Sustainable Energy Grids & Networks最新文献_第4页

Optimal management of green hydrogen production in renewable energy systems using deep reinforcement learning methods 利用深度强化学习方法优化可再生能源系统中绿色制氢的管理

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-05 DOI: 10.1016/j.segan.2025.102075

Donguk Yang , Junki Shim , Jinkun Lee , Seongim Choi

This research focuses on developing a deep reinforcement learning (DRL) framework to optimize green hydrogen production within renewable energy systems. By integrating a DRL-based model, the study aims to enhance real-time management of energy supply, storage, and distribution, involving an electrolyzer and balancing energy flows from photovoltaic (PV) sources, an energy storage system (ESS) and grid power. Utilizing real-world data, the DRL model adapts dynamically to fluctuations in renewable energy output and market prices, thereby optimizing operational efficiency. The study compares various DRL algorithms, including proximal policy optimization (PPO), soft actor-critic (SAC), and advantage actor-critic (A2C), assessing their performance in maximizing predefined reward functions. The findings demonstrate the robustness of the PPO algorithm, demonstrating significant reward accumulation and adaptability in managing dynamic environments. This validation is supported by empirical data and learning curves, confirming the DRL model’s proficiency in optimizing energy use and enhancing operational performance in green hydrogen systems. The integration of DRL with the framework for green hydrogen and renewable energy suggests a comprehensive solution that improves energy efficiency, operational costs, and sustainability initiatives. The research highlights the potential of advanced machine learning techniques for enhanced operational efficiency of renewable energy systems.

本研究的重点是开发一个深度强化学习（DRL）框架，以优化可再生能源系统中的绿色氢气生产。通过集成基于drl的模型，该研究旨在增强能源供应、存储和分配的实时管理，包括电解槽和平衡来自光伏（PV）源、储能系统（ESS）和电网的能量流。DRL模型利用实际数据，动态适应可再生能源产量和市场价格的波动，从而优化运行效率。该研究比较了各种DRL算法，包括近端策略优化（PPO）、软行为者批评（SAC）和优势行为者批评（A2C），评估了它们在最大化预定义奖励函数方面的表现。研究结果证明了PPO算法的鲁棒性，在管理动态环境中展示了显著的奖励积累和适应性。这一验证得到了经验数据和学习曲线的支持，证实了DRL模型在优化能源使用和提高绿色氢系统运行性能方面的熟练程度。DRL与绿色氢和可再生能源框架的整合提出了一个全面的解决方案，可以提高能源效率、运营成本和可持续性举措。该研究强调了先进的机器学习技术在提高可再生能源系统运行效率方面的潜力。

{"title":"Optimal management of green hydrogen production in renewable energy systems using deep reinforcement learning methods","authors":"Donguk Yang , Junki Shim , Jinkun Lee , Seongim Choi","doi":"10.1016/j.segan.2025.102075","DOIUrl":"10.1016/j.segan.2025.102075","url":null,"abstract":"<div><div>This research focuses on developing a deep reinforcement learning (DRL) framework to optimize green hydrogen production within renewable energy systems. By integrating a DRL-based model, the study aims to enhance real-time management of energy supply, storage, and distribution, involving an electrolyzer and balancing energy flows from photovoltaic (PV) sources, an energy storage system (ESS) and grid power. Utilizing real-world data, the DRL model adapts dynamically to fluctuations in renewable energy output and market prices, thereby optimizing operational efficiency. The study compares various DRL algorithms, including proximal policy optimization (PPO), soft actor-critic (SAC), and advantage actor-critic (A2C), assessing their performance in maximizing predefined reward functions. The findings demonstrate the robustness of the PPO algorithm, demonstrating significant reward accumulation and adaptability in managing dynamic environments. This validation is supported by empirical data and learning curves, confirming the DRL model’s proficiency in optimizing energy use and enhancing operational performance in green hydrogen systems. The integration of DRL with the framework for green hydrogen and renewable energy suggests a comprehensive solution that improves energy efficiency, operational costs, and sustainability initiatives. The research highlights the potential of advanced machine learning techniques for enhanced operational efficiency of renewable energy systems.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102075"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-objective reinforcement learning for electric vehicle charging 电动汽车充电的多目标强化学习

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-06 DOI: 10.1016/j.segan.2025.102083

Maximiliano Trimboli, Luis Avila

The transportation sector is a significant contributor to global greenhouse gas emissions, and Electric Vehicles (EVs) have emerged as a promising solution to mitigate this impact by reducing emissions and integrating renewable energy sources. However, battery charging remains a major obstacle to widespread EV adoption, as charging speed is constrained by battery specifications, C-rate limits, and the need to prevent degradation due to thermal and electrochemical stress. To address these challenges, this work proposes a Multi-Objective Reinforcement Learning (MORL) approach for optimal EV battery charging. Unlike traditional methods that rely on hand-crafted scalar rewards, MORL enables the agent to learn control policies that dynamically balance multiple, often conflicting, objectives—such as fast charging and thermal safety—based on user-defined preferences. Leveraging the architecture of a Deep RL agent, the proposed method adapts its charging strategy in real-time, applying high currents when thermal conditions are favorable and reducing them near critical thresholds. Experimental results show the policy’s adaptability: faster charging is achieved when temperature constraints are relaxed, while more conservative profiles emerge when battery longevity is prioritized. This highlights the potential of MORL to enhance both the safety and efficiency of EV charging.

