Spatial Graph Convolutional Neural Network via Structured Subdomain Adaptation and Domain Adversarial Learning for Bearing Fault Diagnosis

TL;DR

This paper introduces DSAGCN, a graph convolutional neural network that models data structure and uses adversarial and local discrepancy methods for unsupervised bearing fault diagnosis across changing conditions.

Contribution

The paper proposes a novel deep subdomain adaptation graph convolutional neural network that considers data structure and aligns subdomains for improved fault diagnosis.

Findings

DSAGCN outperforms comparison models on bearing datasets.

Aligning structured subdomains enhances fault diagnosis accuracy.

Combining adversarial and LMMD methods improves domain adaptation.

Abstract

Unsupervised domain adaptation (UDA) has shown remarkable results in bearing fault diagnosis under changing working conditions in recent years. However, most UDA methods do not consider the geometric structure of the data. Furthermore, the global domain adaptation technique is commonly applied, which ignores the relation between subdomains. This paper addresses mentioned challenges by presenting the novel deep subdomain adaptation graph convolution neural network (DSAGCN), which has two key characteristics: First, graph convolution neural network (GCNN) is employed to model the structure of data. Second, adversarial domain adaptation and local maximum mean discrepancy (LMMD) methods are applied concurrently to align the subdomain's distribution and reduce structure discrepancy between relevant subdomains and global domains. CWRU and Paderborn bearing datasets are used to validate the DSAGCN method's efficiency and superiority between comparison models. The experimental results demonstrate the significance of aligning structured subdomains along with domain adaptation methods to obtain an accurate data-driven model in unsupervised fault diagnosis.