最新刊期
卷 50 ， 期 18 ， 2026
- 摘要：With the opening-up and development of incremental distribution businesses and integrated application of digital-intelligent technologies, energy flow and information flow in the new distribution system frequently interact and are complexly coupled among multiple entities, causing cross-domain cyber-physical operational risks. To address the challenges of coordinated operation and resilience enhancement of multiple entities under cross-domain operational risks, a distributed-coordinated regulation mode for the efficiency improvement of energy-information is explored in a step-by-step manner across three levels: data, mechanism and application. At the data level, the spatio-temporal multi-dimensional value of measurement data is extracted to jointly analyze cross-domain operational risks among multiple entities. At the mechanism level, heterogeneous energy and information resources in multiple entities are coordinated to establish a deep defense system as well as a secure and trustworthy interactive decision-making mechanism. At the application level, the system’s resilient operation capability is analyzed from the perspective of cyber‑physical coupling, enabling closed‑loop near‑optimal operation in complicated dynamic environments. Building on existing studies, emerging social-system risks and future study directions are discussed from macro and micro perspectives.关键词：new distribution system;multiple entities;cyber-physical coupling;distributed-coordinated regulation;operational decision-making;resilience28|2|0更新时间：2026-09-22
Review·Perspective
- 摘要：The large-scale clean energy base is a critical component of China’s new power system. Their inherent characteristics—strong uncertainty, low flexibility, and numerous controllable resources—pose dual challenges of timeliness and complexity for intra-day dispatching. This paper proposes an intra-day scheduling method for wind-solar-hydro-thermal-storage multi-energy complementarity based on improved constrained reinforcement learning. First, an intra-day optimal scheduling model for power system is established, incorporating the coordination of wind, solar, hydro, thermal, storage, and DC transmission, and reformulated as a constrained Markov decision process. Second, a policy network architecture embedded with temporal memory units is designed to capture the temporal dependencies of system states, while self-supervised auxiliary tasks are introduced into the value network to enhance feature representation capability. Third, the Lagrangian relaxation method is employed to derive the optimal scheduling policy under safety constraints, and an action correction module based on expert experience is incorporated to jointly ensure satisfaction of both single-step constraints and long-term coupling constraints. Finally, the effectiveness of the proposed method is validated using a modified IEEE 39-bus system and an actual power grid.关键词：clean energy;new power system;multi-energy complementarity;wind-solar-hydro-thermal-storage;intra-day scheduling;constrained reinforcement learning17|2|0更新时间：2026-09-22
- 摘要：There are various sources of carbon emissions in the photovoltaic-storage-charging stations and a lack of coordinated scheduling strategies for different types of electric energy, resulting in the underutilization of low-carbon resources in the station. Therefore, a two-stage low-carbon scheduling strategy for photovoltaic-storage-charging stations and electric vehicles (EVs) considering dynamic electricity prices is proposed. Firstly, the electric energy in the photovoltaic-storage-charging stations is classified according to whether it contains carbon, and the dynamic electricity price based on clean-carbon electricity coordination is formulated to guide the charging and discharging behaviors of EVs. Secondly, the carbon emission in the stations is modeled in detail. The carbon emissions generated by EVs and energy storage discharging and the dynamic changes of their carbon emission intensity are considered, while considering the carbon emission of electricity purchase in the photovoltaic-storage-charging stations. Finally, a station-vehicle two-stage joint low-carbon scheduling model based on the two-region commuting of EVs is established, and the real-time deviation is adjusted for the intra-day fast-charging load and renewable energy fluctuation. Simulation results demonstrate that the proposed strategy can effectively improve the local consumption rate of photovoltaics. Compared with the scheduling in a single photovoltaic-storage-charging station, the cross-regional scheduling of EVs can realize the spatial-temporal transfer of photovoltaics in the stations and further reduce the carbon emission.关键词：photovoltaic-storage-charging station;electric vehicle (EV);low-carbon scheduling;clean-carbon electricity coordination;dynamic electricity price;carbon emission16|0|0更新时间：2026-09-22