交通运输部门是全球温室气体排放的重要贡献者，电动汽车（ev）已经成为一种有希望的解决方案，通过减少排放和整合可再生能源来减轻这种影响。然而，电池充电仍然是电动汽车广泛采用的主要障碍，因为充电速度受到电池规格、c -速率限制以及防止热应力和电化学应力导致的退化的需要的限制。为了解决这些挑战，本研究提出了一种多目标强化学习（MORL）方法来优化电动汽车电池充电。与依赖手工制作的标量奖励的传统方法不同，MORL使智能体能够根据用户定义的偏好学习动态平衡多个经常相互冲突的目标（例如快速充电和热安全）的控制策略。利用Deep RL代理的架构，所提出的方法可以实时调整其充电策略，在热条件有利时施加大电流，并在临界阈值附近降低电流。实验结果表明，该策略具有较强的适应性：当温度约束较宽松时，充电速度较快；而当电池寿命优先考虑时，充电曲线较为保守。这凸显了MORL在提高电动汽车充电安全性和效率方面的潜力。

{"title":"Multi-objective reinforcement learning for electric vehicle charging","authors":"Maximiliano Trimboli, Luis Avila","doi":"10.1016/j.segan.2025.102083","DOIUrl":"10.1016/j.segan.2025.102083","url":null,"abstract":"<div><div>The transportation sector is a significant contributor to global greenhouse gas emissions, and Electric Vehicles (EVs) have emerged as a promising solution to mitigate this impact by reducing emissions and integrating renewable energy sources. However, battery charging remains a major obstacle to widespread EV adoption, as charging speed is constrained by battery specifications, C-rate limits, and the need to prevent degradation due to thermal and electrochemical stress. To address these challenges, this work proposes a Multi-Objective Reinforcement Learning (MORL) approach for optimal EV battery charging. Unlike traditional methods that rely on hand-crafted scalar rewards, MORL enables the agent to learn control policies that dynamically balance multiple, often conflicting, objectives—such as fast charging and thermal safety—based on user-defined preferences. Leveraging the architecture of a Deep RL agent, the proposed method adapts its charging strategy in real-time, applying high currents when thermal conditions are favorable and reducing them near critical thresholds. Experimental results show the policy’s adaptability: faster charging is achieved when temperature constraints are relaxed, while more conservative profiles emerge when battery longevity is prioritized. This highlights the potential of MORL to enhance both the safety and efficiency of EV charging.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102083"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738304","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unified peer-to-peer energy and frequency response reserve trading in isolated multi-microgrid systems 孤立多微电网系统中统一点对点能量和频率响应储备交易

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-05 DOI: 10.1016/j.segan.2025.102080

Chao Sun , Yun Liu , Ziyu Chen , Jizhong Zhu

Peer-to-peer (P2P) energy trading in a multi-microgrid (MMG) system can incentivize energy sharing and reduce the overall operational cost. However, the MMG system operating in isolated mode may face a reduction in system frequency response reserves, especially the inertia and primary frequency response (IPFR) reserve due to the growing integration of renewable energy resources (RESs) via power electronic inverters. Therefore, the current P2P trading framework ignoring the component of IPFR reserve could lead to frequency insecurity. To overcome these limitations, this paper proposes a two-stage P2P energy and IPFR reserve trading mechanism while considering the participation of synchronous generators (SGs) and inverter-based RES (IBRs) in a MMG system. In the first stage, a frequency-constrained unit commitment (UC) problem is formulated, where the unified transfer function structure is implemented in SGs and IBRs to analyze the frequency dynamic processes. In the second stage, each microgrid autonomously negotiates optimal energy and IPFR reserve trading based on the determined UC results through a fully decentralized ADMM based iterative algorithm, clearly reflecting the costs and prices involved. Case studies on 4-MG and 10-MG systems demonstrate that the proposed scheme ensures frequency-secure operation with good scalability. Results show that an additional cost of 995 CNY per day can avoid an economic loss of 1599 CNY per minute during frequency collapse events, confirming the economic efficiency and frequency-security benefits of the proposed approach.

多微电网（MMG）系统中的点对点（P2P）能源交易可以激励能源共享，降低整体运营成本。然而，在隔离模式下运行的MMG系统可能会面临系统频率响应储备的减少，特别是由于可再生能源（RESs）通过电力电子逆变器的日益整合，惯性和一次频率响应（IPFR）储备的减少。因此，目前的P2P交易框架忽略了IPFR储备的组成部分，可能导致频率不安全。为了克服这些限制，本文提出了一种考虑同步发电机（SGs）和基于逆变器的RES （IBRs）在MMG系统中的参与的两阶段P2P能源和IPFR储备交易机制。首先，提出频率约束单元承诺（UC）问题，在SGs和IBRs中实现统一的传递函数结构，分析频率动态过程；在第二阶段，每个微电网通过完全分散的基于ADMM的迭代算法，根据确定的UC结果自主协商最优能源和IPFR储备交易，清楚地反映所涉及的成本和价格。对4-MG和10-MG系统的实例研究表明，该方案保证了频率安全运行，具有良好的可扩展性。结果表明，每天995元的额外费用可以避免频率崩溃事件中每分钟1599元的经济损失，证实了该方法的经济效率和频率安全效益。

