内蒙古电力技术2025，Vol.43Issue(6)：12-21,10.DOI:10.19929/j.cnki.nmgdljs.2025.0071
基于SAC算法的主动配电网源-荷-储协同优化方法
Collaborative Optimization Method for Source-Load-Storage in Active Distribution Network Based on SAC Algorithm
摘要
Abstract
In active distribution networks,high penetration of distributed energy output significantly increases system volatility due to their inherent fluctuation and randomness.In addition,the bidirectional power interaction mechanism of emerging loads and energy storage systems increases regulatory complexity while enhancing operational flexibility.To address these challenges,this paper proposes a source load-storage-collaborative optimization method based on the soft actor-critic(SAC)algorithm.Using the maximum entropy framework of SAC,the proposed method improves the robustness of system operation while supporting multiple application scenarios,including frequency regulation,peak shaving,valley filling,and demand response.Case analysis on the IEEE 33-node system demonstrates that the method not only enhances the flexibility and economic efficiency of active distribution networks but also improves the operational stability and reliability,showing greater adaptability and practical value compared with conventional optimization methods.关键词
主动配电网/源-荷-储协同优化/深度强化学习/软演员-评论家算法/鲁棒性Key words
active distribution network/source-load-storage collaborative optimization/deep reinforcement learning/soft actor-critic algorithm/robustness分类
信息技术与安全科学引用本文复制引用
WANG Yanni,MA Hengrui,YANG Yan,WANG Bo,FENG Ziyi,YUAN Aotian..基于SAC算法的主动配电网源-荷-储协同优化方法[J].内蒙古电力技术,2025,43(6):12-21,10.基金项目
教育部科技项目"青海高原零碳数据中心能源系统运行优化技术研究"(KFKT-25ERC-02) （KFKT-25ERC-02）
