Electric Power Systems Research最新文献_第5页

A second-order generalized integrator phase-locked loop with a cascaded structure incorporating a low-pass filter 具有级联结构的二阶广义积分器锁相环，其中含有低通滤波器

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2026.112780

Yan Wang , Guangchen Liu , Guizhen Tian , Jianwei Zhang , Yuanyuan Wu , Weihong Zhao

Renewable energy sources (RESs) have achieved widespread global adoption owing to their favourable environmental impacts, together with their inherent sustainability, cost-effectiveness, and controllability. A phase-locked Loop (PLL) is one of the most widely used synchronisation techniques owing to its speed and robustness. This study was developed to address the problem that the traditional second-order generalised integrator-based PLL(SOGI-PLL) fails to achieve correct phase locking under non-ideal grid conditions in single-phase grid synchronization, using a cascaded structure that combines the traditional SOGI with a SOGI incorporating a LPF(SOGI-LPF) .Then, the transfer functions of the improved method and the traditional methods were analysed using Bode diagrams and root locus techniques. The improved PLL was tested on the RTDS/RCP hardware-in-the-loop experimental platform. Moreover, to compare its performance with that of the SOGI and SOGI-LPF methods in terms of indicators such as settling time, and phase error, several scenarios were developed. As a result, the improved PLL demonstrates the fastest dynamic response and, completely rejects four operating conditions—single-phase voltage sag, phase jump, low-order harmonics, and DC offset in the grid. Furthermore, compared with the first two PLLs, the transient phase-locking error of the improved PLL is reduced by up to 80 %.

可再生能源由于其有利的环境影响，以及其固有的可持续性、成本效益和可控性，已在全球得到广泛采用。锁相环（PLL）因其速度快、鲁棒性好而成为应用最广泛的同步技术之一。为了解决传统基于二阶广义积分器的PLL（SOGI-PLL）在单相电网同步非理想条件下无法实现正确锁相的问题，本研究采用了将传统SOGI与包含LPF的SOGI （SOGI-LPF）相结合的级联结构。然后利用博德图和根轨迹技术对改进方法和传统方法的传递函数进行了分析。改进的锁相环在RTDS/RCP硬件在环实验平台上进行了测试。此外，为了将其与SOGI和SOGI- lpf方法在沉降时间和相位误差等指标上的性能进行比较，开发了几种场景。因此，改进的锁相环表现出最快的动态响应，并且完全抑制了电网中的单相电压凹陷、相位跳变、低次谐波和直流偏置四种工作条件。此外，与前两种锁相环相比，改进锁相环的瞬态锁相误差降低了80%。

{"title":"A second-order generalized integrator phase-locked loop with a cascaded structure incorporating a low-pass filter","authors":"Yan Wang , Guangchen Liu , Guizhen Tian , Jianwei Zhang , Yuanyuan Wu , Weihong Zhao","doi":"10.1016/j.epsr.2026.112780","DOIUrl":"10.1016/j.epsr.2026.112780","url":null,"abstract":"<div><div>Renewable energy sources (RESs) have achieved widespread global adoption owing to their favourable environmental impacts, together with their inherent sustainability, cost-effectiveness, and controllability. A phase-locked Loop (PLL) is one of the most widely used synchronisation techniques owing to its speed and robustness. This study was developed to address the problem that the traditional second-order generalised integrator-based PLL(SOGI-PLL) fails to achieve correct phase locking under non-ideal grid conditions in single-phase grid synchronization, using a cascaded structure that combines the traditional SOGI with a SOGI incorporating a LPF(SOGI-LPF) .Then, the transfer functions of the improved method and the traditional methods were analysed using Bode diagrams and root locus techniques. The improved PLL was tested on the RTDS/RCP hardware-in-the-loop experimental platform. Moreover, to compare its performance with that of the SOGI and SOGI-LPF methods in terms of indicators such as settling time, and phase error, several scenarios were developed. As a result, the improved PLL demonstrates the fastest dynamic response and, completely rejects four operating conditions—single-phase voltage sag, phase jump, low-order harmonics, and DC offset in the grid. Furthermore, compared with the first two PLLs, the transient phase-locking error of the improved PLL is reduced by up to 80 %.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112780"},"PeriodicalIF":4.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Temporary measures for distribution network restoration after natural disasters 自然灾害后配电网恢复临时办法

