AttGAN-DPCNN: An Extremely Imbalanced Fault Diagnosis Method for Complex Signals From Multiple Sensors

The intelligent fault diagnosis in engineering scenarios frequently struggles to achieve the desired results due to the presence of data imbalance. Meanwhile, since the data used may come from multiple sensors, and there are compound faults coupled by multiple faults, conventional imbalanced fault diagnosis methods are difficult to facilitate accurate fault diagnosis. To address this issue, from the perspective of data augmentation (DA) and feature processing, this article proposes AttGAN-DPCNN, an imbalanced fault diagnosis scheme that leverages an attention-based generative adversarial network (AttGAN) and dynamic perceptual convolutional neural network (DPCNN). First, the original generative adversarial network (GAN) is enhanced by incorporating an attention-based multilevel information fusion structure and a class-conditional loss function, thereby facilitating the generation of high-quality multisensor signals while effectively balancing the data distribution. Second, in order to further improve the feature recognition ability of the model for multisensor complex signals, a lightweight and high-performance dynamic perceptual network is constructed to adaptively mine the deep features of signals via 4-D dynamic convolution. Finally, the performance of the method is validated on the bogie dataset, achieving accuracies of 97.1%, 98.0%, 97.1%, and 96% for the multiclass extreme imbalance and three single-class extreme imbalance cases, respectively. Meanwhile, its robustness is evaluated on the standard bearing dataset, which reaches an accuracy of 97.4% in the multiclass imbalance case and consistently surpassing 99% in all single-class imbalance cases. This study provides a reliable solution to the problem of extremely imbalanced data for complex signals in engineering scenarios.