基于多尺度排列熵的自调心双列滚动轴承故障诊断

The fault diagnosis of self-aligning double row rolling bearing based on multi-scale permutation entropy

针对热模锻压力机传动系统中支承高速轴的自调心双列滚动轴承的故障诊断问题,提出采用多尺度排列熵(MPE)的故障诊断方法。首先利用故障模拟试验台与信号采集系统采集自调心双列滚动轴承不同故障状态的振动信号;其次对MPE关键参数进行选取,利用MATLAB提取了振动信号不同尺度因子的MPE特征并构建MPE特征向量样本;最后利用MATLAB中的LSSVM工具箱对自调心双列滚动轴承的不同故障进行模式识别。识别结果表明,MPE特征可以强而有力地表征自调心双列滚动轴承状态,LS-SVM分类器对滚动轴承的不同故障具有较高的识别准确率。

Aiming at the fault diagnosis of self-aligning double row rolling bearings supporting high speed shafts in hot die forging press transmission system,a multi-scale permutation entropy(MPE)fault diagnosis method is proposed.The fault simulation test bench and the signal acquisition system are used to collect the vibration signals of different fault states of the self-aligning double row rolling bearing.Then the key parameters of MPE are selected,and the MPE features of different scale factors of vibration signals are extracted by MATLAB and MPE feature vector samples are constructed.Finally,LS-SVM,LIBSVM and other toolboxes in MATLAB are used to identify the different faults of self-aligning double row rolling bearings.The recognition results show that the MPE feature can strongly characterize the self-aligning double row rolling bearings and the classifiers such as LS-SVM have a high recognition accuracy for different faults of rolling bearing state.