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2026.112788

Anmar Arif

Natural disasters can cause significant damage to distribution networks, resulting in power outages and widespread disruptions. To expedite power restoration under these conditions, utilities commonly employ temporary measures such as portable generators and temporary distribution lines. However, deciding when and where to implement these measures can be both complex and costly. In this paper, we present a mixed-integer linear programming (MILP) model that optimizes the selection and deployment of portable power sources and temporary lines for repairing and restoring unbalanced distribution systems following natural disasters. Our approach features a pre-processing step that systematically identifies potential (i.e., candidate) lines and buses suitable for temporary measures, thereby confining the main optimization to a tractable set of possibilities. Through case studies on the IEEE 123-bus distribution system, we demonstrate that the proposed formulation yields cost-effective and timely post-disaster restoration plans, integrating portable substations, generators, temporary lines, crew routing, and network reconfiguration.

自然灾害会对配电网造成重大破坏，导致停电和大范围的破坏。为了在这种情况下加快电力恢复，公用事业公司通常采用临时措施，如便携式发电机和临时配电线路。然而，决定何时何地实施这些措施既复杂又昂贵。本文提出了一种混合整数线性规划（MILP）模型，该模型可以优化移动电源和临时线路的选择和部署，以用于自然灾害后不平衡配电系统的修复和恢复。我们的方法具有一个预处理步骤，系统地识别适合临时措施的潜在（即候选）线路和总线，从而将主要优化限制在一组可处理的可能性上。通过对IEEE 123总线配电系统的案例研究，我们证明了所提出的方案可以产生成本效益高且及时的灾后恢复计划，包括便携式变电站、发电机、临时线路、机组路由和网络重构。

引用次数: 0

Advanced power management of electric vehicle fast charging station united with multi-microgrid in autonomous and interconnected operations 自主互联运行下多微网联合的电动汽车快速充电站先进电源管理

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2026.112784

Hossien Faraji, Amir Khorsandi, Seyed Hossein Hosseinian

This article discusses the power management of two microgrids (MGs) in both off-grid and interconnected modes, incorporating a 200 kW DC fast charging (DCFC) station with four charging locations (CLs). In off-grid mode, each MG supplies power to its internal loads using distributed energy resources. The DCFC station, equipped with a battery energy storage (BES) unit, allows for the exchange of power with electric vehicles (EVs) parked at the CLs, enabling simultaneous charging or discharging of two EVs at 50 kW each. Each MG and the DCFC station operates with its own central control system (CCS). In off-grid mode, the CCS manages local controllers and connects to the main central control system (MCCS) for interconnected operations. In interconnected mode, the MCCS coordinates the charging or discharging of multiple EVs through peer-to-peer power exchange. If the BES fails, the DCFC station can draw power from the MGs, and a load-switching capability between the MGs enhances reliability. Non-linear simulations using MATLAB/SIMULINK demonstrate the effectiveness of the proposed strategies. In interconnected mode, the MGs stabilized the charging power of three EVs at 50 kW, while one EV dropped below 25 kW without their cooperation. The control system successfully transferred up to 25 kW of additional load between MGs during generation interruptions. Throughout, the DC bus voltage remained stable at 660 V, the load inverter frequency in MG 1 stayed between 49.9 and 50.1 Hz, and the RMS load voltage in MG 2 was maintained at 220 V.

本文讨论了两个微电网（mg）在离网和互联模式下的电源管理，其中包括一个200千瓦直流快速充电（DCFC）站和四个充电点（cl）。在离网模式下，每个MG通过分布式能源向其内部负载供电。DCFC充电站设有电池储能装置，可与停在电池储能站的电动汽车交换电力，使两辆电动汽车同时以50千瓦的功率充电或放电。每个MG和DCFC站都有自己的中央控制系统（CCS）。离网模式下，CCS管理本地控制器，并与主MCCS （central control system）互联运行。在互联模式下，MCCS通过点对点的电力交换，协调多辆电动汽车的充放电。当BES发生故障时，DCFC站可以从主控模块中获取电力，主控模块之间的负载切换能力提高了可靠性。利用MATLAB/SIMULINK进行了非线性仿真，验证了所提策略的有效性。在互联模式下，mg将3辆电动汽车的充电功率稳定在50 kW，而在没有mg的配合下，1辆电动汽车的充电功率降至25 kW以下。在发电中断期间，控制系统成功地在mg之间转移了高达25千瓦的额外负载。整个过程中，直流母线电压稳定在660v， MG 1中负载逆变频率保持在49.9 ~ 50.1 Hz之间，MG 2中负载电压均数维持在220v。

