Robust Optimal Configuration of Distributed Energy Storage Considering Distribution Network Topology Reconfiguration

ZHANG Yiming; SHI Linjun; WU Feng; WU Chenyu; YAO Shun

doi:10.7500/AEPS20260212001

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Robust Optimal Configuration of Distributed Energy Storage Considering Distribution Network Topology Reconfiguration

更新时间：2026-09-15

- Robust Optimal Configuration of Distributed Energy Storage Considering Distribution Network Topology Reconfiguration
- Automation of Electric Power Systems Pages: 1-23(2026)
- 作者机构：
  
  1.河海大学电气与动力工程学院，江苏省南京市 211100
  2.国网江苏省电力有限公司淮安供电分公司，江苏省淮安市 223001
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(U23B20140)
- DOI：10.7500/AEPS20260212001
  CLC：
- Received：12 February 2026，
  
  Revised：2026-05-13，
  
  Online First：02 September 2026，
- 稿件说明：
移动端阅览
张益铭,史林军,吴峰等.考虑配电网拓扑重构的分布式储能鲁棒优化配置[J/OL].电力系统自动化, http://dx.doi.org/10.7500/AEPS20260212001.

ZHANG Yiming,SHI Linjun,WU Feng,et al.Robust Optimal Configuration of Distributed Energy Storage Considering Distribution Network Topology Reconfiguration[J/OL].Automation of Electric Power Systems, http://dx.doi.org/10.7500/AEPS20260212001.
张益铭,史林军,吴峰等.考虑配电网拓扑重构的分布式储能鲁棒优化配置[J/OL].电力系统自动化, http://dx.doi.org/10.7500/AEPS20260212001. DOI：

ZHANG Yiming,SHI Linjun,WU Feng,et al.Robust Optimal Configuration of Distributed Energy Storage Considering Distribution Network Topology Reconfiguration[J/OL].Automation of Electric Power Systems, http://dx.doi.org/10.7500/AEPS20260212001. DOI：

摘要

针对传统储能规划方法多依赖单一静态基准网架，难以适应配电网运行期间频繁的拓扑重构，且在面对长周期规划中的不确定性时缺乏海量场景高效推演能力的问题，提出一种基于配电网拓扑重构场景簇的分布式储能鲁棒优化配置方法。首先，构建面向泛化能力的多样化训练场景簇，并评估各节点的储能潜力与最大可安装容量，从而克服传统方法对单一拓扑的依赖。在此基础上，提出双层“决策-评估”闭环学习框架，上层构建以全生命周期成本最小化为目标函数的分布式储能选址定容决策模型，并采用图神经网络编码与指针网络解码结构及Beta参数化的“近端策略优化-图神经网络”序贯决策求解方法，有效缓解维度爆炸问题。下层基于分布鲁棒优化的评估与奖励反馈机制，构建以运行成本最小化为目标的分布式储能运行优化模型，并将求解结果反馈给上层决策模块。仿真结果表明，与仅在单一拓扑下的基准方法相比，所提方法在收敛速度、经济性与鲁棒性方面均具有显著优势。

Abstract

Aiming at the drawbacks of conventional energy storage planning methods

which mostly rely on a single static base grid structure

fail to adapt to frequent topological reconfiguration during distribution network operation

and lack the capability for efficient massive scenario deduction when addressing uncertainties in long-term planning

this paper proposes a robust optimal configuration method for distributed energy storage based on scenario clusters of distribution network topological reconfiguration. First

diversified training scenario clusters oriented to generalization capability are constructed

and the energy storage potential and the maximum installable capacity of each node are evaluated

so as to overcome the dependence of traditional methods on a single topology. On this basis

a bi-level “decision-evaluation” closed-loop learning framework is put forward. At the upper level

a decision model for the siting and sizing of distributed energy storage is established with the objective of minimizing the full-life-cycle cost. A sequential decision-making solution method of graph neural network encoding and pointer network decoding architecture

together with a Beta-parameterized proximal policy optimization-graph neural network

is adopted to effectively alleviate the curse of dimensionality. At the lower level

an evaluation and reward feedback mechanism based on distributionally robust optimization is developed

and an optimal operation model of distributed energy storage targeting the minimum operational cost is constructed. The solution results are fed back to the upper-level decision module. Simulation results verify that compared with the benchmark method implemented under a single topology only

the proposed method achieves prominent advantages in convergence speed

economic efficiency and robustness.

