基于泛化图论与DDPG算法的高渗透率光伏配电网能量路由策略研究

李红; 王家威; 杨剑锋; 伏一平

doi:10.19912/j.0254-0096.tynxb.2025-0698

PDF(2050 KB)

太阳能学报 ›› 2026, Vol. 47 ›› Issue (1) : 127-135. DOI: 10.19912/j.0254-0096.tynxb.2025-0698

基于泛化图论与DDPG算法的高渗透率光伏配电网能量路由策略研究

李红, 王家威, 杨剑锋, 伏一平

作者信息 +

RESEARCH ON ENERGY ROUTING STRATEGY FOR HIGH-PENETRATION PHOTOVOLTAIC DISTRIBUTION NETWORK BASED ON GENERALIZED GRAPH THEORY AND DDPG ALGORITHM

Li Hong, Wang Jiawei, Yang Jianfeng, Fu Yiping

Author information +

文章历史 +

摘要

针对基于能量路由器（ER）间接接纳分布式光伏的高渗透率光伏配电网由于本地台区光伏消纳或出力不足导致的弃光及功率缺额问题,引入高渗透率光伏能量路由器互联系统（HP-PVERIS）,提出一种基于泛化图论和深度确定性策略梯度（DDPG）算法的双层相干驱动能量路由策略。第一层通过定义多目标优化模型,针对节点转换损耗、线路损耗、网络拥塞及能量负载均衡问题,利用改进的泛化图论求解路由路径;第二层基于DDPG强化学习算法对多路径进行最小损耗目标的动态功率分配,并将优化后的新拓扑参数反馈至第一层,形成双层驱动的路由分配策略。仿真结果表明,该策略能够有效改善网损、网络拥塞及负载均衡问题,同时提升高渗透率光伏配电网的光伏消纳和互济能力。

Abstract

To address PV curtailment and power deficits in high-penetration PV distribution networks using energy router （ER）, this paper introduces a high-penetration photovoltaic energy router interconnection system （HP-PVERIS） and proposes a two-layer coherence-driven energy routing strategy combining generalized graph theory and deep deterministic policy gradient （DDPG）. The first layer uses an improved generalized graph theory model for multi-objective optimization of node conversion loss, line loss, congestion, and load balancing. The second layer applies DDPG-based dynamic multi-path power allocation, feeding optimized topology parameters back to the first layer to form a bi-layer routing strategy. Simulations show the approach effectively reduces losses, alleviates congestion, balances loads, and enhances PV accommodation and interconnection capacity.

导出引用

李红, 王家威, 杨剑锋, 伏一平. 基于泛化图论与DDPG算法的高渗透率光伏配电网能量路由策略研究[J]. 太阳能学报. 2026, 47(1): 127-135 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0698

Li Hong, Wang Jiawei, Yang Jianfeng, Fu Yiping. RESEARCH ON ENERGY ROUTING STRATEGY FOR HIGH-PENETRATION PHOTOVOLTAIC DISTRIBUTION NETWORK BASED ON GENERALIZED GRAPH THEORY AND DDPG ALGORITHM[J]. Acta Energiae Solaris Sinica. 2026, 47(1): 127-135 https://doi.org/10.19912/j.0254-0096.tynxb.2025-0698

