool as much as possible are deemed as feature vectors.A classification recognition model based on LSSVM is established to improve the accuracy of the tool condition recognition model.The experimental results show that this method could improve the recognition accuracy of tool wear condition.

Keywords:tool condition recognition；feature extraction；wavelet packet analysis；least squares support vector machine (LSSVM)