Knowledge Graph and Deep Learning-Based Fault Diagnosis for Urban Rail Signal Systems

This paper proposes a fault diagnosis method for urban rail signal systems that integrates knowledge graphs and deep learning techniques. By combining Graph Neural Networks (GNN) and the BILSTM+CRF model, the method efficiently extracts and analyzes fault data, thereby improving the accuracy of fault diagnosis. The knowledge graph represents various fault types and corresponding solutions, while the deep learning models optimize the inference process by capturing complex relationships between data points. The proposed approach enables real-time fault detection and solution provision, thereby enhancing the operational efficiency and reliability of urban rail transit systems. Through the integration of BILSTM+CRF and GNN, the method not only significantly improves diagnostic accuracy but also supports rapid fault resolution, reducing handling time and operational costs.