机械与电子2017，Vol.35Issue(6)：8-11,4.
时变奇异值分解在轴承故障特征提取中的应用研究
Application of Time-varying Singular Value Decomposition in Feature Extraction of Bearing Fault
摘要
Abstract
Aiming at the identification of bearing faults, this paper proposes a time-varying singular value decomposition method to extract the spectral feature of bearing faults.The proposed method employs the singular value decomposition to each signal segment realized by a sliding window, and then stores the singular values in a time-varying singular value matrix, each row of which is a time-varying singular value sequence (TSVS).The TSVS has good periodicity, and its frequency is related to the periodic component of the original signal.The experimental results show that the proposed method has significant advantages over the traditional one in feature extraction of bearing faults.关键词
轴承故障诊断/奇异值分解/时变奇异值分解/特征提取Key words
bearing fault diagnosis/singular value decomposition/time-varying singular value decomposition (TSVD)/feature extraction分类
机械制造引用本文复制引用
袁涛,张尚斌,何清波..时变奇异值分解在轴承故障特征提取中的应用研究[J].机械与电子,2017,35(6):8-11,4.基金项目
国家自然科学基金资助项目(51475441) （51475441）
