Anti-noise transfer adversarial convolutions with adaptive threshold for rotating machine fault diagnosis

Fault diagnosis based on deep learning (DL) has been a research hotspot in recent years. However, the current neural networks are getting larger and larger, with more and more parameters and insufficient noise resistance, making it difficult to effectively apply these methods to real working conditions. To address these issues, we propose a novel deep learning method with fewer parameters and better noise resistance based on transfer adversarial subnetwork (TAS) and channel-wise thresholds (CWT), namely, anti-noise transfer adversarial convolutions (ANTAC). In the proposed method, the original data and feature vectors are mapped to reproducing kernel Hilbert space (RKHS) and processed by maximum mean discrepancy (MMD) and Wasserstein distance (WD), which makes the method more capable to distinguish the similar features without producing any additional training parameters. Furthermore, white Gaussian noise (WGN) and the soft thresholding method with CWT are used to reduce data noise and improve the robustness and noise resistance of the network. Finally, the superiority of the proposed method is validated through experiments on different datasets, network structures and the data with different SNRs. The results show that the proposed method has better feature discrimination ability, noise resistance, and fewer parameters compared with other methods. The highest accuracy of the proposed method is 99.90% on the test set. •In order to enhance the network ability in distinguishing similar types of samples, TAS is proposed. Experiments show that the proposed method can effectively improve the network stability and the feature discrimination ability with fewer parameters.•The anti-interference ability and robustness of the network is enhanced by adding a certain disturbance into it. In addition. The soft threshold function with channel-wise threshold is used to improve the denoising ability of the network.•To enhance the network generalization and sensitivity to similar features, AFV loss is calculated by WD and controlled by scaling factor λ, and finally sent into RMSprop optimizer together with classification loss for optimization.•The method can directly process the original signals and the procedures of denoising and feature extraction are finished automatically in the method, which improves the fault diagnosis efficiency and provides a further practical possibility for the deployment of end-to-end fault diagnosis platforms.