{"title":"Unified peer-to-peer energy and frequency response reserve trading in isolated multi-microgrid systems","authors":"Chao Sun , Yun Liu , Ziyu Chen , Jizhong Zhu","doi":"10.1016/j.segan.2025.102080","DOIUrl":"10.1016/j.segan.2025.102080","url":null,"abstract":"<div><div>Peer-to-peer (P2P) energy trading in a multi-microgrid (MMG) system can incentivize energy sharing and reduce the overall operational cost. However, the MMG system operating in isolated mode may face a reduction in system frequency response reserves, especially the inertia and primary frequency response (IPFR) reserve due to the growing integration of renewable energy resources (RESs) via power electronic inverters. Therefore, the current P2P trading framework ignoring the component of IPFR reserve could lead to frequency insecurity. To overcome these limitations, this paper proposes a two-stage P2P energy and IPFR reserve trading mechanism while considering the participation of synchronous generators (SGs) and inverter-based RES (IBRs) in a MMG system. In the first stage, a frequency-constrained unit commitment (UC) problem is formulated, where the unified transfer function structure is implemented in SGs and IBRs to analyze the frequency dynamic processes. In the second stage, each microgrid autonomously negotiates optimal energy and IPFR reserve trading based on the determined UC results through a fully decentralized ADMM based iterative algorithm, clearly reflecting the costs and prices involved. Case studies on 4-MG and 10-MG systems demonstrate that the proposed scheme ensures frequency-secure operation with good scalability. Results show that an additional cost of 995 CNY per day can avoid an economic loss of 1599 CNY per minute during frequency collapse events, confirming the economic efficiency and frequency-security benefits of the proposed approach.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102080"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145738368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A branch-and-bound algorithm for radial distribution system reconfiguration 径向配电系统重构的分支定界算法

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2026-01-27 DOI: 10.1016/j.segan.2026.102138

Izabeli M. Rosa , Gabriel F. Puerta , Rubén Romero , Leonardo H. Macedo

This work presents a branch-and-bound structured algorithm to solve the electrical power distribution system reconfiguration (DSR) problem, whose primary objective is to find a radial network topology that minimizes active power losses. This proposal works independently of a formal mathematical model for the DSR problem, which is a mixed-integer nonlinear programming model. After obtaining a high-quality initial solution with a constructive heuristic algorithm, the proposed algorithm applies four distinct search movements within the branch-and-bound structure to efficiently fathom the vast solution space. The algorithm’s performance is validated using the 33-, 69-, 84-, and 118-bus systems. Results demonstrate that the proposed technique consistently identifies the best-known optimal solution for all cases while drastically reducing the search space. For the 84- and 118-bus systems, the algorithm explored a minuscule fraction of the search space to find the optimal topology, showcasing its exceptional efficiency and scalability for solving complex DSR problems.

本文提出了一种解决配电系统重构问题的分支定界结构化算法，其主要目标是找到一个使有功功率损耗最小的径向网络拓扑结构。该方法独立于DSR问题的正式数学模型，即混合整数非线性规划模型。在利用建设性启发式算法获得高质量的初始解后，该算法在分支定界结构内采用四种不同的搜索运动，以有效地了解巨大的解空间。采用33、69、84和118总线系统验证了算法的性能。结果表明，所提出的技术一致地识别出所有情况下最知名的最优解决方案，同时大大减少了搜索空间。对于84总线和118总线系统，该算法探索了极小部分的搜索空间来找到最优拓扑，展示了其解决复杂DSR问题的卓越效率和可扩展性。

{"title":"A branch-and-bound algorithm for radial distribution system reconfiguration","authors":"Izabeli M. Rosa , Gabriel F. Puerta , Rubén Romero , Leonardo H. Macedo","doi":"10.1016/j.segan.2026.102138","DOIUrl":"10.1016/j.segan.2026.102138","url":null,"abstract":"<div><div>This work presents a branch-and-bound structured algorithm to solve the electrical power distribution system reconfiguration (DSR) problem, whose primary objective is to find a radial network topology that minimizes active power losses. This proposal works independently of a formal mathematical model for the DSR problem, which is a mixed-integer nonlinear programming model. After obtaining a high-quality initial solution with a constructive heuristic algorithm, the proposed algorithm applies four distinct search movements within the branch-and-bound structure to efficiently fathom the vast solution space. The algorithm’s performance is validated using the 33-, 69-, 84-, and 118-bus systems. Results demonstrate that the proposed technique consistently identifies the best-known optimal solution for all cases while drastically reducing the search space. For the 84- and 118-bus systems, the algorithm explored a minuscule fraction of the search space to find the optimal topology, showcasing its exceptional efficiency and scalability for solving complex DSR problems.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102138"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Ensuring security in coupled distribution networks and electric vehicles-integrated energy storage systems: Transfer-safe reinforcement learning 耦合配电网和电动汽车集成储能系统的安全保障：转移安全强化学习

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2026-02-04 DOI: 10.1016/j.segan.2026.102147

