电力系统的调度方法、装置及电子设备
By processing the initial solution and mathematical model through graph convolutional neural networks, constructing a bipartite graph of the objective, and optimizing integer variables to determine feasible solutions, the problem of low power system dispatch efficiency caused by branch and bound algorithms is solved, and more efficient power system dispatch is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA SOUTHERN POWER GRID COMPANY
- Filing Date
- 2024-12-31
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, when using branch and bound algorithms to solve the safety-constrained unit combination problem, the search space increases as the problem size grows, leading to low power system dispatch efficiency.
A graph convolutional neural network is used to process the initial solution and mathematical model, construct the target bipartite graph, and determine the feasible solution corresponding to the integer variables of the generator set through feature extraction and aggregation. The integer variables are optimized by combining a variable neighborhood search algorithm to improve the solution efficiency.
It effectively reduces the solution time, improves the solution efficiency of the safety-constrained unit combination problem, and enhances the dispatch efficiency of the power system.
Smart Images
Figure CN119944633B_ABST