滚动轴承的MSE和PNN故障诊断方法

针对滚动轴承不同运行状态振动信号具有不同复杂性的特点，提出一种新的基于多尺度熵(multiscale entropy, MSE)和概率神经网络(probabilistic neural networks, PNN)的滚动轴承故障诊断方法。该方法首先利用MSE方法对滚动轴承振动信号进行特征提取，并将其作为PNN神经网络的输入，再利用PNN自动识别轴承故障类型及故障程度。实验数据包括不同故障类型和不同故障程度样本，结果表明，相比于小波包分解和PNN结合的诊断方法，提出的方法具有更高的诊断精度，能有效实现滚动轴承故障类型及程度的诊断。

Fault Diagnosis of Rolling Bearings Using MSE and PNN

Considering vibration signals of rolling bearings exhibiting different levels of complexity in different running conditions, a novel fault diagnosis method has been proposed based on the multiscale entropy (MSE) and probabilistic neural networks (PNN). Fault feature vector is firstly extracted from vibrations using MSE and presented to the PNN neural network as inputs. The PNN network is intended to simultaneously determine bearing fault type and severity level.Experimental data were collected from an induction motor bearing involving various fault types and severity levels. The results demonstrated the proposed MSE-PNN paradigm has a higher accuracy and promising prospect in rolling bearing fault diagnosis in comparison with the combination of wavelet packet decomposition and PNN.