A digital twin-based simulation and remote collaboration platform for electrical maintenance operations of rail transit vehicles

With the increasing complexity of rail transit systems, ensuring efficient and accurate electrical maintenance has become a critical challenge. Traditional maintenance methods lack real-time visibility, simulation capabilities, and collaborative tools for fault diagnosis. This paper proposes a digital twin-based simulation and remote collaboration platform designed to support electrical maintenance of rail vehicles. The system integrates a high-fidelity digital twin engine, a fault simulation model based on differential-algebraic equations, and a web-based interface for multi-user collaboration. Experiments were conducted using real and synthetic datasets involving 12 subsystems and over 8,000 time-series samples. Results show a fault detection accuracy of 95.2% and an average latency of 1.34 seconds, along with high user satisfaction ratings. The platform enables real-time anomaly detection, interactive diagnostics, and effective remote support. In conclusion, the proposed system provides a scalable and intelligent solution for rail transit maintenance and holds promise for future integration with predictive analytics and automated digital twin generation.