Adversarial adaptive 1-D convolutional neural networks for bearing fault diagnosis under varying working condition

TL;DR

This paper introduces A2CNN, an adversarial 1-D CNN model that effectively adapts fault diagnosis across varying working conditions by learning domain-invariant features, outperforming traditional methods.

Contribution

The paper proposes a novel adversarial adaptive 1-D CNN with partially untied layers for efficient domain adaptation in bearing fault diagnosis.

Findings

A2CNN achieves high accuracy across different working conditions.

The model demonstrates strong fault-discriminative and domain-invariant capabilities.

Visualization confirms effective feature learning for domain adaptation.

Abstract

Traditional intelligent fault diagnosis of rolling bearings work well only under a common assumption that the labeled training data (source domain) and unlabeled testing data (target domain) are drawn from the same distribution. However, in many real-world applications, this assumption does not hold, especially when the working condition varies. In this paper, a new adversarial adaptive 1-D CNN called A2CNN is proposed to address this problem. A2CNN consists of four parts, namely, a source feature extractor, a target feature extractor, a label classifier and a domain discriminator. The layers between the source and target feature extractor are partially untied during the training stage to take both training efficiency and domain adaptation into consideration. Experiments show that A2CNN has strong fault-discriminative and domain-invariant capacity, and therefore can achieve high accuracy under different working conditions. We also visualize the learned features and the networks to explore the reasons behind the high performance of our proposed model.