{"title":"Advanced power management of electric vehicle fast charging station united with multi-microgrid in autonomous and interconnected operations","authors":"Hossien Faraji, Amir Khorsandi, Seyed Hossein Hosseinian","doi":"10.1016/j.epsr.2026.112784","DOIUrl":"10.1016/j.epsr.2026.112784","url":null,"abstract":"<div><div>This article discusses the power management of two microgrids (MGs) in both off-grid and interconnected modes, incorporating a 200 kW DC fast charging (DCFC) station with four charging locations (CLs). In off-grid mode, each MG supplies power to its internal loads using distributed energy resources. The DCFC station, equipped with a battery energy storage (BES) unit, allows for the exchange of power with electric vehicles (EVs) parked at the CLs, enabling simultaneous charging or discharging of two EVs at 50 kW each. Each MG and the DCFC station operates with its own central control system (CCS). In off-grid mode, the CCS manages local controllers and connects to the main central control system (MCCS) for interconnected operations. In interconnected mode, the MCCS coordinates the charging or discharging of multiple EVs through peer-to-peer power exchange. If the BES fails, the DCFC station can draw power from the MGs, and a load-switching capability between the MGs enhances reliability. Non-linear simulations using MATLAB/SIMULINK demonstrate the effectiveness of the proposed strategies. In interconnected mode, the MGs stabilized the charging power of three EVs at 50 kW, while one EV dropped below 25 kW without their cooperation. The control system successfully transferred up to 25 kW of additional load between MGs during generation interruptions. Throughout, the DC bus voltage remained stable at 660 V, the load inverter frequency in MG 1 stayed between 49.9 and 50.1 Hz, and the RMS load voltage in MG 2 was maintained at 220 V.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112784"},"PeriodicalIF":4.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Hierarchical distributed scheduling of distribution networks using 5G base station clusters with fault-tolerant consensus-based allocation 基于容错共识分配的5G基站集群分布网络分层分布式调度

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2026.112775

Chongyu Liu, Hao Lei, Changyuan Zhou, Zhi-Wei Liu

The rapid integration of renewable energy into distribution networks has posed challenges due to its intermittency and uncertainty. This paper proposes a hierarchical control framework that leverages the regulation capability of 5G base station (BS) clusters to enhance energy management in distribution networks. The upper layer employs a distributed dual decomposition algorithm to achieve economic coordination between the distribution system operator (DSO) and BS clusters. The lower layer performs distributed allocation of traffic loads and backup energy storage system (BESS) power within each cluster, enabling fast and reliable tracking of bus-level power targets. Furthermore, the method incorporates a fault-tolerant consensus mechanism to ensure stable operation under random device faults. Numerical simulations on the IEEE 33 bus system verify that the proposed framework effectively reduces voltage violations, mitigates photovoltaic curtailment, and improves computational efficiency and robustness compared with existing consensus-based methods.

由于可再生能源的间歇性和不确定性，可再生能源在配电网中的快速整合带来了挑战。本文提出了一种分层控制框架，利用5G基站集群的调节能力来增强配电网的能源管理。上层采用分布式对偶分解算法，实现DSO和BS集群之间的经济协调。下层对各集群内的业务负载和BESS （backup energy storage system）功率进行分布式分配，实现对总线级功率目标的快速、可靠跟踪。此外，该方法还引入了容错共识机制，以确保在随机设备故障下稳定运行。在IEEE 33总线系统上的数值仿真验证了所提出的框架与现有的基于共识的方法相比，有效地减少了电压违规，减轻了光伏弃风，提高了计算效率和鲁棒性。

{"title":"Hierarchical distributed scheduling of distribution networks using 5G base station clusters with fault-tolerant consensus-based allocation","authors":"Chongyu Liu, Hao Lei, Changyuan Zhou, Zhi-Wei Liu","doi":"10.1016/j.epsr.2026.112775","DOIUrl":"10.1016/j.epsr.2026.112775","url":null,"abstract":"<div><div>The rapid integration of renewable energy into distribution networks has posed challenges due to its intermittency and uncertainty. This paper proposes a hierarchical control framework that leverages the regulation capability of 5G base station (BS) clusters to enhance energy management in distribution networks. The upper layer employs a distributed dual decomposition algorithm to achieve economic coordination between the distribution system operator (DSO) and BS clusters. The lower layer performs distributed allocation of traffic loads and backup energy storage system (BESS) power within each cluster, enabling fast and reliable tracking of bus-level power targets. Furthermore, the method incorporates a fault-tolerant consensus mechanism to ensure stable operation under random device faults. Numerical simulations on the IEEE 33 bus system verify that the proposed framework effectively reduces voltage violations, mitigates photovoltaic curtailment, and improves computational efficiency and robustness compared with existing consensus-based methods.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112775"},"PeriodicalIF":4.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Spatio-temporal graph fusion framework for accurate city-wide EV charging forecasting 面向城市电动汽车充电预测的时空图融合框架

