Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 最新索引

基于多智能体深度强化学习的配电网电压分区协同控制

来源:电工电气发布时间:2025-03-03 10:03 浏览次数:6

基于多智能体深度强化学习的配电网电压分区协同控制

尹昕,曹丽鹏,王玉森
(国网山西省电力公司长治供电公司,山西 长治 046000)
 
    摘 要:为充分利用配电网中多类型调节资源的调节能力,提高新能源高比例接入下配电网的分区自治能力,提出了一种基于多智能体深度强化学习(MADRL)的配电网电压多分区协同控制策略。采用多智能体对配电网分区协同电压控制问题进行建模,并运用改进的反事实多智能体柔性动作-评价(COMASAC)深度强化学习模型求解配电网分区协同电压控制问题。通过实际配电网典型日运行数据的仿真算例,验证了所提基于多智能体深度强化学习方法在提高配电网电压稳定性与降低网络损耗方面的优势。
    关键词: 多智能体;深度强化学习;配电网电压;分区协同控制;网络损耗
    中图分类号:TM714.2 ;TM734     文献标识码:A     文章编号:1007-3175(2025)02-0063-09
 
Partition Cooperative Control of Distribution Network Voltage
Based on Multi-Agent Deep Reinforcement Learning
 
YIN Xin, CAO Li-peng, WANG Yu-sen
(State Grid Shanxi Electric Power Company Changzhi Power Supply Company, Changzhi 046000, China)
 
    Abstract: In order to fully utilize the regulation capability of multiple types of regulation resources in the distribution network and improve the zonal autonomy capability of the distribution network under the high proportion of new energy access, this paper proposes a multi-zonal cooperative control strategy for distribution network voltage based on multi-agent deep reinforcement learning (MADRL). The problem of partition cooperative voltage control in distribution network is modeled using a multi-agent approach. Subsequently, an improved counterfactual multi-agent soft actor-critic (COMASAC) deep reinforcement learning model is applied to solve the zonal cooperative voltage control problem in distribution networks.Finally, simulation examples using typical day operational data from actual distribution networks demonstrate the advantages of the proposed multi-agent deep reinforcement learning method in improving voltage stability and reducing network losses in distribution networks.
    Key words: multi-agent; deep reinforcement learning; distribution network voltage; partition cooperative control; network loss
 
参考文献
[1] 陈晓婷,何元明,王菁,等. 新型电力系统背景下配电网分布式光伏可接入容量研究[J] . 电工技术,2023(17) :44-47.
[2] 刘玉玲,张峰,张刚,等. 考虑分布式电源的配电网灾中-灾后两阶段协同韧性恢复决策[J]. 电力系统自动化,2024,48(2) :48-59.
[3] 闫丽梅,丁泽华. 基于谱聚类的主动配电网多时间尺度无功优化策略[J]. 浙江电力,2024,43(2) :58-68.
[4] 杜红卫,尉同正,夏栋,等. 基于集群动态划分的配电网无功电压自律-协同控制[J] . 电力系统自动化,2024,48(10) :171-181.
[5] GAO Y, YU N.Deep Reinforcement Learning in Power Distribution Systems: Overview, Challenges, and Opportunities[C]//IEEE Power and Energy Society Innovative Smart Grid Technologies Conference,2021.
[6] 吴文传,张伯明,孙宏斌,等. 主动配电网能量管理与分布式资源集群控制[J] . 电力系统自动化,2020,44(9) :111-118.
[7] 路小俊,吴在军,李培帅,等. 面向光伏集群的配电网模型-数据联合驱动无功/电压控制[J]. 电力系统自动化,2024,48(9) :97-106.
[8] 朱鹏程,刘曌煜,孙可,等. 基于多分块交替方向乘子法的蜂巢状配电网分布式优化调度[J] . 中国电力,2023,56(6) :90-100.
[9] 赵冬梅,王浩翔,陶然. 基于改进交替方向乘子法的输配电网分散协调鲁棒优化调度模型[J] . 电网技术,2023,47(3) :1138-1150.
[10] 寇凌峰,吴鸣,李洋,等. 主动配电网分布式有功无功优化调控方法[J] . 中国电机工程学报,2020,40(6) :1856-1864.
[11] 姜涛,张东辉,李雪,等. 含分布式光伏的主动配电网电压分布式优化控制[J] . 电力自动化设备, 2021,41(9) :102-109.
[12] LI Peishuai, ZHANG Cuo, WU Zaijun, et al.Distributed Adaptive Robust Voltage/VAR Control with Network Partition in Active Distribution Networks[J].IEEE Transactions on Smart Grid,2020,11(3) :2245-2256.
[13] 倪爽,崔承刚,杨宁,等. 基于深度强化学习的配电网多时间尺度在线无功优化[J] . 电力系统自动化,2021,45(10) :77-85.
[14] LIU Hangyue, ZHANG Cuo, CHAI Qingmian, et al.Robust Regional Coordination of Inverter-Based Volt/Var Control via Multi-Agent Deep Reinforcement Learning[J].IEEE Transactions on Smart Grid,2021,12(6) :5420-5433.
[15] 胡丹尔,彭勇刚,韦巍,等. 多时间尺度的配电网深度强化学习无功优化策略[J] . 中国电机工程学报,2022,42(14) :5034-5044.
[16] 张晓,吴志,郑舒,等. 基于贝叶斯深度强化学习的主动配电网电压控制[J] . 电力系统自动化,2024,48(20) :81-90.
[17] LI S , HU W , DI Cao , et al . A Multiagent Deep Reinforcement Learning Based Approach for the Optimization of Transformer Life Using Coordinated Electric Vehicles[J].IEEE Transactions on Industrial Iinformatics,2022,18(11) :7639-7652.
[18] CAO Di, HU Weihao, ZHAO Junbo, et al.A Multi-Agent Deep Reinforcement Learning Based Voltage Regulation Using Coordinated PV Inverters[J].IEEE Transactions on Power Systems,2020,35(5) :4120-4123.
[19] FOERSTER J, FARQUHAR G, AFOURAS T, et al.Counterfactual Multi-Agent Policy Gradients[C]//Proceeding of the AAAI Conference on Artificial Intelligence,2018.
[20] 王尔申,刘帆,宏晨,等. 基于 MASAC 的无人机集群对抗博弈方法[J] . 中国科学(信息科学),2022,52(12) :2254-2269.