Voltage Optimal Control of Distribution Network Based on Deep Reinforcement Learning

被引：0

作者：

Quan H. ^{[1
]}

Peng X. ^{[1
]}

Liu H. ^{[1
]}

Zhou P. ^{[1
]}

Wu Z. ^{[1
]}

Su H. ^{[1
]}

机构：

[1] School of Automation, Guangdong University of Technology, Guangdong Province, Guangzhou

来源：

Dianwang Jishu/Power System Technology | 2023年 / 47卷 / 05期

基金：

中国国家自然科学基金;

关键词：

data-driven; Markov decision process; multi-agent deep reinforcement learning; photovoltaic inverter; voltage optimal control of distribution network;

D O I：

10.13335/j.1000-3673.pst.2022.1472

中图分类号：

学科分类号：

摘要：

The access of large-scale distributed generation makes the distribution network voltage optimization control strategy different from the traditional distribution network. Aiming at the problem of lack of coordination between photovoltaic inverter voltage regulation in local control, this paper proposes a voltage control method of distribution network based on multi-agent deep reinforcement learning. Firstly, the partially observable Markov decision process is designed according to the voltage control model, and then the multi-agent twin delayed deep deterministic policy gradient algorithm is used to solve it, and the coordinated reactive power control of PV inverters is realized according to the framework of centralized training and decentralized execution. The method can make intelligent decisions on the amount of reactive power regulation for each inverter, generating the real-time voltage control strategy according to the random change of generation and load. It has good real-time performance and control economy. Finally, the effectiveness of the proposed method is verified by simulation examples. © 2023 Power System Technology Press. All rights reserved.

引用

页码：2029 / 2038

页数：9

共 28 条

[11]

Qi LI, QIAO Ying, ZHANG Yujing, Continuous reactive power optimization of distribution network using deep reinforcement learning[J], Power System Technology, 44, 4, pp. 1473-1480, (2020)

[12]

NI Shuang, CUI Chenggang, YANG Ning, Multi-time-scale online optimization for reactive power of distribution network based on deep reinforcement learning[J], Automation of Electric Power Systems, 45, 10, pp. 77-85, (2021)

[13]

TOUBEAU J F, ZAD B B，, HUPEZ M, Deep reinforcement learning-based voltage control to deal with model uncertainties in distribution networks[J], Energies, 13, 15, (2020)

[14]

DENG Qingtang, HU Daner, CAI Tiantian, Reactive power optimization strategy of distribution network based on multiagent deep reinforcement learning[J], Advanced Technology of Electrical Engineering and Energy, 41, 2, pp. 10-20, (2022)

[15]

YE Yujian, YUAN Quan, TANG Yi, Decentralized coordination parameters optimization in microgrids mitigating demand response synchronization effect of flexible loads[J], Proceedings of the CSEE, 42, 5, pp. 1748-1759, (2022)

[16]

KABIRI R, HOLMES D G，, MCGRATH B P, LV grid voltage regulation using transformer electronic tap changing，with PV inverter reactive power injection[J], IEEE Journal of Emerging and Selected Topics in Power Electronics, 3, 4, pp. 1182-1192, (2015)

[17]

WANG Ke, YAO Jianguo, YU Peiyao, Architecture and key technologies of intelligent decision-making of power grid look-ahead dispatch based on deep reinforcement learning[J/OL], Proceedings of the CSEE, pp. 1-9, (2022)

[18]

LOWE R, WU Yi, TAMAR A, Multi-agent actor-critic for mixed cooperative-competitive environments[C], Proceedings of the 31st International Conference on Neural Information Processing Systems, pp. 6382-6393, (2017)

[19]

ZHANG Jinyuan, PU Tianjiao, LI Ye, Research on optimal dispatch strategy of distributed generators based on multi-agent deep reinforcement learning[J], Power System Technology, 46, 9, pp. 3496-3504, (2022)

[20]

FUJIMOTO S, VAN HOOF H, MEGER D．, Addressing function approximation error in actor-critic methods[C], Proceedings of the 35th International Conference on Machine Learning, pp. 1582-1591, (2018)

← 1 2 3 →