Voltage optimization strategy for a distribution network based on deep reinforcement learning considering regionalization and imitation learning

被引：0

作者：

Li, Shidan ^{[1
]}

Li, Hang ^{[1
]}

Li, Guojie ^{[1
]}

Han, Bei ^{[1
]}

Xu, Jin ^{[1
]}

Li, Ling ^{[2
]}

Wang, Hongtao ^{[3
]}

机构：

[1] Key Laboratory of Control of Power Transmission and Conversion, Ministry of Education, Shanghai Jiao Tong University, Shanghai

[2] Shanghai PeiKe Technology Co., Ltd., Shanghai

[3] Jiaxing Power Supply Company, State Grid Zhejiang Electric Power Co., Ltd., Jiaxing

来源：

Dianli Xitong Baohu yu Kongzhi/Power System Protection and Control | 2024年 / 52卷 / 22期

关键词：

deep reinforcement learning; guidance signal; imitation learning; voltage optimization of distribution network; zoned loss reduction;

D O I：

10.19783/j.cnki.pspc.240117

中图分类号：

学科分类号：

摘要：

The current deep reinforcement learning (DRL) method has some issues with voltage optimization, such as challenging credit allocation and low exploration efficiency. These all lead to poor performance in model training speed and optimization effect. Considering regionalization and imitation learning, a voltage optimization strategy based on the guidance signal-based multi-agent deep deterministic policy gradient (GS-MADDPG) is proposed. First, electric vehicle (EV) clusters, distributed generation (DG) and reactive power regulators are taken as decision agents to build the reinforcement learning optimization model. Secondly, the external reward is decoupled through regionalization of the distribution network, and combined with imitation learning, an internal reward is introduced through the guidance signal to help agents search for optimization quickly. Finally, an example test is conducted on the improved IEEE 33-node distribution network. The results indicate that the proposed voltage optimization strategy has higher sample utilization, more stable convergence, and higher model training efficiency than the traditional DRL method, and improves the voltage optimization effect. © 2024 Power System Protection and Control Press. All rights reserved.

引用

页码：1 / 11

页数：10

共 31 条

[1] CHEN Guoping, DONG Yu, LIANG Zhifeng, Analysis and reflection on high-quality development of new energy with Chinese characteristics in energy transition, Proceedings of the CSEE, 40, 17, pp. 5493-5506, (2020)
[2] WANG Zhiwei, ZHANG Zhenyu, LI Zheng, Et al., Optimized DSSC configuration to enhance load-carrying capacity of the new energy grid with high permeability, Power System Protection and Control, 51, 6, pp. 179-187, (2023)
[3] ZHANG Xu, LI Yang, LIU Xiao, Et al., Distribution network planning of new AC/DC energy storage system with high penetration new energy, Southern Power System Technology, 16, 4, pp. 60-67, (2022)
[4] YANG Jinhai, WU Jiahui, WANG Haiyun, Et al., Dynamic security region analysis of power system under different penetration rate of new energy, Electric Power Construction, 43, 4, pp. 58-68, (2022)
[5] XIONG Zhuangzhuang, Two-stage robust optimization of dynamic reactive power optimization in active distribution network, (2019)
[6] WANG Yongjie, WU Wenchuan, ZHANG Boming, Et al., Active and reactive power coordinated robust optimization for active distribution networks, Automation of Electric Power Systems, 40, 9, pp. 29-34, (2016)
[7] LE FLOCH C, BELLETTI F, MOURA S., Optimal charging of electric vehicles for load shaping: a dual-splitting framework with explicit convergence bounds, IEEE Transactions on Transportation Electrification, 2, 2, pp. 190-199, (2016)
[8] SCHULLER A, FLATH C M, GOTTWALT S., Quantifying load flexibility of electric vehicles for renewable energy integration, Applied Energy, 151, pp. 335-344, (2015)
[9] ZHENG Neng, DING Xiaoqun, GUAN Zhicheng, Et al., Coordinated optimization of active power and reactive power in distribution network based on scenario method, Power System Technology, 43, 5, pp. 1640-1651, (2019)
[10] GUO Qingyuan, WU Jiekang, MO Chao, Et al., A model for multi-objective coordination optimization of voltage and reactive power in distribution networks based on mixed integer second-order cone programming, Proceedings of the CSEE, 38, 5, pp. 1385-1396, (2018)

← 1 2 3 4 →