DRL-Based Active and Reactive Power Coordinated Control for Flexible Interconnected Power Distribution Systems with Soft Open Points

作者：Wenlin Wang, Yongbiao Ling, Dongyue Zhang, Jie Zhang, Zhengli Hu, Shanjing Wan, Jixuan Wang, Qibing Wang, Jin Huang, Xiaodong Yang · 年份：2023 · DOI：10.1109/epee59859.2023.10351973 · 被引用次数：1 · 研究领域：Optimal Power Flow Distribution、Microgrid Control and Optimization、Smart Grid Energy Management

This paper integrates deep reinforcement learning (DRL) with voltage regulation, system losses and photovoltaic randomness for interconnected distribution networks. First, the soft open point (SOP) model and the system power flow constraints model are formed, then the decision-making of SOP action is established as a Markov decision process, and a multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed to find the optimal active power transmission and reactive power compensation of SOP, and regard them as actions to find out the control scheme with the minimum system losses and voltage deviation. The results show that this model can effectively solve the above mentioned aspects and maximize the economy of the system.