Peixiao Fan , Yuxin Wen , Jun Yang , Song Ke

The integration of distribution networks (DN) with electrochemical energy storage systems (ESS), including large-scale user-side storage such as electric vehicles (EVs), introduces complex interactions. Faults in DNs, such as abnormal electrical states, can propagate to ESS and EVs, triggering critical events like thermal runaway. Conversely, the dynamic states of ESS and EVs, including high-speed charging and discharging, can destabilize DNs, exacerbating operational risks. These bidirectional interactions pose significant challenges to ensuring the security and stability of DN-ESS systems. However, the mechanisms underlying these interactions between DNs and ESS remain insufficiently comprehended. To bridge this gap, this paper proposes a novel coordinated protection strategy based on a transfer safe reinforcement learning (TSRL). Firstly, a state evolution model of ESS, including EV-integrated systems, is meticulously established to clarify the coupling effects and transmission paths of abnormal operation states. Secondly, security operation and inherent protection models for both the ESS and the DN are comprehensively developed, with the security operation boundaries for both systems being carefully designed to ensure robustness. Subsequently, the risk state evaluation indicators for the coupled operation are systematically selected, and specific phased control measures are thoughtfully designed to address varying operational scenarios. Building upon these foundations, TSRL is designed to enhance adaptability across different EV and ESS configurations via transfer learning while enforcing strict safety constraints on control actions through a safety layer. This approach ensures secure, efficient, and adaptive control strategies under dynamic and uncertain conditions. Finally, simulations verify that the proposed strategy can reliably ensure the operational safety of both the DN and EV-integrated ESS, accurately identify system operation risks, and effectively block the transmission of abnormal electrical quantities within the coupled system, ensuring the secure and stable operation of EV-integrated ESS in DNs.

配电网络（DN）与电化学储能系统（ESS）的集成，包括大规模用户侧存储，如电动汽车（ev），引入了复杂的相互作用。DNs故障，如异常的电气状态，可以传播到ESS和电动汽车，引发热失控等关键事件。相反，ESS和电动汽车的动态状态，包括高速充放电，会使DNs不稳定，加剧运营风险。这些双向交互对确保DN-ESS系统的安全性和稳定性提出了重大挑战。然而，DNs和ESS之间相互作用的机制仍然没有得到充分的理解。为了弥补这一差距，本文提出了一种基于转移安全强化学习（TSRL）的新型协调保护策略。首先，精心建立了ESS（包括电动汽车集成系统）的状态演化模型，明确了异常运行状态的耦合效应和传递路径。其次，全面开发了ESS和DN的安全运行和固有保护模型，并精心设计了两个系统的安全运行边界，确保了系统的鲁棒性。随后，系统选择耦合运行的风险状态评价指标，并针对不同运行场景精心设计具体的阶段性控制措施。基于这些基础，TSRL旨在通过迁移学习增强不同EV和ESS配置的适应性，同时通过安全层对控制动作实施严格的安全约束。该方法保证了动态和不确定条件下的安全、高效和自适应控制策略。最后，仿真验证了所提出的策略能够可靠地保证DN和ev集成ESS的运行安全，准确识别系统运行风险，有效阻断耦合系统内异常电量的传输，确保ev集成ESS在DN中安全稳定运行。

{"title":"Ensuring security in coupled distribution networks and electric vehicles-integrated energy storage systems: Transfer-safe reinforcement learning","authors":"Peixiao Fan , Yuxin Wen , Jun Yang , Song Ke","doi":"10.1016/j.segan.2026.102147","DOIUrl":"10.1016/j.segan.2026.102147","url":null,"abstract":"<div><div>The integration of distribution networks (DN) with electrochemical energy storage systems (ESS), including large-scale user-side storage such as electric vehicles (EVs), introduces complex interactions. Faults in DNs, such as abnormal electrical states, can propagate to ESS and EVs, triggering critical events like thermal runaway. Conversely, the dynamic states of ESS and EVs, including high-speed charging and discharging, can destabilize DNs, exacerbating operational risks. These bidirectional interactions pose significant challenges to ensuring the security and stability of DN-ESS systems. However, the mechanisms underlying these interactions between DNs and ESS remain insufficiently comprehended. To bridge this gap, this paper proposes a novel coordinated protection strategy based on a transfer safe reinforcement learning (TSRL). Firstly, a state evolution model of ESS, including EV-integrated systems, is meticulously established to clarify the coupling effects and transmission paths of abnormal operation states. Secondly, security operation and inherent protection models for both the ESS and the DN are comprehensively developed, with the security operation boundaries for both systems being carefully designed to ensure robustness. Subsequently, the risk state evaluation indicators for the coupled operation are systematically selected, and specific phased control measures are thoughtfully designed to address varying operational scenarios. Building upon these foundations, TSRL is designed to enhance adaptability across different EV and ESS configurations via transfer learning while enforcing strict safety constraints on control actions through a safety layer. This approach ensures secure, efficient, and adaptive control strategies under dynamic and uncertain conditions. Finally, simulations verify that the proposed strategy can reliably ensure the operational safety of both the DN and EV-integrated ESS, accurately identify system operation risks, and effectively block the transmission of abnormal electrical quantities within the coupled system, ensuring the secure and stable operation of EV-integrated ESS in DNs.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102147"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147395618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Development of grid-compliance metric for reliable integration of fast charging stations in power networks 电网快速充电站可靠集成的电网顺应性指标研究

