基于提升小波包变换的滚动轴承包络分析诊断方法  

An Envelop Analysis Approach for Ball Bearing Based on Lifting Wavelet Packet Transform

作　　者：段晨东[1] 郭研[2] 

机构地区：[1]长安大学电子与控制工程学院,西安市710061 [2]长安大学工程机械学院

出　　处：《农业机械学报》2008年第5期192-196,共5页Transactions of the Chinese Society for Agricultural Machinery

基　　金：陕西省自然科学基金资助项目(项目编号:2005E205)

摘　　要：提出了一种基于提升小波包变换提取滚动轴承损伤特征的方法。在提升模式的框架下,根据小波包变换的思想,对细节信号进一步采用提升模式进行分解,实现了提升小波包变换。为了有效地获取滚动轴承的损伤特征,选择合适的小波包基函数,把振动信号按给定的尺度分解,以该尺度信号能量最大的小波包信号分解频带作为最佳小波包,再用解调分析法提取特征频率。并采用仿真和实验信号对该方法进行了验证。采用该方法对无损伤和含有损伤的滚动轴承进行分析,取得了较好的诊断效果。At the basis of the lifting wavelet packet transform （WPT）, a method was proposed for extracting ball bearing local defect feature. The lifting WPT was implemented by using the approximation signal obtained from the lifting scheme for further decomposition, In order to extract ball bearing defect feature efficiently, the proposed method decomposed vibration signal to a proper level by adopting a selected basis, then a wavelet packet frequency-band possessing the most energy on this decomposition level was taken as the optimum packet. Furthermore, the decomposition signal of the optimum packet was applied to extract the defect characteristic frequency via the envelop analysis approach. The proposed method was evaluated by employing simulated and experimental signals. Good and faulty ball bearings were tested and analyzed by experiment, respectively. The results proved that the method presents ideal performance, which provides an automatic monitoring and diagnostic way for ball bearing.

关 键 词：滚动轴承 损伤特征 包络分析 提升模式 小波包变换 

分 类 号：TH133.33[机械工程—机械制造及自动化] TP206.3[自动化与计算机技术—检测技术与自动化装置]