Deep learning fault diagnosis method based on global optimization GAN for unbalanced data

By F. Zhou, Shuai Yang, Hamido Fujita, Danmin Chen, Chenglin Wen. Knowledge-Based Systems, 2020

515 citations recorded by the source corpus.

Abstract Deep learning can be applied to the field of fault diagnosis for its powerful feature representation capabilities. When a certain class fault samples available are very limited, it is inevitably to be unbalanced. The fault feature extracted from unbalanced data via deep learning is inaccurate, which can lead to high misclassification rate. To solve this problem, new generator and discriminator of Generative Adversarial Network (GAN) are designed in this paper to generate more discriminant fault samples using a scheme of global optimization. The generator is designed to generate those fault feature extracted from a few fault samples via Auto Encoder (AE) instead of fault data sample. The training of the generator is guided by fault feature and fault diagnosis error instead of the statistical coincidence of traditional GAN. The discriminator is designed to filter the unqualified generated samples in the sense that qualified samples are helpful for more accurate fault diagnosis. The experimental results of rolling bearings verify the effectiveness of the proposed algorithm.

Subfields: Artificial Intelligence, Control and Systems Engineering

View Paper: https://www.semanticscholar.org/paper/04eaa55f50dff99af2accbbd93813ea35101e8ce

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jay H. Lee
2. Xiaodong Jia
3. Shibin Wang
4. Tianfu Li
5. José Barata
6. K. Sun
7. Qin Wang
8. Olga Fink

In Control and Systems Engineering

1. Chuang Sun
2. Ruqiang Yan
3. Xuefeng Chen
4. Jay H. Lee
5. Shibin Wang
6. Qin Wang
7. Olga Fink
8. Markus Svensén

Citations in the corpus

• Potential, challenges and future directions for deep learning in prognostics and health management applications (2020, background): “Recently, several approaches have been proposed to generate faulty samples or fault features with generative neural networks (Yu et al., 2019; Zhou et al., 2020).”
• Deep Learning Algorithms for Rotating Machinery Intelligent Diagnosis: An Open Source Benchmark Study (2020, methodology): “For imbalanced learning, generation adversarial network (GAN) was used to combine with AE models to generate new labeled samples in [43–45].”
• Industrial Artificial Intelligence in Industry 4.0 - Systematic Review, Challenges and Outlook (2020, background): “Some promising examples of this can be found in the literature, addressing the class imbalance problem in applications in which the data of faulty operational conditions is scarce and difficult to obtain [82]– [84].”