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-20 DOI: 10.1016/j.segan.2025.102088

Salman Harasis , Irfan Khan , Ahmed Massoud

With the fast deployment of electric bus fleets in public transportation, efficient energy consumption and grid impact have become major concerns and challenges. Moreover, due to the accelerated need for fast charging, low-voltage distribution networks show insufficient hosting capacity for E-transportation systems, which encourages stations to connect to the medium voltage lines. In addition to the voltage level constraints, many factors associated with fast charging affect the grid interaction level and the maximum charging power that can be applied to charge the fleets. Although several studies have analyzed voltage deviations, harmonics, and renewable support individually, there remains a lack of a comprehensive grid-compliance evaluation methodology that can holistically quantify these impacts for large-scale charging stations. Therefore, this paper proposes a reliable grid interaction framework for E-bus fleets and develops a novel grid impact metric to ensure efficient charging power with minimal grid impact in a PV grid-connected system. The measures include voltage profile, charging power, and grid-injected harmonics. This work examines an optimal charging strategy to address fast charging challenges, featuring novel performance indices that quantify the grid impact and PV power generation. The proposed strategy is demonstrated by evaluating the charging station deployed at the IEEE 34-node network under different voltage levels. The proposed IGIM is demonstrated on the IEEE 34-node test feeder, where results show that MV connection significantly outperforms LV in terms of grid hosting capacity, which reduces voltage deviations by more than 50 %. Harmonic analysis reveals that constant-current mode charging up to 80 % SoC complies better with IEEE-519 limits than constant-voltage mode. In addition, PV-assisted charging increases self-consumption by up to 60 %.

随着电动公交车队在公共交通中的快速部署，高效的能源消耗和对电网的影响已经成为主要的问题和挑战。此外，由于对快速充电的需求加快，低压配电网络对电子交通系统的承载能力不足，这鼓励了车站连接到中压线路。除了电压水平的限制外，与快速充电相关的许多因素也会影响电网交互水平和可用于为车队充电的最大充电功率。尽管有几项研究分别分析了电压偏差、谐波和可再生能源支持，但仍然缺乏一种全面的电网合规性评估方法，可以全面量化这些对大型充电站的影响。因此，本文提出了一种可靠的电动公交车队电网交互框架，并开发了一种新的电网影响指标，以确保光伏并网系统中有效的充电功率和最小的电网影响。这些措施包括电压分布、充电功率和电网注入谐波。这项工作研究了一种解决快速充电挑战的最佳充电策略，具有量化电网影响和光伏发电的新型性能指标。通过评估不同电压水平下部署在IEEE 34节点网络中的充电站，验证了所提出的策略。提出的IGIM在IEEE 34节点测试馈线上进行了验证，结果表明，中压连接在电网承载能力方面明显优于低压连接，减少了50%以上的电压偏差 %。谐波分析表明，恒流模式充电高达80 % SoC比恒压模式更符合IEEE-519限制。此外，pv辅助充电增加了高达60% %的自我消耗。

{"title":"Development of grid-compliance metric for reliable integration of fast charging stations in power networks","authors":"Salman Harasis , Irfan Khan , Ahmed Massoud","doi":"10.1016/j.segan.2025.102088","DOIUrl":"10.1016/j.segan.2025.102088","url":null,"abstract":"<div><div>With the fast deployment of electric bus fleets in public transportation, efficient energy consumption and grid impact have become major concerns and challenges. Moreover, due to the accelerated need for fast charging, low-voltage distribution networks show insufficient hosting capacity for E-transportation systems, which encourages stations to connect to the medium voltage lines. In addition to the voltage level constraints, many factors associated with fast charging affect the grid interaction level and the maximum charging power that can be applied to charge the fleets. Although several studies have analyzed voltage deviations, harmonics, and renewable support individually, there remains a lack of a comprehensive grid-compliance evaluation methodology that can holistically quantify these impacts for large-scale charging stations. Therefore, this paper proposes a reliable grid interaction framework for E-bus fleets and develops a novel grid impact metric to ensure efficient charging power with minimal grid impact in a PV grid-connected system. The measures include voltage profile, charging power, and grid-injected harmonics. This work examines an optimal charging strategy to address fast charging challenges, featuring novel performance indices that quantify the grid impact and PV power generation. The proposed strategy is demonstrated by evaluating the charging station deployed at the IEEE 34-node network under different voltage levels. The proposed IGIM is demonstrated on the IEEE 34-node test feeder, where results show that MV connection significantly outperforms LV in terms of grid hosting capacity, which reduces voltage deviations by more than 50 %. Harmonic analysis reveals that constant-current mode charging up to 80 % SoC complies better with IEEE-519 limits than constant-voltage mode. In addition, PV-assisted charging increases self-consumption by up to 60 %.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102088"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An energy management strategy for integrated energy system based on data-driven and game theory methods 基于数据驱动和博弈论方法的综合能源系统能源管理策略

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2026-01-02 DOI: 10.1016/j.segan.2025.102118

