To address the complex inspection scenarios at a State Grid 220kV substation, a collaborative inspection system was deployed, incorporating wheeled inspection robots, rail-mounted inspection robots, and fixed cameras. This system, combined with technologies such as digital twins, deep learning, and edge computing, created an online intelligent inspection system for the substation. The system possesses capabilities for equipment status perception, defect detection, proactive early warning, and risk management, effectively reducing the workload of maintenance personnel, improving inspection efficiency and intelligence, meeting the application requirements for unmanned, digitalized, and intelligent substation inspections, and achieving intelligent and coordinated operation of the entire system.