GCG: Graph Convolutional network and gated recurrent unit method for high-speed train axle temperature forecasting

Abstract

Nowadays, with the growing scale of high-speed railways and the increasing number of high-speed trains, the research on train equipment fault diagnosis and health management becomes more and more significant. Bearings are parts which are prone to be the failure equipment on high-speed trains. The temperature of a faulty bearing will increase suddenly during the working process, which may lead to potential accidents. So the axle temperature prediction has become a key research direction. This paper proposes a new organization form of axle temperature data, which connects axle temperature measurement points according to their locations so as to form a graph. Then, based on the Graph Convolutional Network (GCN) and Gated Recurrent Units (GRU) models, a new framework named GCG which combines the GCN and GRU is proposed to extract features and predict axle temperature. Finally, the experiments are conducted based on actual data. The results show that the prediction accuracy and tracking sensitivity are better than other advanced methods.