中图分类号： TM734

参考文献

[1] 白泽, 刘灏, 毕天姝. 基于图注意力网络的配电网分布式光伏实时消纳能力评估方法[J]. 电力自动化设备, 2024, 44(12): 92-99.BAI Z, LIU H, BI T S. Real-time distributed PV consumption capacity assessment method for distribution network based on graph attention network[J]. Electric power automation equipment, 2024, 44(12): 92-99.
[2] 唐雅洁, 李俊豪, 林达, 等. 考虑分布式光伏高效消纳与负荷损失最小的区域配电网应急资源协同配置策略[J]. 电力自动化设备, 2024, 44(7): 78-85.TANG Y J, LI J H, LIN D, et al. Coordinated allocation strategy of emergency resource for regional distribution network considering distributed photovoltaic efficient absorption and minimum load loss[J]. Electric power automation equipment, 2024, 44(7): 78-85.
[3] 白迪, 张铁岩, 杨政. 基于可控等效阻抗的能量路由器无功电压控制策略研究[J]. 太阳能学报, 2022, 43(1): 329-334.BAI D, ZHANG T Y, YANG Z. Research on reactive voltage control strategy of energy routers based on controllable equivalent impedance[J]. Acta energiae solaris sinica, 2022, 43(1): 329-334.
[4] 戴玮. 基于能量路由器的分布式光伏多级消纳策略研究[D]. 北京: 北京交通大学, 2019.DAI W. Research on multi-level distributed photovoltaic consumption strategies based on energy routers[D]. Beijing: Beijing Jiaotong University, 2019.
[5] DU Y F, YIN X G, LAI J M, et al.Energy optimization and routing control strategy for energy router based multi-energy interconnected energy system[J]. International journal of electrical power & energy systems, 2021, 133: 107110.
[6] JIANG X Y, SUN C, CAO L L, et al.Semi-decentralized energy routing algorithm for minimum-loss transmission in community energy internet[J]. International journal of electrical power & energy systems, 2022, 135: 107547.
[7] WANG R C, WU J D, QIAN Z N, et al.A graph theory based energy routing algorithm in energy local area network[J]. IEEE transactions on industrial informatics, 2017, 13(6): 3275-3285.
[8] 林川, 赵海, 刘晓, 等. 能源互联网路由策略研究[J]. 电工技术学报, 2015(11): 37-44.LIN C, ZHAO H, LIU X, et al. Research on routing strategy of energy Internet[J]. Transactions of China Electrotechnical Society, 2015(11): 37-44.
[9] GUO H, WANG F, LI L, et al.A minimum loss routing algorithm based on real-time transaction in energy internet[J]. IEEE transactions on industrial informatics, 2019, 15(12): 6446-6456.
[10] 陈璨, 樊小伟, 张文浩, 等. 促进分布式光伏消纳的两阶段源网荷储互动优化运行策略[J]. 电网技术, 2022, 46(10): 3786-3799.CHEN C, FAN X W, ZHANG W H, et al. Two-staged generation-grid-load-energy storage interactive optimization operation strategy for promotion of distributed photovoltaic consumption[J]. Power system technology, 2022, 46(10): 3786-3799.
[11] JIANG X Y, SUN C, CAO L L, et al.Decentralized local energy trading with cooperative energy routing in energy local area network[J]. International journal of electrical power & energy systems, 2023, 152: 109209.
[12] JI X T, LIU D, XIAO F, et al.A novel graph theory based two-stage minimum cost routing mechanism in energy Internet[J]. International journal of electrical power & energy systems, 2022, 142: 108346.
[13] 曾博, 穆宏伟, 孟自帅, 等. 考虑通信灵活性的5G基站与含光伏电源配电网多目标区间规划[J]. 太阳能学报, 2024, 45(4): 423-433.ZENG B, MU H W, MENG Z S, et al. Multi-objective interval planning for 5G base stations and distribution networks with photovoltaic power sources considering communication flexibility[J]. Acta energiae solaris sinica, 2024, 45(4): 423-433.
[14] 立梓辰, 吴远. 多能形式能源路由器运行工况集的能量流控制[J]. 太阳能学报, 2023, 44(10): 68-79.LI Z C, WU Y. Energy flow control of multi-energy router operating condition set[J]. Acta energiae solaris sinica, 2023, 44(10): 68-79.
[15] 郭慧, 汪飞, 张笠君, 等. 基于撮合交易的能源互联网最小网损路由算法[J]. 电力系统自动化, 2018, 42(14): 172-179.GUO H, WANG F, ZHANG L J, et al. Matchmaking tradeoff based minimum loss routing algorithm in energy internet[J]. Automation of electric power systems, 2018, 42(14): 172-179.
[16] 宋元明, 刘亚杰, 王锐, 等. 基于微波无线传能的动态无线传能链路多目标规划问题[J]. 控制与决策, 2021, 36(12): 3039-3048.SONG Y M, LIU Y J, WANG R, et al. Multi-objective planning of dynamic wireless energy transmission link based on microwave wireless energy transmission[J]. Control and decision, 2021, 36(12): 3039-3048.
[17] 曾博, 徐富强, 刘裕, 等. 考虑可再生能源与需求响应协同增效的能量枢纽多目标区间优化规划方法[J]. 中国电机工程学报, 2021, 41(21): 7212-7225.ZENG B, XU F Q, LIU Y, et al. Multi-objective interval optimization approach for energy hub planning with consideration of renewable energy and demand response synergies[J]. Proceedings of the CSEE, 2021, 41(21): 7212-7225.
[18] 徐良. 能量路由器的用电策略及路由策略研究[D]. 宁波: 宁波大学, 2022.XU L. Research on power consumption policy and routing strategy of energy router[D]. Ningbo: Ningbo University, 2022.
[19] JIA M, WU J, ZHANG L, et al.DDPG-based intelligent computation offloading and resource allocation for LEO satellite edge computing network[J]. China communications, 2025, 22(3): 1-15.
[20] 肖泽青, 华昊辰, 曹军威. 人工智能在能源互联网中的应用综述[J]. 电力建设, 2019, 40(5): 63-70.XIAO Z Q, HUA H C, CAO J W. Overview of the application of artificial intelligence in energy Internet[J]. Electric power construction, 2019, 40(5): 63-70.
[21] 刘顺成, 何禹清, 乔学博, 等. 基于合作博弈的交直流配电网光储协调规划方法[J]. 太阳能学报, 2024, 45(1): 534-544.LIU S C, HE Y Q, QIAO X B, et al. Photovoltaic-storage coordinated planning method of AC/DC distribution network based on cooperative game[J]. Acta energiae solaris sinica, 2024, 45(1): 534-544.
[22] YANG X, ZHAO B.Optimal neuro-control strategy for nonlinear systems with asymmetric input constraints[J]. IEEE/CAA journal of automatica sinica, 2020, 7(2): 575-583.
[23] XUE S, LUO B, LIU D R, et al.Neural network-based event-triggered integral reinforcement learning for constrained H∞ tracking control with experience replay[J]. Neurocomputing, 2022, 513: 25-35.