Ensemble-Enhanced Graph Autoencoder with GAT and Transformer-Based Encoders for Robust Fault Diagnosis

2025-04-13

AI Summary

To address poor generalization of industrial machinery fault classification under varying operating conditions, this paper proposes an end-to-end diagnostic method based on dynamic graph modeling. First, Shannon entropy-guided adaptive windowing and Dynamic Time Warping (DTW) jointly construct a dynamic graph structure from vibration time-series data. Second, a collaborative graph encoder integrating Graph Attention Networks (GAT) and Graph Transformers learns robust graph representations. Finally, joint optimization of a graph autoencoder and an ensemble classifier enables accurate fault identification. Key contributions include: (i) an entropy-driven adaptive graph construction paradigm; (ii) a GAT–Transformer hybrid graph encoding architecture; and (iii) an end-to-end graph autoencoder–ensemble joint diagnosis framework. Evaluated on the CWRU dataset, the method achieves an average F1-score of 0.99—substantially outperforming CNN and LSTM baselines (0.94–0.97)—and attains 0.99 F1 for the most challenging fault class (Class 8), compared to only 0.71 for Bi-LSTM. Strong cross-domain generalization is further validated on the HUST dataset.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Intelligent Robots: State Estimation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Fault classification in industrial machinery is vital for enhancing reliability and reducing downtime, yet it remains challenging due to the variability of vibration patterns across diverse operating conditions. This study introduces a novel graph-based framework for fault classification, converting time-series vibration data from machinery operating at varying horsepower levels into a graph representation. We utilize Shannon's entropy to determine the optimal window size for data segmentation, ensuring each segment captures significant temporal patterns, and employ Dynamic Time Warping (DTW) to define graph edges based on segment similarity. A Graph Auto Encoder (GAE) with a deep graph transformer encoder, decoder, and ensemble classifier is developed to learn latent graph representations and classify faults across various categories. The GAE's performance is evaluated on the Case Western Reserve University (CWRU) dataset, with cross-dataset generalization assessed on the HUST dataset. Results show that GAE achieves a mean F1-score of 0.99 on the CWRU dataset, significantly outperforming baseline models-CNN, LSTM, RNN, GRU, and Bi-LSTM (F1-scores: 0.94-0.97, p<0.05, Wilcoxon signed-rank test for Bi-LSTM: p<0.05) -- particularly in challenging classes (e.g., Class 8: 0.99 vs. 0.71 for Bi-LSTM). Visualization of dataset characteristics reveals that datasets with amplified vibration patterns and diverse fault dynamics enhance generalization. This framework provides a robust solution for fault diagnosis under varying conditions, offering insights into dataset impacts on model performance.

Problem

• Classify machinery faults under varying operating conditions
• Convert time-series vibration data into graph representation
• Enhance fault diagnosis robustness with ensemble graph autoencoder

Innovation

• Graph Autoencoder with transformer-based encoders
• Dynamic Time Warping for graph edge definition
• Ensemble classifier for robust fault diagnosis