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2026.112765

Xufeng Wu , Zuowei Chen , Hefang Jiang , Shoukang Luo , Honghao Liang , Siming Li , Chenyang Xu , Bojin Wang , Wensai Xuan , Yi Zhao , Lin Lin , Hao Wang

Accurately forecasting city-wide electric vehicle (EV) charging demand is hindered by complex spatio-temporal dependencies and the static fusion limitations of conventional models. To address these challenges, this paper proposes DyConfuse-Net, a novel Dynamic Context-driven Multi-component Hierarchical Fusion Network. DyConfuse-Net overcomes the rigidity of existing hybrid models by integrating four parallel prediction branches: Last Observation (LO), Bi-directional LSTM (BiLSTM), a Fully Connected Network (FCNN), and Differentiated Feature Projection (DFP). The core innovation is a dynamic fusion strategy that adaptively weights the contribution of each branch based on the specific spatio-temporal context of the input data. Experiments on the large-scale UrbanEV benchmark dataset demonstrate that DyConfuse-Net achieves state-of-the-art performance, with a Mean Absolute Percentage Error (MAPE) of 9.680% and a Mean Absolute Error (MAE) of 0.0155, significantly outperforming mainstream approaches. The results validate that our dynamic, hierarchical fusion framework offers a more robust and accurate solution for complex urban energy forecasting tasks.

复杂的时空依赖关系和传统模型静态融合的局限性阻碍了对城市电动汽车充电需求的准确预测。为了解决这些挑战，本文提出了dyconfusion - net，一种新颖的动态上下文驱动的多组件分层融合网络。DyConfuse-Net通过集成四个并行预测分支：最后一次观测（LO）、双向LSTM （BiLSTM）、完全连接网络（FCNN）和差分特征投影（DFP），克服了现有混合模型的刚性。其核心创新是一种动态融合策略，该策略根据输入数据的特定时空背景自适应地加权每个分支的贡献。在大规模UrbanEV基准数据集上的实验表明，DyConfuse-Net达到了最先进的性能，平均绝对百分比误差（MAPE）为9.680%，平均绝对误差（MAE）为0.0155，显著优于主流方法。结果验证了我们的动态、分层融合框架为复杂的城市能源预测任务提供了更稳健和准确的解决方案。

{"title":"Spatio-temporal graph fusion framework for accurate city-wide EV charging forecasting","authors":"Xufeng Wu , Zuowei Chen , Hefang Jiang , Shoukang Luo , Honghao Liang , Siming Li , Chenyang Xu , Bojin Wang , Wensai Xuan , Yi Zhao , Lin Lin , Hao Wang","doi":"10.1016/j.epsr.2026.112765","DOIUrl":"10.1016/j.epsr.2026.112765","url":null,"abstract":"<div><div>Accurately forecasting city-wide electric vehicle (EV) charging demand is hindered by complex spatio-temporal dependencies and the static fusion limitations of conventional models. To address these challenges, this paper proposes <strong>DyConfuse-Net</strong>, a novel Dynamic Context-driven Multi-component Hierarchical Fusion Network. DyConfuse-Net overcomes the rigidity of existing hybrid models by integrating four parallel prediction branches: Last Observation (LO), Bi-directional LSTM (BiLSTM), a Fully Connected Network (FCNN), and Differentiated Feature Projection (DFP). The core innovation is a dynamic fusion strategy that adaptively weights the contribution of each branch based on the specific spatio-temporal context of the input data. Experiments on the large-scale UrbanEV benchmark dataset demonstrate that DyConfuse-Net achieves state-of-the-art performance, with a Mean Absolute Percentage Error (MAPE) of 9.680% and a Mean Absolute Error (MAE) of 0.0155, significantly outperforming mainstream approaches. The results validate that our dynamic, hierarchical fusion framework offers a more robust and accurate solution for complex urban energy forecasting tasks.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112765"},"PeriodicalIF":4.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Optimal distribution network reconfiguration with non-adaptive overcurrent relay re-coordination 非自适应过流继电器重新协调的配电网优化重构