Xun Xu, Zhenguo Shao, Feixiong Chen, Guoyang Cheng

To address the challenge of imperfect market coordination in multi-energy-coupled integrated energy systems (IESs) under uncertainty, especially the unresolved conflicts among stakeholders and the insufficient protection of disadvantaged participants within current market frameworks, an energy management strategy based on data-driven and game theory methods is proposed. Firstly, to optimize the benefits for both individual and collective stakeholders, a tri-level multi-energy management model is developed using multi-game framework, providing a novel approach to capturing interactions among diverse entities. Secondly, to handle the uncertainty of renewable energy, a data-driven distributionally robust chance constraint (DRCC) method is introduced, which uniquely combines dynamic Bayesian network (DBN) with imprecise Dirichlet model (IDM) and applies it to mixed ambiguity set that integrates desirable properties of different ambiguity sets. Finally, fixed-point theory is used to establish the existence of game equilibrium, and a Gauss-Seidel algorithm with adaptive inertia weight, combined with the alternating direction method of multipliers, is proposed to solve the multi-game model while ensuring the privacy of all parties. Case studies demonstrate that the DBN-IDM reduces the conservatism of parameter selection for the DRCC, and the proposed energy management strategy and improved Gauss-Seidel algorithm enhance participant benefits and accelerate convergence.

针对不确定条件下多能耦合综合能源系统市场协调不完善的挑战，特别是当前市场框架下利益相关者之间的冲突未解决和弱势参与者保护不足的问题，提出了一种基于数据驱动和博弈论方法的能源管理策略。首先，为了优化个体利益相关者和集体利益相关者的利益，利用多博弈框架建立了一个三级多能量管理模型，提供了一种捕捉不同实体之间相互作用的新方法。其次，针对可再生能源的不确定性，提出了一种数据驱动的分布式鲁棒机会约束（DRCC）方法，该方法将动态贝叶斯网络（DBN）与不精确狄利克雷模型（IDM）独特地结合起来，并将其应用于综合了不同模糊集所需特性的混合模糊集。最后，利用不动点理论建立了博弈均衡的存在性，并提出了一种自适应惯性权重的Gauss-Seidel算法，结合乘数交替方向法，在保证各方隐私的前提下求解多博弈模型。实例研究表明，DBN-IDM降低了DRCC参数选择的保守性，提出的能量管理策略和改进的Gauss-Seidel算法提高了参与者的利益，加快了收敛速度。

{"title":"An energy management strategy for integrated energy system based on data-driven and game theory methods","authors":"Xun Xu, Zhenguo Shao, Feixiong Chen, Guoyang Cheng","doi":"10.1016/j.segan.2025.102118","DOIUrl":"10.1016/j.segan.2025.102118","url":null,"abstract":"<div><div>To address the challenge of imperfect market coordination in multi-energy-coupled integrated energy systems (IESs) under uncertainty, especially the unresolved conflicts among stakeholders and the insufficient protection of disadvantaged participants within current market frameworks, an energy management strategy based on data-driven and game theory methods is proposed. Firstly, to optimize the benefits for both individual and collective stakeholders, a tri-level multi-energy management model is developed using multi-game framework, providing a novel approach to capturing interactions among diverse entities. Secondly, to handle the uncertainty of renewable energy, a data-driven distributionally robust chance constraint (DRCC) method is introduced, which uniquely combines dynamic Bayesian network (DBN) with imprecise Dirichlet model (IDM) and applies it to mixed ambiguity set that integrates desirable properties of different ambiguity sets. Finally, fixed-point theory is used to establish the existence of game equilibrium, and a Gauss-Seidel algorithm with adaptive inertia weight, combined with the alternating direction method of multipliers, is proposed to solve the multi-game model while ensuring the privacy of all parties. Case studies demonstrate that the DBN-IDM reduces the conservatism of parameter selection for the DRCC, and the proposed energy management strategy and improved Gauss-Seidel algorithm enhance participant benefits and accelerate convergence.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102118"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145926047","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel time-varying control method of renewable energy sources for smart grid efficiency enhancement 一种提高智能电网效率的可再生能源时变控制方法

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-11 DOI: 10.1016/j.segan.2025.102102

Xiaotong Ji , Dan Liu , Chang Ye , Ji Han , Bokai Zhou , Jiaming Guo , Bocheng Long , Yuqi Ao , Liangli Xiong

The rapid integration of renewable energy sources (RESs), such as photovoltaic (PV) and wind power generation (WPG), poses significant challenges to smart grids. Traditional control methods based on static or piecewise-linearized models are insufficiently adaptive to nonlinear and time-varying system behavior. This paper proposes a novel time-varying control strategy to enhance RES efficiency and coordination in smart grids. First, a control model is formulated considering both operational costs and system losses. To address system nonlinearities, a real-time sensitivity-based linearization scheme is developed to dynamically update the optimization model parameters as operating conditions evolve. Then, the optimality conditions of the time-varying optimization problem are derived, and a distributed control algorithm based on graph theory and finite-time convergence theory is proposed. The convergence of the algorithm is rigorously established through theoretical analysis. Finally, case studies are conducted on the IEEE 33-bus system and a real-world grid. The results demonstrate that the proposed method maintains generation–load deviation below 0.15 %, reduces operation cost and power loss by up to 8.5 % and 10.2 % compared with consensus, deep reinforcement learning (DRL), and droop control, and achieves RES consumption rates exceeding 85 % for WPG and 70 % for PV across representative scenarios.

