摘要
利用小波包分析技术适于对非平稳信号进行特征提取和支持向量机在小样本情况下具有较强分类能力的特点,提出了一种基于小波包分析和支持向量机相结合进行刀具切削故障诊断的方法。该方法采用小波包分析对其提取特征向量,利用支持向量机故障分类器实现对刀具切削故障分类。试验结果表明,小波包分析和支持向量机能对刀具故障进行有效诊断,故障预报正确率为90%。
Wavelet package analysis(WPA) is suitable for analyzing non-stationary signals. Support vector machine(SVM) poss esses excellent classification capacity for small samples. According to these features, a method of diagnosis of cutting faults based on WPA and SVM was presented. The characteristics of the cutting tool were extracted by WPA, and the cutting faults were classified by the SVM classifier. Experiment results show that both WAP and SVM are suitable for diagnosis of cutting faults, and the rate of successful prediction about the faults is 90%.
出处
《振动．测试与诊断》
EI
CSCD
2008年第3期273-276,共4页
Journal of Vibration,Measurement & Diagnosis
基金
四川省教育厅青年基金资助项目(编号:2006B043)
关键词
刀具
故障诊断
小波包分析
支持向量机
cutting tool fault diagnosis wavelet package analysis support vector machine
