基于电力大数据的配电网故障预测方法及系统
By constructing an offline trajectory dataset and nearest-neighbor state graph based on power big data, and performing structural entropy minimization partitioning, the problem of multi-source data integration for distribution network fault prediction was solved, achieving accurate prediction and stability improvement of distribution network faults.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LIAOYANG POWER SUPPLY COMPANY OF STATE GRID LIAONING ELECTRIC POWER SUPPLY
- Filing Date
- 2025-10-27
- Publication Date
- 2026-07-17
AI Technical Summary
Existing fault prediction methods for distribution networks rely on manual inspections and monitoring of single devices, making it difficult to integrate multi-source power data. This results in incomplete and inaccurate fault predictions, failing to meet the needs for accurate fault prediction and effective management of distribution networks.
Based on power big data, by constructing an offline trajectory dataset, a nearest neighbor state graph, and a structural entropy minimization partition, we obtain high-level, mid-level, and low-level sub-partition trees, perform hierarchical diffusion prediction modeling, construct a distribution network fault predictor, and introduce a structural entropy term for loss analysis.
It has improved the accuracy and reliability of power distribution network fault prediction, met the needs of accurate fault assessment and effective control, and enhanced the stability of prediction results.
Smart Images
Figure CN121350576B_ABST