Power grid fault intelligent rescue forward scheduling system based on guangming big model

CN122155183APending Publication Date: 2026-06-05国网福建省电力有限公司营销服务中心
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
国网福建省电力有限公司营销服务中心
Filing Date
2026-02-09
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing power grid fault detection and emergency repair scheduling methods are inadequate in terms of fault prediction capability, resource allocation efficiency, and dynamic response. In particular, they cannot adapt to the emergency repair needs in complex scenarios when there are multiple faults or related system faults, resulting in scheduling delays or uneven resource allocation.

Method used

A forward-looking intelligent power grid fault repair system based on the Guangming large-scale model is adopted. Through multi-modal data perception and fusion, spatiotemporal neural network evolution prediction, and multi-scenario simulation digital twin modules, it achieves comprehensive perception, accurate prediction, and efficient repair of power grid faults. The system includes a multi-modal data perception and fusion module, a spatiotemporal neural network evolution prediction module, and a multi-scenario simulation digital twin module. Combined with closed-loop feedback and model self-optimization mechanisms, it enables dynamic analysis and optimized resource scheduling of power grid faults.

Benefits of technology

It enables accurate prediction of power grid faults and global optimal scheduling of emergency repair resources, improving the reliability and stability of power grid operation. It can effectively cope with multi-point concurrent faults in complex power grid environments and improve the efficiency of emergency repair and the real-time nature of resource allocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122155183A_ABST
    Figure CN122155183A_ABST
Patent Text Reader

Abstract

The application discloses a power grid fault intelligent repair forward scheduling system based on a Guangming big model, and relates to the technical fields of artificial intelligence and power dispatching. The system comprises a multi-modal data sensing and fusion module, which is used for collecting and standardizing multi-dimensional power grid operation data to generate real-time state data flow; a space-time graph neural network evolution prediction module, which deduces a plurality of possible fault evolution paths based on initial fault information and global state to form a probabilistic future state set; a multi-scenario deduction digital twin module, which simulates each path to quantify expected system loss; and a closed-loop feedback and model self-optimization module, which compares actual results with prediction and simulation results to optimize model parameters. Based on the Guangming big model, the system introduces advanced intelligent prediction algorithms and dynamic resource allocation mechanisms to realize comprehensive sensing, accurate prediction and efficient repair of power grid faults, thereby improving the reliability and stability of power grid operation.
Need to check novelty before this filing date? Find Prior Art