关键词

Keywords

references

应益强，王正风，吴旭，等 . 计及新能源随机特性的电网深度调峰多目标策略［J］. 电力系统保护与控制， 2020 ， 48 （ 6 ）： 34 - 42 ．

Ying Yiqiang ， Wang Zhengfeng ， Wu Xu ， et al ． Multi-objective strategy for deep peak shaving of power grid considering uncertainty of new energy ［J］. Power System Protection and Control ， 2020 ， 48 （ 6 ）： 34 - 42 ．

Hemmati R ， Saboori H . Emergence of hybrid energy storage systems in renewable energy and transport applications-a review ［J］. Renewable and Sustainable Energy Reviews ， 2016 ， 65 ： 11 - 23 ．

李更丰，孙少华，别朝红，等 . 面向新型电力系统弹性提升的储能优化配置与灵活调度研究综述［J］. 高电压技术， 2023 ， 49 （ 10 ）： 4084 - 4095 ．

Li Gengfeng ， Sun Shaohua ， Bie Zhaohong ， et al . Review on optimal configuration and flexible scheduling research of energy storage for resilience improvement of new power system ［J］. High Voltage Engineering ， 2023 ， 49 （ 10 ）： 4084 - 4095 ．

Jing Rui ， Wang Ming ， Zhang Zhi ， et al . Distributed or centralized？ Designing district-level urban energy systems by a hierarchical approach considering demand uncertainties ［J］. Applied Energy ， 2019 ， 252 ： 113424 ．

白雪岩，樊艳芳，刘雨佳，等 . 考虑可靠性及灵活性的风光储虚拟电厂分层容量配置［J］. 电力系统保护与控制， 2022 ， 50 （ 8 ）： 11 - 24 ．

Bai Xueyan ， Fan Yanfang ， Liu Yujia ， et al . Wind power storage virtual power plant considering reliability and flexibility tiered capacity configuration ［J］. Power System Protection and Control ， 2022 ， 50 （ 8 ）： 11 - 24 ．

Song Y ， Zheng Y ， Liu T ， et al . A new formulation of distribution network reconfiguration for reducing the voltage volatility induced by distributed generation ［J］. IEEE Transactions on Power Systems ， 2020 ， 35 （ 1 ）： 496 - 507 ．

Huang Bo ， Zhao Tao ， Yue Ming ， et al . Bi-level adaptive storage expansion strategy for microgrids using deep reinforcement learning ［J］. IEEE Transactions on Smart Grid ， 2024 ， 15 （ 2 ）： 1362 - 1375 .

Chen Lei ， He Hao ， Jing Rui ， et al . Energy management in integrated energy system with electric vehicles as mobile energy storage： an approach using bi-level deep reinforcement learning ［J］. Energy ， 2024 ， 307 ： 132757 .

KundačIna O B ， Vidović P M ， Petković M R . Solving dynamic distribution network reconfiguration using deep reinforcement learning ［J］. Electrical Engineering ， 2022 ， 104 ： 1487 - 1501 .

Jo S ， Oh J Y ， Yoon Y T ， et al . Self-healing radial distribution network reconfiguration based on deep reinforcement learning ［J］. Results in Engineering ， 2024 ， 22 ： 102026 .

Arulkumaran K ， Deisenroth M P ， Brundage M ， et al . Deep reinforcement learning： a brief survey ［J］. IEEE Signal Processing Magazine ， 2017 ， 34 （ 6 ）： 26 - 38 .

Hedayatnia A ， Ghafourian J ， Sepehrzad R ， et al . Two-stage data-driven optimal energy management and dynamic real-time operation in networked microgrid based on a deep reinforcement learning approach ［J］. International Journal of Electrical Power & Energy Systems ， 2024 ， 160 ： 110142 .

Li Hao ， Li Gang ， Lie Taotao ， et al . Constrained large-scale real-time EV scheduling based on recurrent deep reinforcement learning ［J］. International Journal of Electrical Power & Energy Systems ， 2023 ， 144 ： 108603 .

