Machine fault diagnosis with small sample based on variational information constrained generative adversarial network. (October 2022)

• Record Type: Journal Article
• Authors: Liu, Shaowei; Jiang, Hongkai; Wu, Zhenghong; Liu, Yunpeng; Zhu, Ke
• Is Part Of: Advanced engineering informatics. Volume 54(2022)
• Publication Date: 2022-10
• Subjects: Fault diagnosis -- Generative adversarial network -- Rolling bearing -- Variational information constraint -- Small sample; Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2022.101762
• Languages: English
• ISSNs: 1474-0346

Abstract

Highlights: The encoder is incorporated into the discriminator to map the deep features. A variational information constraint technique is used to constrain the gradients. A representation matching module is added to avoid the mode collapse problem. Abstract: In actual engineering scenarios, limited fault data leads to insufficient model training and over-fitting, which negatively affects the diagnostic performance of intelligent diagnostic models. To solve the problem, this paper proposes a variational information constrained generative adversarial network (VICGAN) for effective machine fault diagnosis. Firstly, by incorporating the encoder into the discriminator to map the deep features, an improved generative adversarial network with stronger data synthesis capability is established. Secondly, to promote the stable training of the model and guarantee better convergence, a variational information constraint technique is utilized, which constrains the input signals and deep features of the discriminator using the information bottleneck method. In addition, a representation matching module is added to impose restrictions on the generator, avoiding the mode collapse problem and boosting the sample diversity. Two rolling bearing datasets are utilized to verify the effectiveness and stability of the presented network, which demonstrates that the presented network has an admirable ability in processing fault diagnosis with few samples, and performs better than state-of-the-art approaches.