Bidirectional Local–Global Interaction-Guided GAN With Discriminator Gradient Gap Regularization for Bearing Fault Diagnosis With Unbalanced Datasets

Abstract

In complex industrial environments, mechanical devices predominantly operate under normal conditions, resulting in a shortage of available fault samples. This imbalance significantly impedes the effectiveness of intelligent fault diagnosis approaches. To overcome the challenge, an innovative generative adversarial network (GAN) model is proposed in this study, termed the bidirectional local–global interaction-guided GAN with discriminator gradient gap regularization (Bi-Interaction GAN). First, the generator is designed to incorporate a bidirectional interaction mechanism that accounts for the interaction between local and global features of vibration signals, effectively capturing the deep relationships between local and global information in the bearing vibration signals with limited datasets. Subsequently, the stability of the model is enhanced by introducing a novel loss function that integrates discriminator gradient gap regularization with the Wasserstein distance. This approach aims to minimize the gradient discrepancy between real and generated samples in the discriminator, thereby facilitating a more stable training process. Extensive comparative experiments are carried out to validate the proposed approach using a widely utilized dataset and realistic experimental configuration. The results indicate that the Bi-Interaction GAN outperforms existing state-of-the-art (SOTA) GAN models in generating superior-quality samples while achieving high classification accuracy.

Topics & Concepts