滚动轴承故障程度诊断的HMM方法研究

为提高滚动轴承故障诊断率,提出基于时频域指标的HMM轴承故障程度诊断方法。利用轴承故障模拟试验台,采集不同剥落程度滚动体的振动信号,分别提取均方值、有效值、方差、修正样本方差、标准差、频域中心及带宽共7个时域和频域指标作为特征向量训练HMM,得到基于HMM的诊断分类器。利用该分类器对330组待检滚动轴承振动信号进行分析,诊断正确率达90%以上,说明该方法能有效提取故障特征。

HMM in Fault Severity Diagnosis for Rolling Bearings

In order to improve the fault diagnosis rate of rolling bearings,a new method for diagnosing rolling bearing faults using the theory of HMM is presented based on time-frequency domain indexes.Using a test rig of rolling bearing fault simulation,the vibration signals of different sizes of spalling faults in bearing ball are collected.The model is trained based on feature vectors consisted by seven indexes extracted from time domain and frequency domain of the signals.The index includes mean square value,RMS,variance,modified sample variance,standard deviation,frequency center and bandwidth.Then,a diagnosis classifier based on the trained model is obtained.The classifier is applied to test 320 sets of vibration signals with different fault severity,the diagnosis accuracy rate is over 90%.The result demonstrates that the proposed method is effective to extract fault features.