Deep Morphological Shrinkage Convolutional Autoencoder-Based Feature Learning of Vibration Signals for Gearbox Fault Diagnosis

Fault diagnosis is significant to guarantee the safety and reliability of the machinery. Local faults will make the collected vibration signals deviate from the normal signals. To extract fault-related features from vibration signals, many deep-learning-based methods have been proposed for machinery fault diagnosis recently. However, due to the extreme conditions (e.g., high background noise and limited labeled training samples), it is a challenging task to implement feature extraction from the collected signals with nonlinear and nonstationary characteristics. To implement feature extraction and noise reduction of vibration signals, this article proposes a novel network, that is, deep morphological shrinkage convolutional autoencoder (DMSCAE) for gearbox fault diagnosis considering the insufficient labeled training samples. First, a morphological convolutional autoencoder is proposed for noise filtering and feature extraction. Second, a multibranch structure with different structural elements (SEs) is used in the morphological layer to extract impulsive components. Finally, a soft thresholding-based shrinkage is employed to filter ineffective features, where an adaptative method is developed to adjust the threshold automatically in the backpropagation procedure. The experiments on two gearbox fault diagnosis tests are conducted to verify the performance of DMSCAE. The results indicate that DMSCAE obtains a better performance for fault diagnosis than other DNNs, for example, efficient channel attention network (ECANet) and self-calibrated convolutional network (SCNet).