Power distribution network fault restoration method based on monte carlo tree search and neural network

By constructing a repair guidance model based on Monte Carlo tree search and neural network reinforcement learning framework, the problem of existing technologies being unable to adapt to complex real-world situations is solved, and effective and economical fault repair guidance for distribution network fault recovery is achieved.

CN115000954BActive Publication Date: 2026-06-19SHENZHEN POWER SUPPLY BUREAU +1
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHENZHEN POWER SUPPLY BUREAU
Filing Date
2022-07-07
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing power distribution network fault recovery methods are difficult to adapt effectively to various unexpected events and compromises when faced with complex real-world situations, resulting in poor practical application results.

Method used

A reinforcement learning framework based on Monte Carlo tree search and neural networks is adopted to represent the fault recovery process through the active power flow matrix and the regional repair time matrix, and a repair guidance model is constructed to adjust the action guidance plan in real time to adapt to the actual situation.

Benefits of technology

It enables the provision of effective fault repair guidance in actual maintenance processes facing various unexpected events and compromises, thereby improving the economic efficiency and applicability of fault recovery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115000954B_ABST
    Figure CN115000954B_ABST
Patent Text Reader

Abstract

This invention relates to the field of distribution network fault repair technology, specifically to a distribution network fault recovery method based on Monte Carlo tree search and neural networks, comprising the following steps: Step 1, obtaining the active power flow matrix and regional repair time matrix of the distribution network, and characterizing the distribution network fault recovery process using the active power flow matrix and regional repair time matrix; wherein, the regional repair time matrix is ​​obtained by processing the historical repair times of each regional feeder of the distribution network; Step 2, confirming the reference values, constraints, and state space of the action space for fault recovery; the constraints are resource constraints for fault repair, the state space is a set of distribution network states, and the action space is a set of repair decisions. This method enables maintenance guidance schemes to be applied to actual maintenance processes full of various unexpected events and compromises.
Need to check novelty before this filing date? Find Prior Art