光伏（PV）和风力发电（WPG）等可再生能源（RESs）的快速整合对智能电网提出了重大挑战。传统的基于静态或分段线性化模型的控制方法对非线性时变系统行为的适应性不足。本文提出了一种新的时变控制策略，以提高智能电网的可再生能源效率和协调性。首先，建立了考虑运行成本和系统损失的控制模型。为了解决系统的非线性问题，提出了一种基于实时灵敏度的线性化方案，根据工况变化动态更新优化模型参数。然后，推导了时变优化问题的最优性条件，提出了一种基于图论和有限时间收敛理论的分布式控制算法。通过理论分析，严格证明了算法的收敛性。最后，对IEEE 33总线系统和实际网格进行了案例研究。结果表明，与共识、深度强化学习（DRL）和下垂控制相比，该方法将发电负荷偏差保持在0.15 %以下，将运行成本和功率损耗分别降低8.5 %和10.2 %，并在代表性场景中实现了WPG超过85 %和PV超过70 %的RES消耗率。

{"title":"A novel time-varying control method of renewable energy sources for smart grid efficiency enhancement","authors":"Xiaotong Ji , Dan Liu , Chang Ye , Ji Han , Bokai Zhou , Jiaming Guo , Bocheng Long , Yuqi Ao , Liangli Xiong","doi":"10.1016/j.segan.2025.102102","DOIUrl":"10.1016/j.segan.2025.102102","url":null,"abstract":"<div><div>The rapid integration of renewable energy sources (RESs), such as photovoltaic (PV) and wind power generation (WPG), poses significant challenges to smart grids. Traditional control methods based on static or piecewise-linearized models are insufficiently adaptive to nonlinear and time-varying system behavior. This paper proposes a novel time-varying control strategy to enhance RES efficiency and coordination in smart grids. First, a control model is formulated considering both operational costs and system losses. To address system nonlinearities, a real-time sensitivity-based linearization scheme is developed to dynamically update the optimization model parameters as operating conditions evolve. Then, the optimality conditions of the time-varying optimization problem are derived, and a distributed control algorithm based on graph theory and finite-time convergence theory is proposed. The convergence of the algorithm is rigorously established through theoretical analysis. Finally, case studies are conducted on the IEEE 33-bus system and a real-world grid. The results demonstrate that the proposed method maintains generation–load deviation below 0.15 %, reduces operation cost and power loss by up to 8.5 % and 10.2 % compared with consensus, deep reinforcement learning (DRL), and droop control, and achieves RES consumption rates exceeding 85 % for WPG and 70 % for PV across representative scenarios.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102102"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Stackelberg game between charging stations and distribution networks with regional load forecasting and intelligent charging strategies 基于区域负荷预测和智能充电策略的充电站与配电网Stackelberg博弈

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-09 DOI: 10.1016/j.segan.2025.102093

Xiaocheng Wang , ZeLong Li , Qiaoni Han , Pengjiao Sun

In recent years, due to improper management of the relationship between charging stations (CSs) and distribution networks (DNs) in many areas, the fluctuation of power grid load has increased, which has affected the overall economic benefits of the power system. After analyzing the clear hierarchical relationship between CSs and DNs and their inherent rationality and selfishness, Stackelberg game is adopted. In this game, the DN tries to minimize its operating costs, while the goal of the CS is to maximize its profits. On the other hand, since it is difficult for DN to be aware of the load of each region in real time, this paper introduces regional load forecasting to help DN make more reasonable electricity pricing and power distribution plans. Moreover, due to the disorder and uncertainty of electric vehicle (EV) charging, the CS needs to control the charging behaviors of EVs, that is, the intelligent charging strategy is introduced to optimize the charging process, so as to ensure the load of the CS and improve its income. Finally, in order to solve the formulated Stackelberg game, the backward induction method is used to determine the optimal electricity purchase quantity of CSs and the optimal electricity price of DN through iteration. The simulation results show that the proposed method reduces the operating cost of DN by 20 % and increases the profit of CS by 18 %, and has significant advantages compared with other methods.

近年来，由于许多地区对充电站与配电网的关系管理不当，导致电网负荷波动增大，影响了电力系统的整体经济效益。在分析了CSs和dn之间清晰的层次关系以及它们内在的合理性和自私自利之后，采用Stackelberg博弈。在这个博弈中，DN的目标是最小化其运营成本，而CS的目标是最大化其利润。另一方面，由于DN难以实时了解各区域的负荷情况，本文引入区域负荷预测，帮助DN制定更合理的电价和配电方案。此外，由于电动汽车充电的无序性和不确定性，CS需要对电动汽车的充电行为进行控制，即引入智能充电策略对充电过程进行优化，从而保证CS的负载，提高CS的收益。最后，为了求解公式化的Stackelberg博弈，采用逆向归纳法，通过迭代确定CSs的最优购电量和DN的最优电价。仿真结果表明，该方法可使DN的运行成本降低20%，使CS的利润提高18%，与其他方法相比具有显著的优势。