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-30 DOI: 10.1016/j.epsr.2025.112668

Rasoul Abdollahi, Reza Mohamadi Chabanloo

Reconfiguring distribution networks is a well-established strategy to reduce power losses and enhance operational efficiency. However, topology changes may alter the roles of primary and backup overcurrent relays and modify short-circuit currents, thereby compromising the reliability of the protection system. This paper addresses this challenge by formulating a joint optimization problem with two objectives: (i) loss-minimizing distribution network reconfiguration and (ii) simultaneous coordination of overcurrent relays. Instead of discarding configurations that violate protection constraints, the proposed method determines a single set of general relay settings and then optimizes reconfiguration under these settings while maintaining coordination requirements. The framework is non-adaptive, requires no telecommunication links, and thus offers practical advantages over adaptive schemes. A Pareto-based multi-objective optimization is developed, combining a Genetic Algorithm (GA) for network reconfiguration with a Linear Programming (LP) model for relay re-coordination. Simulation results on the standard 33-bus distribution system demonstrate that the method significantly expands the set of feasible configurations, achieves notable loss reduction, and ensures reliable relay coordination across multiple operating states.

重新配置配电网是一种行之有效的策略，可以减少电力损失，提高运行效率。然而，拓扑结构的变化可能会改变主、备过流继电器的作用，改变短路电流，从而影响保护系统的可靠性。本文通过制定一个具有两个目标的联合优化问题来解决这一挑战：(i)损失最小化配电网络重构和（ii）同时协调过流继电器。该方法不是丢弃违反保护约束的配置，而是确定一组通用继电器设置，然后在保持协调要求的情况下在这些设置下优化重新配置。该框架是非自适应的，不需要电信链路，因此比自适应方案具有实际优势。将遗传算法（GA）与线性规划（LP）模型相结合，提出了一种基于pareto的多目标优化算法。对标准33母线配电系统的仿真结果表明，该方法显著扩展了可行配置集，显著降低了损耗，保证了继电器在多运行状态下的可靠协调。

{"title":"Optimal distribution network reconfiguration with non-adaptive overcurrent relay re-coordination","authors":"Rasoul Abdollahi, Reza Mohamadi Chabanloo","doi":"10.1016/j.epsr.2025.112668","DOIUrl":"10.1016/j.epsr.2025.112668","url":null,"abstract":"<div><div>Reconfiguring distribution networks is a well-established strategy to reduce power losses and enhance operational efficiency. However, topology changes may alter the roles of primary and backup overcurrent relays and modify short-circuit currents, thereby compromising the reliability of the protection system. This paper addresses this challenge by formulating a joint optimization problem with two objectives: (i) loss-minimizing distribution network reconfiguration and (ii) simultaneous coordination of overcurrent relays. Instead of discarding configurations that violate protection constraints, the proposed method determines a single set of general relay settings and then optimizes reconfiguration under these settings while maintaining coordination requirements. The framework is non-adaptive, requires no telecommunication links, and thus offers practical advantages over adaptive schemes. A Pareto-based multi-objective optimization is developed, combining a Genetic Algorithm (GA) for network reconfiguration with a Linear Programming (LP) model for relay re-coordination. Simulation results on the standard 33-bus distribution system demonstrate that the method significantly expands the set of feasible configurations, achieves notable loss reduction, and ensures reliable relay coordination across multiple operating states.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112668"},"PeriodicalIF":4.2,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Resilient control of DC microgrids under false data injection attacks using adaptive sliding mode unknown input observer 基于自适应滑模未知输入观测器的虚假数据注入攻击下直流微电网弹性控制

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-29 DOI: 10.1016/j.epsr.2026.112781

Ziji Su, Yile Dai, Jiawei Zhang, Miao Yu

The rapid advancement of communication and control technologies has significantly enhanced the intelligence of DC microgrids, yet it also introduces cybersecurity risks such as false data injection attacks (FDIAs). Existing studies on FDIA mitigation still exhibit limitations. For example, their reliance on distributed communication leads to slow convergence. Moreover, most methods assume an ideal primary layer, neglecting actuator faults and parameter perturbations. To this end, this paper proposes a resilient control scheme: at the secondary layer, an auxiliary reference system (ARS) is designed to generate reference values without requiring data transmission from the primary layer, while the primary layer employs a control law based on the super-twisting algorithm to achieve robustness against actuator faults and parameter perturbations. Furthermore, a decentralized adaptive sliding mode unknown input observer is developed to reconstruct ARS states for FDIA mitigation. Compared with existing approaches, the proposed strategy demonstrates superior FDIA suppression capability and robustness against non-ideal conditions at the primary layer. The proposed scheme achieves control recovery within 15 ms, which is significantly faster than the comparative methods (0.45 s and 1.26 s, respectively). Stability analysis and parameter design guidelines are provided. Hardware-in-the-loop results validate the effectiveness and superiority of the proposed scheme.

