智能综合配电箱的边缘侧自愈控制与故障隔离恢复方法
By collecting and analyzing bus voltage, total current, and branch current in the distribution box, and using a time convolutional network model and preset rules to identify event types, a recovery stability index and a relationship matrix are generated to optimize the recovery strategy. This solves the problems of low recovery efficiency and superimposed impact risk under bus voltage disturbance in the existing technology, and achieves efficient and reliable self-healing control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SUZHOU SUTUO COMM TECH
- Filing Date
- 2026-03-27
- Publication Date
- 2026-07-17
AI Technical Summary
When faced with bus voltage disturbances, existing distribution boxes have difficulty distinguishing between different scenarios such as voltage collapse, instantaneous power loss and branch faults. During the recovery process, they lack real-time determination of the mutual exclusion relationship of the impact and the dependence relationship of the business, resulting in low recovery efficiency and easy occurrence of secondary disturbances.
By collecting bus voltage, total current and branch current, a disturbance event window is established. The event type is identified using a time convolutional network model and preset rules. A recovery stability index and relationship matrix are generated, recovery candidate sequences are optimized, multi-level observation and judgment are performed, and the recovery strategy is dynamically adjusted.
It enables accurate identification and differentiated recovery control of different types of events, avoids the risk of overlapping impacts during the recovery process, and improves the reliability and intelligence level of power supply to communication sites.
Smart Images
Figure CN121923374B_ABST