Real-Time Image-Identification-Based Anti-Manmade Misoperation System for Substations

A novel antimisoperation system is proposed to prevent manmade misoperations using real-time signboard image identification technology. A portable video capture device is employed to capture real-time images of the current working bay´s substation signboard. These images, with the aid of image identification, are used to determine whether the operations staff has accessed a wrong bay. The framework and workflow of the system are provided, and the implementation of the signboard image border detection, character partition, and image-matching methods are also introduced in detail. The effectiveness and engineering applicability of the proposed system are demonstrated through lab results and in-field tests. The proposed system can be used in conjunction with an active five-prevention system to further improve the validity and safety of substation operations.