2021 International Conference on Sensing, Measurement and Data Analytics in the era of Artificial Intelligence (ICSMD2021)

83 / 2021-07-21 09:06:47

Rotating Machinery Fault Diagnosis Based on Spatial-Temporal GCN

Multi-sensor can provide more comprehensive and accurate information for mechanical fault diagnosis. Aiming at the weak ability of traditional artificial intelligence (AI) models to model multi-sensor signals, a method of fault diagnosis is proposed based on Spatial-Temporal Graph Convolution Network (ST-GCN) in this paper. The multi-sensor data is firstly modeled as a spatial-temporal graph. The graph topology, i.e. the relationship between different sensors is established adaptively using node features. Then, an ST-GCN model is designed to learn the hierarchical features both in the spatial and temporal dimension. Finally, the fault type is inferred by a softmax classifier based on the entire graph representation. The experiment results on the bearing and gearbox show that the proposed method can extract the information of multi-sensor effectively and contribute to the fault diagnosis of rotating machinery.