摘要
提出了一种基于匹配追踪的齿轮振动信号的特征提取方法.通过仿真计算将该方法与传统的时频分析方法,如短时傅里叶变换进行比较,验证了该方法的优越性.将匹配追踪方法应用到实际的齿轮故障诊断中.结果表明,该方法能准确地提取故障齿轮的冲击和谐波成分.因此,匹配追踪方法应用于齿轮故障诊断是有效的.
A matching pursuit based feature extraction method of gears vibration was presented. According to simulation, this method verifies its advantages compared with the traditional time-frequency analysis methods, such as short time fourier transform. Matching pursuit method was applied to actual gears fault diagnosis. The results show that this method can exactly extract the impulses and harmonic components of damaged gear vibration signals. The matching pursuit method is effective for the fault diagnosis of gear.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2009年第6期910-913,共4页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金资助项目(50675140)
国家高技术研究发展计划(863)项目(2006AA04Z175)
关键词
匹配追踪
特征提取
故障诊断
齿轮
matching pursuit
feature extraction
fault diagnosis
gear