通信和控制技术的快速发展大大提高了直流微电网的智能化水平，但同时也带来了虚假数据注入攻击（FDIAs）等网络安全风险。现有的减缓FDIA的研究仍然显示出局限性。例如，它们对分布式通信的依赖导致了缓慢的收敛。此外，大多数方法假设一个理想的初级层，忽略了执行器故障和参数摄动。为此，本文提出了一种弹性控制方案：在第二层设计辅助参考系统（ARS）生成参考值，无需从第二层传输数据，而第二层采用基于超扭转算法的控制律实现对执行器故障和参数扰动的鲁棒性。在此基础上，提出了一种分散自适应滑模未知输入观测器来重建ARS状态，以缓解FDIA。与现有方法相比，该策略具有较好的FDIA抑制能力和对初级层非理想条件的鲁棒性。该方案在15 ms内实现了控制恢复，明显快于对比方法（分别为0.45 s和1.26 s）。提供了稳定性分析和参数设计指南。硬件在环实验结果验证了该方案的有效性和优越性。

{"title":"Resilient control of DC microgrids under false data injection attacks using adaptive sliding mode unknown input observer","authors":"Ziji Su, Yile Dai, Jiawei Zhang, Miao Yu","doi":"10.1016/j.epsr.2026.112781","DOIUrl":"10.1016/j.epsr.2026.112781","url":null,"abstract":"<div><div>The rapid advancement of communication and control technologies has significantly enhanced the intelligence of DC microgrids, yet it also introduces cybersecurity risks such as false data injection attacks (FDIAs). Existing studies on FDIA mitigation still exhibit limitations. For example, their reliance on distributed communication leads to slow convergence. Moreover, most methods assume an ideal primary layer, neglecting actuator faults and parameter perturbations. To this end, this paper proposes a resilient control scheme: at the secondary layer, an auxiliary reference system (ARS) is designed to generate reference values without requiring data transmission from the primary layer, while the primary layer employs a control law based on the super-twisting algorithm to achieve robustness against actuator faults and parameter perturbations. Furthermore, a decentralized adaptive sliding mode unknown input observer is developed to reconstruct ARS states for FDIA mitigation. Compared with existing approaches, the proposed strategy demonstrates superior FDIA suppression capability and robustness against non-ideal conditions at the primary layer. The proposed scheme achieves control recovery within 15 ms, which is significantly faster than the comparative methods (0.45 s and 1.26 s, respectively). Stability analysis and parameter design guidelines are provided. Hardware-in-the-loop results validate the effectiveness and superiority of the proposed scheme.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112781"},"PeriodicalIF":4.2,"publicationDate":"2026-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Noise-resilient fault and power quality event detection, classification, and localization in renewable-integrated distribution networks using EMD–SVM 基于EMD-SVM的可再生集成配电网噪声恢复故障和电能质量事件检测、分类和定位

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-28 DOI: 10.1016/j.epsr.2026.112774

Dhanunjayudu N , Eswaramoorthy K. Varadharaj , Mohana Rao M , Krishnaiah J

An effective framework is presented for sub-cycle detection, classification, and localization of faults and power-quality (PQ) events in renewable-integrated distribution systems. Using only single-end substation measurements, the proposed EMD–SVM/SVR-based approach enables fast event diagnosis with a decision latency of 0.33 cycle. Robustness is achieved through lightweight preprocessing and short-window analysis, ensuring stable operation under nonstationary and noisy conditions. The method attains a classification accuracy of 97.8% at 20 dB SNR and fault resistances up to 7 Ω, while reliably distinguishing PQ disturbances and switching events with zero false detections. Comparative evaluation against existing EMD, DWT, and DMD based techniques demonstrates superior detection speed, robustness, and real-time feasibility, making the proposed framework suitable for practical protection of renewable-integrated distribution networks.

