电力系统调度决策:一种示教学习辅助加速的安全强化学习方法
Power System Dispatch:An Accelerated Safe Reinforcement Learning Approach by Incorporating Learning From Demonstration
仪忠凯 1梁寿愚 2王伟 1蒋蔚 1杨程 1辛焱1
作者信息
- 1. 阿里巴巴达摩院(杭州)科技有限公司,浙江省 杭州市 310000
- 2. 中国南方电网有限责任公司,广东省 广州市 510000
- 折叠
摘要
随着可再生能源占比攀升和电网运行环境愈加复杂,亟需构建知识-数据融合的新型电力系统调度模式.鉴于此,首先采用模仿学习的方法对专家知识库中的案例进行拟合,构建示教学习模型,为电力系统调度运行提供示教调度引导指令.在此基础上,提出一种基于示教学习辅助加速的安全强化学习方法,能用于支撑电力系统实时快速决策.通过引入示教学习辅助加速机制,所提方法的收敛速度显著加快,调度策略迅速趋优,降低系统运行成本,缓解潮流越限风险.案例分析验证所提方法在提升强化学习收敛效率和促进电力系统安全经济运行方面的优势.
Abstract
With the growing penetration of renewable energy generation and the increasing complexity of the power system environment,it is necessary to formulate a hybrid knowledge-data-driven dispatch strategy for modern power systems.In light of this,an imitation learning approach is proposed to exploit the expert knowledge,which provides demonstrations for power system economic dispatch strategy using neural networks.Furthermore,an accelerated safe reinforcement learning approach is proposed by incorporating learning from demonstration,which can make fast decisions in real-time operation.By incorporating the learning from demonstration approach,the algorithm convergence speed is significantly accelerated,the dispatch strategy is optimized,the operation cost is reduced,and the power flow violation risk is alleviated.Numerical simulation results verify the advantages of the proposed approach in improving the convergence efficiency of the reinforcement learning algorithm and promoting the security and economy of the power systems.
关键词
电力系统/经济调度/安全强化学习/示教学习Key words
power systems/economic dispatch/safe reinforcement learning/learning from demonstration引用本文复制引用
基金项目
国家重点研发计划项目(2022YFB2403500)
出版年
2024
国家科技期刊平台
NSTL
万方数据