- 摘要：Distribution networks with the participation of distributed resource aggregators (DRAs) are confronted with problems including insufficient proactivity in congestion management and low efficiency of cross-entity coordination,which necessitates development of a rapid dispatch approach for multi-entity collaboration. To address this issue, this paper constructs a day-ahead and intra-day two-stage coordinated scheduling framework. In the day-ahead stage, this paper conducts differentiated modeling for DRAs. It introduces the asynchronous parallel alternating direction method of multipliers (ADMM). This method achieves a rapid solution while protecting the privacy of all parties through the limited interaction of boundary power and price signals. In the intra-day stage, this paper proposes an event-driven rolling optimization strategy. It designs a congestion relief correction price that incentivizes and guides DRAs to dynamically adjust the exchanged power. When necessary, it achieves congestion elimination through minimum load shedding. Case studies demonstrate that the proposed two-stage coordinated management method significantly enhances the capability of rapidly eliminating distribution network congestion. The asynchronous parallel ADMM reduces the computation time by 82.1% compared with the conventional synchronous parallel scheme, which verifies the feasibility and superiority of the proposed method.关键词：distribution network;congestion management;distributed resource aggregators (DRA);coordinated scheduling;distributed optimization;event-driven;alternating direction method of multipliers10|0|0更新时间：2026-09-22
-
Analysis of Demand Response Capability and Potential in Distribution Networks Based on Deep Learning
摘要：In the background of the continuous advancement of electricity market-oriented reform, in order to improve the utilization efficiency of demand response in distribution networks, a method for analyzing demand response capability and potential in distribution networks based on deep learning is proposed. Firstly, a convolutional neural network (CNN)-bi-directional long short-term memory (BiLSTM)-Transformer two-layer prediction model is constructed to achieve accurate and dynamic prediction of demand response capability. Secondly, load models for various industries and user types are established. The demand response potential is analyzed from three perspectives of user side, distribution network side, and development trends. A long- and short-term demand response potential quantification method based on spatial envelope domain is proposed. Meanwhile, with the objective of minimizing the comprehensive cost of the distribution network, an optimization model based on short-term demand response potential is constructed, and a targeted incentive pricing strategy is proposed. Finally, simulation case studies verify the effectiveness and superiority of the proposed method in prediction accuracy, potential assessment, and market application.关键词：distribution network;demand response;electricity market;deep learning;dynamic forecasting;incentive price;Transformer8|0|0更新时间：2026-09-22 - 摘要：In the day-ahead invitation-based demand response, the equivalent load forecasting results of virtual power plants (VPPs) directly affect their bidding decisions and operation revenues. However, traditional forecasting methods aim at maximizing forecasting accuracy but neglect the asymmetric relationship between forecasting deviations and decision-induced revenue loss, which may lead to revenue reduction. To address this, this paper proposes a revenue-oriented multi-period equivalent load forecasting method for VPPs participating in demand response. Firstly, this paper reveals the inconsistency between equivalent load forecasting accuracy improvement and VPP revenue enhancement, and constructs a two-stage decision-making model for VPP demand response. Secondly, a framework of “accuracy-oriented training + decision-oriented fine-tuning” is designed, utilizing a surrogate network to effectively characterize the revenue loss caused by forecasting deviations of the decision-making model. Finally, to tackle the learning difficulty for surrogate networks under multi-period coupling decision-making, a local decision-oriented loss function learning method with multi-period decoupling representation is proposed to approximately characterize the multi-period local decision-oriented loss function by the aggregated learning of single-period local losses. Numerical case results demonstrate that, compared with accuracy-oriented methods, the obtained forecasting results can significantly enhance the demand response revenue of VPPs.关键词：virtual power plant;demand response;revenue;load forecasting;decision making;multi-period decoupling;surrogate network8|0|0更新时间：2026-09-22
- 摘要：Integrated demand response (IDR) serves as an effective means to achieve the supply-demand balance in the context of the new power system. To improve the accuracy of user response models for IDR and enable integrated energy system providers (IESPs) to formulate more economical and efficient incentive strategies, this paper incorporates the influences of users’ anchoring effect and ratchet effect on their response behaviors. First, an anchoring effect matrix is established to effectively model the user anchoring effect. Then, ratchet coefficients are introduced into the conventional user response model to characterize the ratchet effect. Meanwhile, an IESP optimization model considering incomplete information is constructed. Based on the maximum likelihood theory and the Newton iteration principle, the proposed model addresses the incomplete information problem caused by the unobservability of user ratchet coefficients. The proposed model is established as a bi-level optimization problem with users and IESPs as the two decision-making subjects, which is further transformed into a single-level optimization problem for solution. Simulation results demonstrate that the proposed model can improve the benefits of users participating in IDR, cut down the operational cost of IESPs, and mitigate the power imbalance risk of the power grid. It can not only effectively characterize users’ anchoring and ratchet effects but also efficiently handle the incomplete information on the user side.关键词：integrated energy system (IES);integrated demand response (IDR);supply-demand balance;anchoring effect;ratchet effect;incomplete information8|0|0更新时间：2026-09-22