提出了可再生集成配电系统故障和电能质量事件的子周期检测、分类和定位的有效框架。仅使用单端变电站测量，所提出的基于EMD-SVM / svr的方法可以实现快速事件诊断，决策延迟为0.33个周期。鲁棒性通过轻量级预处理和短窗口分析实现，确保在非平稳和噪声条件下稳定运行。该方法在信噪比为20 dB，故障电阻高达7 Ω的情况下，分类准确率达到97.8%，同时可靠地区分PQ干扰和开关事件，零误检。与现有的基于EMD、DWT和DMD技术的比较评估表明，该框架具有更高的检测速度、鲁棒性和实时性，适用于可再生综合配电网的实际保护。

引用次数: 0

A geospatial approach to detect and prioritize regularization of hook-based non-technical losses 一种地理空间方法来检测和优先考虑基于钩子的非技术损失的正则化

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-27 DOI: 10.1016/j.epsr.2026.112769

Matheus Menezes , Italo F.S. Silva , Felipe M. Feyh , Igor G. Vargas , Manuela B. de Lima , Pedro H.L. Dias , Romulo R. Viana , Guilherme V.B. Rubim , Carlos J.S. Moura , Auriane A.M. dos Santos , Lucas P.A. Pinheiro , Patrick C. Araujo , Pedro V. Bernhard , Thiago P. Freire , Joana K.A. Silva , Eduardo F.P. Dutra , Levi C. Santos , Weslley K.R. Figueredo , João D.S. Almeida , Luis J.E.R. Cabrejos , Aristófanes C. Silva

Illegal electricity connections are a major source of non-technical losses, especially in areas lacking advanced metering infrastructure. These connections increase operational costs, reduce grid reliability, and transfer economic overload to regular consumers. We propose a building-level, hybrid geospatial methodology to detect and prioritize the regularization of potential clandestine connections caused by hook-based theft. Our approach combines satellite imagery, geospatial filtering, computer vision, and human-in-the-loop validation to identify and cluster irregular buildings based on their financial return on regularization. Validated with real data from Pará, Brazil, the method achieved an average accuracy of 83.05% across over 10,000 inspected sites, with regional peaks of 94.%. Clustering algorithms such as DBSCAN efficiently grouped illegal connections by geographic and financial features, enabling more effective allocation of on-site operations. This framework is adaptable to distinct scenarios, and supports large-scale, data-driven non-technical losses mitigation, improving revenue recovery, grid safety, and tariff fairness.

非法接电是非技术损失的主要来源，特别是在缺乏先进计量基础设施的地区。这些连接增加了运营成本，降低了电网的可靠性，并将经济负荷转移给了普通消费者。我们提出了一种建筑级的混合地理空间方法来检测和优先考虑由基于钩子的盗窃引起的潜在秘密连接的正规化。我们的方法结合了卫星图像、地理空间过滤、计算机视觉和人在环验证，根据正则化的财务回报来识别和聚类不规则建筑。通过巴西par的真实数据验证，该方法在超过10,000个被检查地点的平均准确率为83.05%，区域峰值为94%。DBSCAN等聚类算法可以根据地理和财务特征对非法连接进行有效分组，从而更有效地分配现场操作。该框架可适应不同的场景，并支持大规模、数据驱动的非技术损失缓解，提高收入恢复、电网安全和电价公平。

{"title":"A geospatial approach to detect and prioritize regularization of hook-based non-technical losses","authors":"Matheus Menezes , Italo F.S. Silva , Felipe M. Feyh , Igor G. Vargas , Manuela B. de Lima , Pedro H.L. Dias , Romulo R. Viana , Guilherme V.B. Rubim , Carlos J.S. Moura , Auriane A.M. dos Santos , Lucas P.A. Pinheiro , Patrick C. Araujo , Pedro V. Bernhard , Thiago P. Freire , Joana K.A. Silva , Eduardo F.P. Dutra , Levi C. Santos , Weslley K.R. Figueredo , João D.S. Almeida , Luis J.E.R. Cabrejos , Aristófanes C. Silva","doi":"10.1016/j.epsr.2026.112769","DOIUrl":"10.1016/j.epsr.2026.112769","url":null,"abstract":"<div><div>Illegal electricity connections are a major source of non-technical losses, especially in areas lacking advanced metering infrastructure. These connections increase operational costs, reduce grid reliability, and transfer economic overload to regular consumers. We propose a building-level, hybrid geospatial methodology to detect and prioritize the regularization of potential clandestine connections caused by hook-based theft. Our approach combines satellite imagery, geospatial filtering, computer vision, and human-in-the-loop validation to identify and cluster irregular buildings based on their financial return on regularization. Validated with real data from Pará, Brazil, the method achieved an average accuracy of 83.05% across over 10,000 inspected sites, with regional peaks of 94.%. Clustering algorithms such as DBSCAN efficiently grouped illegal connections by geographic and financial features, enabling more effective allocation of on-site operations. This framework is adaptable to distinct scenarios, and supports large-scale, data-driven non-technical losses mitigation, improving revenue recovery, grid safety, and tariff fairness.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112769"},"PeriodicalIF":4.2,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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

