基于微分经验模式分解和隐马尔科夫模型的滚动轴承故障诊断方法  

Fault Diagnosis Method for Rolling Bearing Based on Differential-based Empirical Mode Decomposition and Hidden Markov Model

作　　者：孟宗 闫晓丽 

机构地区：[1]河北省测试计量技术及仪器重点实验室,河北秦皇岛066004 [2]国家冷轧板带装备及工艺工程技术研究中心,河北秦皇岛066004

出　　处：《计量学报》2015年第5期482-486,共5页Acta Metrologica Sinica

基　　金：国家自然科学基金(51105323);河北省自然科学基金(E2015203356);河北省高等学校科学研究计划重点项目(ZD2015049)

摘　　要：提出基于微分经验模式分解（DEMD）和隐马尔科夫模型（HMM）的旋转机械故障诊断方法，并应用到滚动轴承故障诊断中。首先，对故障信号进行基于微分的经验模式分解，提取瞬时能量作为故障特征向量；然后将故障特征向量输入HMM分类器进行模式识别，输出各状态似然概率值；以最大似然概率所对应的故障状态作为诊断结果，最终实现滚动轴承故障诊断。滚动轴承点蚀故障的诊断实验证明了该方法的有效性。与基于EMD-HMM的故障诊断方法相比，基于DEMD-HMM的故障诊断方法更适用于滚动轴承故障诊断。Based on the differential-based empirical mode decomposition（DEMD） and hidden Markov model（ HMM）, a new method for rotating machinery fault diagnosis is proposed. The method is applied to rolling bearing fault diagnosis. First of all, fault signals are decomposed by DEMD, the instantaneous energy distribution of each signal is extracted to form the fault feature vectors, and then input the feature vectors into the HMM classifier for malfunction recognition, the maximum likelihood probability which is output by HMM classifier is in the fault state. Finally, different fault types are recognized. A practical fault signal of a rolling bearing with corrosive pitting is applied to test the method. Experimental result showed that the method of DEMD-HMM is superior to the method of EMD-HMM and can identify the rolling bearing fault accurately and effectively.

关 键 词：计量学 轴承故障诊断 微分经验模式分解 隐马尔科夫模型 

分 类 号：TB936[一般工业技术—计量学]