- 摘要：Accurate operation state estimation for urban fast-charging stations is a crucial prerequisite for the efficient utilization of electric vehicle charging networks. Existing studies lack the consideration of the actual operation mechanisms of fast-charging stations, failing to explore the spatio-temporal interaction characteristics of multi-station operation states. To address this, by transforming state estimation into a sample generation process that integrates operation knowledge and historical data of fast-charging stations, this paper proposes an operation state estimation method for urban fast-charging stations based on knowledge-embedded hypergraph deep learning architecture. First, the dynamic operation states of fast-charging stations and their multi-spatial distribution are represented as a hypergraph model, and the temporal convolution attention network and hypergraph convolution network are used to capture temporal similarity and spatial correlation in graph-structured feature information. Second, operation strategies of fast-charging stations, physical rules, and constraints are decoupled into a knowledge loss function, which is integrated into the sample learning process of generative adversarial networks to construct a knowledge-data dual-driven loss function. Through the “encoding-decoding” paradigm, the conditional probability distribution of samples is implicitly learned to estimate the operation state of individual charging piles across different fast-charging stations. Finally, using the actual road network and operating fast-charging stations in Nanjing, China, as a case study, simulation results demonstrate that the proposed method can accurately characterize the spatio-temporal distribution characteristics of operation states for fast-charging stations, providing theoretical support for subsequent electric vehicle charging scheduling.关键词：electric vehicle (EV);charging network;fast-charging station;state estimation;operation knowledge;hypergraph deep learning;generative adversarial network8|0|0更新时间：2026-09-22
- 摘要：With the dramatic evolution of power system operation modes and the growing coupling between transmission and distribution networks, independent state estimation methods for transmission and distribution networks can no longer meet the requirements of unified dispatching and refined management of transmission and distribution networks, giving rise to coordinated transmission-distribution state estimation. Considering the drawbacks of centralized state estimation, such as heavy computational burden and asynchronous measurement updates between transmission and distribution networks, this paper proposes an asynchronous state estimation method for transmission and distribution networks based on Gauss-Newton local belief propagation. Firstly, a dynamic partitioned factor graph model is constructed according to the received boundary information of transmission and distribution networks. On this basis, a Gauss-Newton local belief propagation algorithm is developed to realize dynamic local updating of the factor graph model and fully distributed state estimation for the transmission network, alleviating the computational burden of the transmission network control center. Secondly, the asynchronous state estimation method based on Gauss-Newton local belief propagation is formulated to achieve real-time asynchronous information updating of each boundary node in the transmission network, which fully exploits the distributed and local computing advantages of the algorithm. Under both normal and bad measurement scenarios, the proposed method can effectively reduce the number of message transmitted of the factor graph model while guaranteeing estimation accuracy, thus saving computational resources.关键词：transmission network;distribution network;state estimation;factor graph;Gauss-Newton local belief propagation6|0|0更新时间：2026-09-22
