一种基于时空深度学习的机组组合加速求解方法及系统
By using a generator combination prediction model that alternately integrates GCN and Informer networks, the problems of low computational complexity and low iteration efficiency in generator combination problems in power systems are solved, enabling efficient and accurate generator start-up and shutdown decisions and grid dispatch.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHAANXI UNIV OF SCI & TECH
- Filing Date
- 2025-11-28
- Publication Date
- 2026-07-17
AI Technical Summary
Existing power system unit combination problems cannot simultaneously achieve high accuracy in time dynamic modeling, spatial structure representation capability, and high computational efficiency. Traditional methods have high computational complexity or low iteration efficiency, which cannot meet the needs of power grid dispatching.
A unit combination prediction model based on alternating fusion of graph convolutional network (GCN) and Informer network is adopted. By constructing the power grid topology, spatial features are extracted, and temporal features are captured by combining sparse self-attention mechanism. A constraint perception mechanism and dynamic penalty strategy are introduced to optimize unit start-up and shutdown decisions.
It improves the computational efficiency and decision-making accuracy of power system unit combination optimization, meets power grid security constraints, and provides an efficient dispatching scheme.
Smart Images
Figure CN121440799B_ABST