基于LMD和HMM的旋转机械故障诊断

提出了基于局部均值分解（LMD）和隐马尔科夫模型(HMM)的旋转机械故障诊断方法。首先，对故障信号进行局部均值分解，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，输出各状态的似然概率；以最大似然概率所对应的故障状态为诊断结果。通过滚动轴承点蚀故障诊断试验验证了该方法的有效性，并将其与基于EMD-HMM的故障诊断方法进行了比较。结果表明，基于LMD-HMM的故障诊断方法更适用于旋转机械的故障诊断。

Rotating Machinery Fault Diagnosis Based on Local Mean Decomposition and Hidden Markov Model

Based on LMD and HMM, a new method for rotating machinery fault diagnosis was proposed. The method was applied to rolling bearing fault diagnosis. First of all, fault signals were decomposed by LMD, the instantaneous energy distribution of each signal was 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 was output by HMM classifier was in the fault state. A practical fault signal of a rolling bearing with corrosive pitting was applied to test the method. Experimental results show that the method of LMD-HMM is superior to the method of empirical mode decomposition(EMD)-HMM and can identify the rolling bearing faults accurately and effectively.