- 摘要：As the scale of high-voltage direct current (HVDC) expands, the short-circuit ratio of the receiving-end power system gradually approaches or exceeds its critical value, which may easily trigger DC commutation failures and other issues, thereby endangering the stable operation of the power system. Coordinated planning of source and grid can effectively improve system stability by rationally deploying power sources and lines. To address this problem, this paper proposes a source-grid coordinated planning method considering the hybrid multi-infeed DC equivalent short-circuit ratio (ESCR) constraints. First, the explicit expression of the critical equivalent short-circuit ratio (CESCR) is derived based on the coupling relationship between nodal voltage sensitivity factors with respect to active and reactive power and the CESCR, and an ESCR constraint model is established. Second, by coupling the source-grid coordinated planning model, the evaluation model of short-circuit ratio, and the generation model of short-circuit ratio constraint cut, a coordinated planning model of source and grid with adaptive ESCR adjustment capability is developed. Verification results on the IEEE 24-bus test system and a practical regional power system demonstrate that CESCR can accurately quantify the voltage support capability of hybrid multi-infeed DC. The short-circuit ratio level of the receiver-end system can be elevated by optimizing the network topology, while the coordinated planning scheme of source and grid can optimize power output of generating units and reduce system operating costs.关键词：high-voltage direct current (HVDC);multi-feed DC;coordinated planning;equivalent short-circuit ratio (ESCR);voltage support;sensitivity factor8|0|0更新时间：2026-09-22
- 摘要：With the increasing penetration capacity of power electronic devices such as high voltage direct current (HVDC) converters in power systems, their influence on system frequency strength gradually becomes prominent. However, components such as dead zones and power limitations in DC frequency limit control (FLC) introduce significant nonlinear characteristics in the system frequency dynamics. Conventional methods based on linear power-frequency models fail to accurately analyze the system frequency response and quantitatively evaluate the frequency strength. To this end, this paper proposes an analytical method for the frequency response of power systems that accounts for the nonlinear characteristics of the DC FLC, enabling accurate quantitative evaluation of system frequency strength. Firstly, a common-mode frequency model incorporating FLC is established, in which the frequency regulation dynamics of different types of devices in the system are approximately described using a unified inertia-damping-spring coefficient structure. Secondly, the frequency response is divided into several stages according to the instants when the FLC dead zone or power limitation is reached. The boundary conditions of the frequency response in each stage are analyzed, and the frequency trajectories of each stage are analytically derived. Finally, by synthesizing the responses of all stages, the characteristics of the maximum frequency deviation are quantitatively evaluated, a frequency strength index is established, and the impacts of FLC parameters such as dead zones and power limitations are analyzed. Simulation results verify the effectiveness of the proposed method.关键词：DC frequency limit control;dead zone;power limitation;frequency strength;frequency response;inertia;damping;spring coefficient9|0|0更新时间：2026-09-22
- 摘要：Traditional cars consume fossil fuels while emitting large amounts of carbon and nitrogen oxides, which affects air quality. In contrast, fuel cell hybrid electric vehicle (FCHEV) powered by renewable energy sources are gaining increasing market favor due to their environmental advantages. The stable operation of FCHEV relies on the sustainable management of energy in its hybrid energy storage system (HESS). To this end, a structurally simple adaptive continuous sliding mode control strategy is proposed for a FCHEV-HESS consisting of fuel cells, hybrid photoelectrochemical cells, photovoltaic cells, batteries, and supercapacitors, aiming to improve energy conversion efficiency and robustness. First, a DC-bus-based internal model of the HESS is established, and the output current characteristics of each module are analyzed via power conservation. Then, a HESS energy management strategy is designed according to FCHEV load variations to achieve sustainable power supply under complex driving conditions. Next, adaptive gains and a super-twisting sliding mode algorithm are introduced to design controllers for each module, dynamically regulating the DC-bus voltage and load currents of the HESS. Finally, a hardware-in-the-loop experimental platform is built to validate the effectiveness of the proposed strategy. Results show that, compared with conventional sliding mode strategies, the proposed method improves the current response speed of the FCHEV-HESS by 29.5% and reduces the current tracking error under continuous speed variations by 44.2%.关键词：renewable energy sources;electric vehicle;fuel cell;hybrid energy storage system (HESS);sliding mode control6|0|0更新时间：2026-09-22
Basic Research
- 摘要：The power system optimization engine, as a fundamental mathematical tool for the planning and operation of power systems and the clearing of electricity markets, plays a crucial supporting role in the construction of new power systems and the effective operation of electricity markets. This paper introduces the development and construction of the power system optimization engine and its application progress in China Southern Power Grid (CSG). First, the framework and functional modules of the optimization engine are presented, along with the generic large-scale optimization methods and the principles of efficient solving techniques for power system optimization adopted by the engine. Then, case studies of the power system optimization engine based on the application scenarios in CSG are introduced, and the solving efficiency and accuracy are verified. Finally, the prospects of the optimization engine in power systems are analyzed.关键词：power system operation;power system planning;electricity market;optimization engine;mixed-integer programming;unit commitment9|0|0更新时间：2026-09-22