{"title":"Stackelberg game between charging stations and distribution networks with regional load forecasting and intelligent charging strategies","authors":"Xiaocheng Wang , ZeLong Li , Qiaoni Han , Pengjiao Sun","doi":"10.1016/j.segan.2025.102093","DOIUrl":"10.1016/j.segan.2025.102093","url":null,"abstract":"<div><div>In recent years, due to improper management of the relationship between charging stations (CSs) and distribution networks (DNs) in many areas, the fluctuation of power grid load has increased, which has affected the overall economic benefits of the power system. After analyzing the clear hierarchical relationship between CSs and DNs and their inherent rationality and selfishness, Stackelberg game is adopted. In this game, the DN tries to minimize its operating costs, while the goal of the CS is to maximize its profits. On the other hand, since it is difficult for DN to be aware of the load of each region in real time, this paper introduces regional load forecasting to help DN make more reasonable electricity pricing and power distribution plans. Moreover, due to the disorder and uncertainty of electric vehicle (EV) charging, the CS needs to control the charging behaviors of EVs, that is, the intelligent charging strategy is introduced to optimize the charging process, so as to ensure the load of the CS and improve its income. Finally, in order to solve the formulated Stackelberg game, the backward induction method is used to determine the optimal electricity purchase quantity of CSs and the optimal electricity price of DN through iteration. The simulation results show that the proposed method reduces the operating cost of DN by 20 % and increases the profit of CS by 18 %, and has significant advantages compared with other methods.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102093"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Voltage sensitivity-guided aggregation for virtual power plants via a model-data integration framework 基于模型-数据集成框架的虚拟电厂电压灵敏度引导聚合

IF 5.6 2区工程技术 Q2 ENERGY & FUELS

Sustainable Energy Grids & Networks

Pub Date : 2026-03-01 Epub Date: 2025-12-09 DOI: 10.1016/j.segan.2025.102097

Xu Zhang , Wei Feng , Yanhui Zhang , Xuemei Dai

The interaction between virtual power plants (VPP) and distribution system operators is constrained by privacy preservation and voltage security requirements. Conventional dynamic operating envelopes (DOE) can safeguard privacy and voltage security, but they fail to guide VPP aggregation toward proactively mitigating voltage violations in distribution grids. This paper proposes a voltage sensitivity-guided aggregation driven by a model-data integration framework to address this limitation. The framework integrates a voltage-sensitivity affine model with data-driven uncertainty characterization, enabling aggregation with voltage regulation effects. Specifically, a voltage sensitivity affine model is established at the point of common coupling, where the stochastic factors of distributed energy resources are characterized using Gaussian mixture models combined with error propagation theory. The affine model is subsequently reformulated as a chance-constrained programming model, thus achieving the aggregation for VPP to ensure privacy preservation and voltage regulation. Case studies on the IEEE 33-bus distribution test system demonstrate that the proposed framework reduces aggregation costs and significantly enhances voltage regulation compared with conventional DOE-based aggregation approaches.

虚拟电厂（VPP）与配电系统运营商之间的交互受到隐私保护和电压安全要求的限制。传统的动态运行包络（DOE）可以保护隐私和电压安全，但它们无法引导VPP聚合主动减轻配电网中的电压违规。本文提出了一种由模型-数据集成框架驱动的电压灵敏度导向聚合来解决这一限制。该框架将电压敏感仿射模型与数据驱动的不确定性特征集成在一起，使聚合具有电压调节效果。具体而言，在共耦合点建立电压敏感仿射模型，利用高斯混合模型结合误差传播理论对分布式能源的随机因素进行表征。然后将仿射模型重新表述为机会约束规划模型，从而实现VPP的聚合，以确保隐私保护和电压调节。对IEEE 33总线配电测试系统的实例研究表明，与传统的基于doe的聚合方法相比，该框架降低了聚合成本，并显著提高了电压调节能力。

{"title":"Voltage sensitivity-guided aggregation for virtual power plants via a model-data integration framework","authors":"Xu Zhang , Wei Feng , Yanhui Zhang , Xuemei Dai","doi":"10.1016/j.segan.2025.102097","DOIUrl":"10.1016/j.segan.2025.102097","url":null,"abstract":"<div><div>The interaction between virtual power plants (VPP) and distribution system operators is constrained by privacy preservation and voltage security requirements. Conventional dynamic operating envelopes (DOE) can safeguard privacy and voltage security, but they fail to guide VPP aggregation toward proactively mitigating voltage violations in distribution grids. This paper proposes a voltage sensitivity-guided aggregation driven by a model-data integration framework to address this limitation. The framework integrates a voltage-sensitivity affine model with data-driven uncertainty characterization, enabling aggregation with voltage regulation effects. Specifically, a voltage sensitivity affine model is established at the point of common coupling, where the stochastic factors of distributed energy resources are characterized using Gaussian mixture models combined with error propagation theory. The affine model is subsequently reformulated as a chance-constrained programming model, thus achieving the aggregation for VPP to ensure privacy preservation and voltage regulation. Case studies on the IEEE 33-bus distribution test system demonstrate that the proposed framework reduces aggregation costs and significantly enhances voltage regulation compared with conventional DOE-based aggregation approaches.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"45 ","pages":"Article 102097"},"PeriodicalIF":5.6,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145791163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0