Integration and Collaboration Mechanism of Power Operation Safety Monitoring System with Multi-Technology Integration

Le résumé fourni par la source

The power operation safety monitoring system is faced with the following problems: isolated multi-source data, poor real-time early warning. This paper proposes the integrated and cooperative mechanism of power operation safety monitoring system composed of multiple technologies. The system adopts the three-layer collaborative architecture of "end-edge-cloud," the terminal layer integrates UWB/IMU fusion positioning, panoramic vision, wearable physiological monitoring data, and environmental sensors to reach the accurate acquisition of multi-source data, the edge layer deploys heterogeneous data fusion and event-level correlation algorithms to reach millisecond-level event response and the preliminary identification of events through spatiotemporal alignment, attention mechanism weighted fusion and Kalman filtering, and the cloud platform layer build an event-driven collaborative mechanism and digital twin, realize the message linkage between subsystems based on the MQTT protocol, and cross-verify multi-source evidence to Experimental results show that the average early warning processing time of the system is 2.7~3.8 seconds, the average false alarm rate of the system is 1.3%, the false alarm times are 0~2, and the number of false alarms is 0~2; the daily active detection rate is 96.1%, which greatly improves the intelligence level and collaborative efficiency of power operation safety management.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.