- 摘要：To address the strong coupling between resource allocation and collaborative scheduling in collaborative inspection of distribution lines by vehicles and unmanned aerial vehicles (UAVs), and the difficulty in balancing operational efficiency and resource utilization, this paper proposes a joint optimization method for resource allocation and execution scheduling. First, a unified modeling framework for resource allocation and path scheduling is constructed, in which the number of UAVs and the layout of parking points are transformed from a priori parameters into optimization variables. Then, within a bi-level multi-objective evolutionary framework, the outer level performs resource allocation search, while the inner level handles vehicle routing and UAV task scheduling. Based on the execution results, a feedback evaluation mechanism for resource allocation is established to achieve joint optimization between resource allocation and execution scheduling. Finally, the performance of the proposed method is verified through typical distribution inspection scenarios, engineering application scenarios, and UAV failure conditions. Results demonstrate that the proposed method reduces the total operation time while effectively suppressing resource redundancy. Compared with the non-dominated sorting genetic algorithm Ⅱ, the total operation time and comprehensive cost are reduced by approximately 3.7% and 5.4%, respectively, and the resource redundancy rate decreases from 11.6% to 8.0%. Under high UAV failure ratios, the system recovery scheduling cost increases steadily, which validates the effectiveness and robustness of the proposed method.关键词：distribution network;line inspection;unmanned aerial vehicle (UVA);inspection vehicle;resource allocation;multi-objective optimization;path scheduling;feedback11|0|0更新时间：2026-09-22
- 摘要：To address the challenges of insufficient multi-modal data fusion and sparse sample data in offshore wind power operation and maintenance scenarios, this paper proposes a technical scheme that deeply integrates knowledge graph and few-shot learning, focusing on resolving the key problems of knowledge organization, few-shot entity recognition, and knowledge completion in the offshore wind power field. Firstly, a four-dimensional offshore wind power knowledge graph encompassing environment, equipment, fault, and operation and maintenance is constructed to achieve systematic organization of heterogeneous data. Secondly, a integrated framework combining meta-learning and contrastive self-supervised learning is designed, which decomposes query intentions into five sub-tasks: fault diagnosis, operational guidance, information retrieval, environmental correlation, and comprehensive information. A contrastive loss function is employed to optimize feature representations, thereby enhancing entity recognition capability in few-shot scenarios. Thirdly, an edge-labeled graph neural network is introduced to explicitly model edge features such as semantic relations, similarity, and confidence between entities. Through iterative updating of node and edge features, weighted voting-based completion of missing entities in the knowledge graph is achieved. Finally, a closed-loop iterative mechanism based on the knowledge graph is established to ensure knowledge matching accuracy and model generalization performance. Experimental results demonstrate that, under few-shot data conditions, the proposed technical scheme achieves significant improvements in entity recognition accuracy, association completion performance, and query rewriting accuracy over traditional methods, providing technical support for knowledge engineering in the offshore wind power field.关键词：offshore wind power;operation and maintenance;knowledge graph;few-shot learning;artificial intelligence (AI)8|0|0更新时间：2026-09-22
- 摘要：To address issues such as the excessive number of candidate line combinations and the difficulty in defining obstacle zones in the topology optimization of power collection systems for offshore wind farms, a dimensionality reduction optimization method for power collection systems of offshore wind farms is proposed, incorporating refined obstacle zone avoidance. First, to mitigate the subjectivity of threshold parameter settings during regional division of various wind farms, based on the spatial layout properties of offshore wind turbines, a partitioning strategy based on the improved k-ring-Voronoi diagram is designed, utilizing kernel density estimation for adaptive quantification and adjustment of neighborhood division to reduce candidate line scale and minimize redundant cross-constraints. Second, to effectively avoid irregular obstacle zones, auxiliary path points are introduced to optimize avoidance paths of obstacle zones by selecting auxiliary nodes through convex hull edge delineation, effectively preserving geometric features of obstacle zones. Finally, a mixed-integer linear programming (MILP) model is further constructed, integrating a preprocessing framework to achieve coordinated global optimization and local obstacle avoidance. Additionally, an accelerated strategy for MILP incorporating topological prior knowledge is designed to enhance convergence performance for large-scale optimization problems. Validation through case studies of wind farms with varying scales demonstrates that the proposed method can significantly reduce the computational complexity while maintaining topology optimization accuracy, thereby improving planning efficiency.