基于EMD重构和SVM的滚动轴承故障诊断方法研究  

A Study of Rolling Bearing Fault Diagnosis Based on EMD Reconstruction and SVM

作　　者：周小龙[1] 杨恭勇 梁秀霞 李家飞 

机构地区：[1]东北电力大学工程训练教学中心,吉林吉林132012 [2]河南信宇石油机械制造股份有限公司,河南濮阳457001

出　　处：《东北电力大学学报》2016年第6期71-76,共6页Journal of Northeast Electric Power University

摘　　要：针对滚动轴承振动信号非平稳性和故障特征难以提取的问题,提出一种基于经验模态分解重构和支持向量机相结合的故障诊断方法。首先,采用经验模态分解,将滚动轴承振动信号分解成一系列固有模态函数;其次,根据伪固有模态函数剔除法选取对故障特征敏感的模态函数进行信号重构;最后,以重构信号的有效值和峭度值作为支持向量机分类器的输入来识别滚动轴承的工作状态和故障类型。试验结果表明,该方法能有效地识别和诊断出滚动轴承的运行状态和故障类型,具有很高的工程实用价值。Aiming at the non-stationary characteristic of the rolling bearing vibration signal and the difficulty to get the fault features in its fault diagnosis, a rolling bearing fault diagnosis method based on empirical mode decomposition reconstruction and support vector machine is proposed. Firstly, the fault signal is decomposed by empirical mode decomposition. Then, the intrinsic mode functions are obtained and sensitive intrinsic mode functions are selected by the sensitivity evaluation method. Finally, the valid and kurtosis values of the reconstruction signal as input vectors of support vector machine, and identify the rolling bearing fault pattern and condition. The experiment shows that this method can identify rolling bearing fault patterns effectively and it has a practical value.

关 键 词：经验模态分解 固有模态函数 支持向量机 滚动轴承 故障诊断 

分 类 号：TH17[机械工程—机械制造及自动化] TH133.33