A Hierarchical Sparse Discriminant Autoencoder for Bearing Fault Diagnosis

Although some traditional autoencoders and their extensions have been widely used in the research of intelligent fault diagnosis of rotating parts, their feature extraction capabilities are limited without label information. In response to this problem, this research proposes a hierarchical sparse discriminant autoencoder (HSDAE) method for fault diagnosis of rotating components, which is a new semi-supervised autoencoder structure. By considering the sparsity of autoencoders, a hierarchical sparsity strategy was proposed to improve the stacked sparsity autoencoders, and the particle swarm optimization algorithm was used to obtain the optimal sparsity parameters to improve network performance. In order to enhance the classification of the autoencoder, a class aggregation and class separability strategy was used, which is an additional discriminative distance that was added as a penalty term in the loss function to enhance the feature extraction ability of the network. Finally, the reliability of the proposed method was verified on the bearing data set of Case Western Reserve University and the bearing data set of the laboratory test platform. The results of comparison with other methods show that the HSDAE method can enhance the feature extraction ability of the network and has reliability and stability for different data sets.

A frequency-weighted energy operator and swarm decomposition for bearing fault diagnosis

The weak fault characteristics of rolling bearings are difficult to identify due to strong background noise. To address this issue, a bearing fault detection scheme combining swarm decomposition (SWD) and frequency-weighted energy operator (FWEO) is presented. First, SWD is applied to decompose the bearing fault signal into single mode components. Then, a new evaluation index termed LEP is constructed by combining the advantages of envelope entropy, Pearson correlation coefficient and L-kurtosis, and it is utilized to choose the sensitive component containing the richest bearing fault characteristics. Finally, FWEO is employed for extracting the bearing fault features from the sensitive component. Simulation and experimental analyses indicate that the LEP index has better performance than the L-kurtosis index in determining the sensitive component. The method has the effect of suppressing noise and enhancing impulse characteristics, which is superior to the SWD-based envelope demodulation method.