Lowe R ， Wu Y ， Tamar A ， et al . Multi-agent actor-critic for mixed cooperative-competitive environments ［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems ， December 4-9 ， 2017 ， Red Hook， USA ： 6382 - 6393 .

Lee X Y ， Sarkar S ， Wang Y . A graph policy network approach for Volt-Var Control in power distribution systems ［J］. Applied Energy ， 2022 ， 323 ： 119530 .

Jacob R A ， Paul S ， Chowdhury S ， et al . Real-time outage management in active distribution networks using reinforcement learning over graphs ［J］. Nature Communications ， 2024 ， 15 ： 4766 .

张沛，陈玉鑫，王光华，等 . 基于图强化学习的配电网故障恢复决策［J］. 电力系统自动化， 2024 ， 48 （ 2 ）： 151 - 158 .

Zhang Pei ， Chen Yuxin ， Wang Guanghua ， et al . Distribution network fault recovery decision-making based on graph reinforcement learning ［J］. Automation of Electric Power Systems ， 2024 ， 48 （ 2 ）： 151 - 158 .

Jabr R A ． Distributionally robust CVaR constraints for power flow optimization ［J］. IEEE Transactions on Power Systems ， 2020 ， 35 （ 5 ）： 3764 - 3773 ．

Wei Wei ， Wang Yudong ， Huang Xu ， et al . Data-driven distributionally robust day-ahead dispatch for active distribution networks based on improved conditional generative adversarial network ［J］. Sustainable Energy， Grids and Networks ， 2024 ， 38 ： 101402 ．

程跃珂，王钰洁 . 考虑光伏出力的储能系统容量优化配置研究［J］. 太阳能学报， 2025 ， 46 （ 11 ）： 61 - 68 ．

Cheng Yueke ， Wang Yujie ． Research on capacity optimization configuration of energy storage system considering photovoltaic output ［J］. Acta Energiae Solaris Sinica ， 2025 ， 46 （ 11 ）： 61 - 68 ．

Dong Lei ， Li Jia ， Pu Tianjiao ， et al . Distributionally robust optimization model of active distribution network considering uncertainties of source and load ［J］. Journal of Modern Power Systems and Clean Energy ， 2019 ， 7 （ 6 ）： 1585 - 1595 ．

何为，高红均，王仁浚，等 . 考虑合环安全性的配电网重构运行优化方法［J/OL］. 电网技术［ 2025-12-14 ］. https：//doi.org/10.13335/j.1000-3673.pst.2025.1060 https://doi.org/10.13335/j.1000-3673.pst.2025.1060 .

He Wei ， Gao Hongjun ， Wang Renjun ， et al . Reconfiguration operation optimization method for distribution network considering loop closing safety ［J/OL］. Power System Technology ［ 2025-12-14 ］. https：//doi.org/10.13335/j.1000-3673.pst.2025.1060 https://doi.org/10.13335/j.1000-3673.pst.2025.1060 .

王勇飞，李晓飞，孙雨欣，等 . 基于核主成分分析与长短时记忆网络的水电机组监测预警［J］. 振动与冲击， 2024 ， 43 （ 24 ）： 287 - 294 ．

Wang Yongfei ， Li Xiaofei ， Sun Yuxin ， et al ． Monitoring and early warning of hydropower units based on kernel principal component analysis and long short-term memory network ［J］． Journal of Vibration and Shock ， 2024 ， 43 （ 24 ）： 287 - 294 ．

Zhao Chao ， Li Xiang ， Cai Yang . A power grid topology detection method based on edge graph attention neural network ［J］. Electric Power Systems Research ， 2025 ， 239 ： 111219 .