Wind power scenario generation via multi-scale condition adaptive diffusion model 基于多尺度条件自适应扩散模型的风电情景发电

IF 4.2 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Electric Power Systems Research

Pub Date : 2026-01-27 DOI: 10.1016/j.epsr.2026.112753

Jiawei Zhang , Shuhao Liu , Zeyi Shi , Yuancheng Li

The large-scale integration of wind power poses significant challenges to power system operation, as the intermittency and stochasticity of wind power generation can adversely affect grid stability. While diffusion models have been applied to wind power scenario generation, their performance remains sensitive due to an inadequate capacity for processing conditional information. To address these issues, this manuscript proposes an multi-scale condition adaptive diffusion model(MS-CADM). Specifically, a multi-scale embedding network is proposed to extract complex conditions at multiple hierarchical levels, while a time step adaptation Transformer architecture is introduced to decouple the time-step condition unique to diffusion models from the external control conditions, which makes model to learn different controllable conditions accurately. During training, a learnable variance term is incorporated to enhance posterior expressiveness, and a random conditional masking strategy is applied for regularization to stabilize training. On Global Energy Forecasting Competition 2014 (GEFCom2014) wind power dataset, MS-CADM reduces MAE by 4.26% and RMSE by 1.91% compared to state-of-the-art methods and achieves competitive performance across six metrics, demonstrating its ability to effectively leverage conditional information and achieve more efficient convergence.

风电的大规模并网对电力系统的运行提出了重大挑战，风力发电的间歇性和随机性会对电网的稳定性产生不利影响。虽然扩散模型已应用于风力发电情景，但由于条件信息处理能力不足，其性能仍然敏感。为了解决这些问题，本文提出了一个多尺度条件自适应扩散模型（MS-CADM）。具体而言，提出了一种多尺度嵌入网络，在多个层次上提取复杂条件；引入时间步长自适应Transformer架构，将扩散模型特有的时间步长条件与外部控制条件解耦，使模型能够准确地学习不同的可控条件。在训练过程中，引入可学习方差项增强后验表达性，并采用随机条件掩蔽策略进行正则化以稳定训练。在2014年全球能源预测大赛（GEFCom2014）风电数据集上，与最先进的方法相比，MS-CADM将MAE降低了4.26%，RMSE降低了1.91%，并在六个指标上取得了具有竞争力的表现，证明了其有效利用条件信息和实现更有效收敛的能力。

{"title":"Wind power scenario generation via multi-scale condition adaptive diffusion model","authors":"Jiawei Zhang , Shuhao Liu , Zeyi Shi , Yuancheng Li","doi":"10.1016/j.epsr.2026.112753","DOIUrl":"10.1016/j.epsr.2026.112753","url":null,"abstract":"<div><div>The large-scale integration of wind power poses significant challenges to power system operation, as the intermittency and stochasticity of wind power generation can adversely affect grid stability. While diffusion models have been applied to wind power scenario generation, their performance remains sensitive due to an inadequate capacity for processing conditional information. To address these issues, this manuscript proposes an multi-scale condition adaptive diffusion model(MS-CADM). Specifically, a multi-scale embedding network is proposed to extract complex conditions at multiple hierarchical levels, while a time step adaptation Transformer architecture is introduced to decouple the time-step condition unique to diffusion models from the external control conditions, which makes model to learn different controllable conditions accurately. During training, a learnable variance term is incorporated to enhance posterior expressiveness, and a random conditional masking strategy is applied for regularization to stabilize training. On Global Energy Forecasting Competition 2014 (GEFCom2014) wind power dataset, MS-CADM reduces MAE by 4.26% and RMSE by 1.91% compared to state-of-the-art methods and achieves competitive performance across six metrics, demonstrating its ability to effectively leverage conditional information and achieve more efficient convergence.</div></div>","PeriodicalId":50547,"journal":{"name":"Electric Power Systems Research","volume":"255 ","pages":"Article 112753"},"PeriodicalIF":4.2,"publicationDate":"2026-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146078806","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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