# Multi-Level Temporal Graph Networks with Local-Global Fusion for Industrial Fault Diagnosis

arXiv: https://arxiv.org/abs/2604.18765 · HTML: https://mlsanity.com/paper/2604.18765/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 41 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Builds on (most cited first, 9 of 9)

- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://mlsanity.com/paper/1606.09375/) (2016)
- [Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting](https://mlsanity.com/paper/1709.04875/) (2017)
- [Hierarchical Graph Representation Learning with Differentiable Pooling](https://mlsanity.com/paper/1806.08804/) (2018)
- [Multivariate Time-series Anomaly Detection via Graph Attention Network](https://mlsanity.com/paper/2009.02040/) (2020)
- [Data-Driven Fault Diagnosis Analysis and Open-Set Classification of Time-Series Data](https://mlsanity.com/paper/2009.04756/) (2020)
- [Graph neural network-based fault diagnosis: a review](https://mlsanity.com/paper/2111.08185/) (2021)
- [Graph Neural Networks with Trainable Adjacency Matrices for Fault Diagnosis on Multivariate Sensor Data](https://mlsanity.com/paper/2210.11164/) (2022)
- [Enhanced CNN with Global Features for Fault Diagnosis of Complex Chemical Processes](https://mlsanity.com/paper/2210.01727/) (2022)

## Abstract (arXiv)

Fault detection and diagnosis are critical for the optimal and safe operation of industrial processes. The correlations among sensors often display non-Euclidean structures where graph neural networks (GNNs) are widely used therein. However, for large-scale systems, local, global, and dynamic relations extensively exist among sensors, and traditional GNNs often overlook such complex and multi-level structures for various problems including the fault diagnosis. To address this issue, we propose a structure-aware multi-level temporal graph network with local-global feature fusion for industrial fault diagnosis. First, a correlation graph is dynamically constructed using Pearson correlation coefficients to capture relationships among process variables. Then, temporal features are extracted through long short-term memory (LSTM)-based encoder, whereas the spatial dependencies among sensors are learned by graph convolution layers. A multi-level pooling mechanism is used to gradually coarsen and learn meaningful graph structures, to capture higher-level patterns while keeping important fault related details. Finally, a fusion step is applied to combine both detailed local features and overall global patterns before the final prediction. Experimental evaluations on the Tennessee Eastman process (TEP) demonstrate that the proposed model achieves superior fault diagnosis performance, particularly for complex fault scenarios, outperforming various baseline methods.