薛欣，朱天晨，孙庆赟，等．融合图神经网络的高效子图匹配算法［J］. 计算机研究与发展， 2025 ， 62 （ 3 ）： 694 - 708 ．

Xue Xin ， Zhu Tianchen ， Sun Qingyun ， et al . An efficient subgraph matching algorithm integrating graph neural networks ［J］. Journal of Computer Research and Development ， 2025 ， 62 （ 3 ）： 694 - 708 ．

Chou P W ， Maturana D ， Scherer S . Improving stochastic policy gradients in continuous control with deep reinforcement learning using the Beta distribution ［C］// Proceedings of the 34th International Conference on Machine Learning ， August 6-11 ， 2017 ， Sydney， Australia ： 834 - 843 ．

Kang H ， Jung S ， Jeong J ， et al ． A bi-level reinforcement learning model for optimal scheduling and planning of battery energy storage considering uncertainty in the energy-sharing community ［J］. Sustainable Cities and Society ， 2023 ， 94 ： 104538 ．

Ren Chao ， Wei Zhinong ， Zhou Yizhou ， et al . Distributionally robust CVaR optimization for resilient distribution system planning with consideration for long-term and short-term uncertainties ［J］. Reliability Engineering & System Safety ， 2024 ， 251 ： 110378 ．

李懿鑫，李正烁．考虑恶意攻击和随机性注入的配电网-聚合商双层灵活性评估方法［J/OL］. 中国电机工程学报［ 2025-12-29 ］. https：//link.cnki.net/urlid/11. 2107.TM.20250724.1433.006 https://link.cnki.net/urlid/11.2107.TM.20250724.1433.006 ．

Li Yixin ， Li Zhengshuo . A bi-level flexibility evaluation method for distribution network-aggregator considering malicious attacks and stochastic injection ［J/OL］. Proceedings of the CSEE ［ 2025-12-29 ］. https：//link.cnki.net/urlid/11.2107.TM.20250724.1433.006 https://link.cnki.net/urlid/11.2107.TM.20250724.1433.006 ．

Bobo L ， Venzke A ， Chatzivasileiadis S . Second-order cone relaxations of the optimal power flow for active distribution grids： comparison of methods ［J］. International Journal of Electrical Power & Energy Systems ， 2021 ， 127 ： 106625 ．

Khajeh H ， Parthasarathy C ， Doroudchi E ， et al . Optimized siting and sizing of distribution-network-connected battery energy storage system providing flexibility services for system operators ［J］. Energy ， 2023 ， 285 ： 129490 .

Salazar V C E ， Gallego P L A ， GóMez C O . A multi-period MILP model for optimal integration of battery energy storage systems and distributed generation in power distribution systems ［J］. Results in Engineering ， 2025 ， 27 ： 106806 .

Chen Minwu ， Gong Xin ， Liang Zongyou ， et al . Bi-hierarchy capacity programming of co-phase TPSS with PV and HESS for minimum life cycle cost ［J］. International Journal of Electrical Power & Energy Systems ， 2023 ， 147 ： 108904 ．

Xu Xiaofeng ， Wang Kai ， Ma Wenhao ， et al ． Multi-objective particle swarm optimization algorithm based on multi-strategy improvement for hybrid energy storage optimization configuration ［J］. Renewable Energy ， 2024 ， 223 ： 120086 ．

Mohseni S ， Brent A C ． Probabilistic sizing and scheduling co-optimisation of hybrid battery/super-capacitor energy storage systems in micro-grids ［J］. Journal of Energy Storage ， 2023 ， 73 ： 109172 ．

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Multi-scale Topology-aware Graph Neural Network Method for Distribution Line Parameter Identification

Charging Load Forecasting Method for Electric Vehicles Considering Dynamic Reconfiguration of Distribution Networks

Fault Location for Active Distribution Network with High Penetration Rate Based on Improved Spatio-Temporal Graph Neural Network

Multi-market Two-stage Trading Decisions with Participation of Electric Vehicle Clusters Under Diversified Business Modes

Fostering Endogenous Security via Regulation Responsibility: Exploration of New Electricity Market Paradigm

Related Author

LI Bingle

YE Chengjin

DING Yi

FAN Junqiu

TANG Xueyong

CHEN Chun

CHEN Yuxiang

CAO Yijia

Related Institution

College of Electrical Engineering, Zhejiang University

Guizhou Power Grid Co., Ltd.

State Key Laboratory of Disaster Prevention & Reduction for Power Grid, Changsha University of Science & Technology

State Grid Economic and Technological Research Institute Co., Ltd.

Electric Power Research Institute of State Grid Jiangxi Electric Power Co., Ltd.

AI问答

⁰