一种基于电力系统可观性的深度强化学习PMU配置方法
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SICHUAN UNIV
- Filing Date
- 2023-08-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to efficiently and economically address the PMU configuration problem in power systems, particularly in large-scale power systems where achieving system observability with the minimum number of PMUs is challenging. Furthermore, existing methods are computationally intensive and time-consuming, failing to meet the requirements for rapid transient voltage identification.
We employ a deep reinforcement learning approach based on power system observability, using a graph convolutional neural network (GCN-DDPG) for power management unit (PMU) configuration. We optimize PMU placement through a reward mechanism and combine numerical stability, anti-interference, and risk resilience to quantify system observability. We then construct a Markov decision process for interactive solution.
It enables efficient and economical configuration of PMUs in large-scale power systems, improves system observability and immunity, provides quantitative observability assessment, and reduces computational complexity and time cost.
Smart Images
Figure CN116992774B_ABST