关键词：offshore wind farm;wind turbine;power collection system;topology optimization;dimensionality reduction;k-ring-Voronoi diagram8|0|0更新时间：2026-09-22
- 摘要：While reducing the risk of commutation failure and enhancing the operation reliability of equipment in the line commutated converter (LCC), the self-adaptive static var generator (SVG) and line commutation converter (SLCC) also saves land area in converter stations, which has broad engineering application prospects. This paper analyzes the fault response and power characteristics of SLCC, reveals the profound impact of SVG integration on LCC. Regarding the issue of passive and lagging responses in SVG caused by the independent control of LCC and SVG in SLCC, an optimal control strategy for SLCC is proposed, which considers the interaction and coupling effects between LCC and SVG. This strategy enables coordinated control between LCC and SVG, fully utilizing the fast adjustment ability of SVG. Based on the electromechanical-electromagnetic hybrid simulation software PSD-PSModel, the simulation results verify the rationality of the theoretical analysis, and the power characteristics of SLCC are improved with the proposed optimal strategy.关键词：line commutation converter (LCC);commutation failure;static var generator (SVG);fault response;power characteristic;reactive power controt optimization10|0|0更新时间：2026-09-22
Application Research
- 摘要：Different from traditional battery energy storage systems, the dynamic reconfigurable battery energy storage system (DRBESS) consists of numerous battery modules with varying states, which are connected via power electronic switches. By controlling the access mode of each module via switches, flexible dynamic reconfiguration of the internal topology is achieved, thereby adjusting the operating condition of each module and ensuring safe and efficient system operation. Under different reconfiguration topologies, the battery system exhibits different exchangeable electrical quantities (including voltage, current, and power), leading to significant variations in its external support capability. To fully leverage the flexible operation advantages of the DRBESS while maximizing support to the energy-consuming entities, this paper proposes a bidirectional interactive coordinated operation strategy between the dynamic reconfigurable battery network (DRBN) and the power conversion system (PCS). The operational requirements of the reconfigurable battery network and the PCS are classified into six typical operating conditions. Through information interconnection and flexible state coordination between the two sides, optimal performance matching under each condition is achieved. A detailed simulation model is built based on MATLAB/Simulink, and effectiveness of the proposed control strategy under all operating conditions is verified.关键词：battery energy storage system;dynamic reconfigurable battery network (DRBN);power conversion system (PCS);power electronic switch;coordinated operation;power matching6|0|0更新时间：2026-09-22
- 摘要：The site-division cascaded ultra-high-voltage direct current (UHVDC) transmission system exhibits unique operating characteristics and protection requirements. This paper introduces the topology of the proposed system and proposes the principle of configuring independent bipolar-zone protections for each of the two converter stations, along with corresponding protection criteria. Under the operating condition, in which both poles are in service and each pole is grounded via the electrode lines of different stations, the failure of protections—such as the bipolar neutral-bus differential protection and the station grounding overcurrent protection—to isolate faults after operation is investigated. A scheme is proposed that, depending on the hot-standby status of the electrode lines, selects either electrode‑line switchover or blocking one pole while restarting the other with phase shift (with bipolar blocking applied for the station grounding overcurrent protection). Moreover, for a single open-circuit fault, the issue that the sound electrode line may still be damaged even after pole balancing is activated by the overload protection is studied, and a power runback scheme for the faulty pole is presented. Finally, real-time digital simulation verifies the feasibility and effectiveness of the proposed schemes.关键词：ultra-high-voltage direct current (UHVDC);converter station;site-division;electrode line;protection;fault7|0|0更新时间：2026-09-22
Engineering Application
- Postal code：310000
- Tel：13632102679
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0