English Version Journal Login
EN
最新刊期
卷 50 ， 期 10 ， 2026
- 摘要：In the context of accelerating the construction of new power systems, distributed smart grids are increasingly encountering scenarios involving islanded operation, leading to increased operation difficulty. There is an urgent need for grid-forming energy storage to support their safe and stable operation. Firstly, the connotations and attributes of distributed smart grids and distributed islanded grids are given. Furthermore, from the perspectives of supply and demand balance, stability, and reliability of distributed islanded grids, the main challenges faced by distributed islanded grids are analyzed and summarized. Secondly, the control methods for grid-forming energy storage are analyzed, specifically including droop control, virtual synchronous generator control, matching control, and virtual oscillator control. Based on this, a multi-level control architecture for grid-forming energy storage is proposed. Then, the applications of grid-forming energy storage in enhancing regulation capability, steady-state operation, and operation under disturbances of distributed islanded grids are discussed. Finally, combining the current research status, thoughts and suggestions are proposed from the aspects of efficient planning, optimal operation, and intelligent control.关键词：new power system;grid-forming energy storage;smart grid;distributed islanded grid;stability;distributed energy;grid-forming control1|1|0更新时间：2026-08-21
- 摘要：Driven by the new round of electricity market reform and the goals of “carbon emission peak and carbon neutrality”, the construction of electricity markets in China continues to deepen, with further improvements required in operational efficiency and risk prevention capabilities. As a financial mechanism embedded in the electricity spot market, virtual bidding can effectively converge day-ahead and real-time market prices through arbitrage behaviors, enhance market competitiveness, and provide market participants with flexible risk management instruments. This paper first reviews the market models and strategy optimization methods of virtual bidding, and then analyzes its impact mechanisms on market operation, including promoting price discovery, mitigating market power, and its interactions with other markets. On this basis, the market design and supervision mechanisms for virtual bidding are summarized. In consideration on the development stage and evolutionary trends of electricity markets in China, the adaptability, potential benefits, and major challenges of introducing virtual bidding are assessed. Finally, a feasible staged and well-regulated implementation pathway is proposed, aiming to provide a reference for the efficient operation and risk prevention of electricity markets in China.关键词：virtual bidding;electricity market;spot market;electricity financial product;trading strategy;risk management;market mechanism6|1|0更新时间：2026-08-21
- 摘要：The inherent control delay of the energy storage converter, together with the relevant technical code requirements for positive- and negative-sequence reactive power injection, may cause both transient and steady-state overcurrent risks, respectively. It is prone to result in failure of fault ride-through. To address this issue, this paper proposes a fault ride-through method for energy storage power stations based on multi-timescale hierarchical control. In this method, the fault process is divided into the instantaneous scale and the short-term scale. On the instantaneous scale, a current-hysteresis-based insulated-gate bipolar transistor (IGBT) blocking method is adopted to suppress the inrush current before the controller responds. On the short-term scale, a multi-objective control framework incorporating negative-sequence reactive power regulation is developed to achieve software-based current limiting and code compliance. Through multi-timescale coordinated control, the current is suppressed throughout the entire fault process, while simultaneously mitigating active power oscillations and maximizing the utilization of converter capacity. Finally, a real-time digital simulation platform is used to build a simulation model of an energy storage power station in Guangdong, China, and the simulation results validate the effectiveness of the proposed strategy.关键词：energy storage power station;converter;fault ride-through;current limiting;active power oscillation;reactive power regulation3|1|0更新时间：2026-08-21
- 摘要：Dynamic reconfiguration of distribution network enables more flexible operational regulation through topology adjustment. However, the existing decision-making technologies struggle to balance computational real-time performance and reliability. Large language models (LLMs), with their powerful semantic understanding, in-context learning, and generative decision-making capabilities, demonstrate significant potential for complex decision-making problems such as dynamic reconfiguration of distribution network. Nevertheless, their direct application to dynamic reconfiguration optimization faces challenges including the absence of structured domain knowledge, insufficient task comprehension and adaptation capabilities, and difficulties in embedding physical constraints, thereby compromising decision reliability. To address these issues, this paper proposes a dynamic reconfiguration method of distribution network based on knowledge graph enhancing reinforcement fine-tuning of LLM. First, a dynamic knowledge graph is constructed to explicitly represent the temporal evolution and causal logical relationship of distribution network operation. Second, high-quality fine-tuning datasets are generated through subgraph sampling of knowledge graph, and the supervision and fine-tuning of the pre-trained LLM is completed to enhance its task comprehension and instruction-following capabilities. Finally, a multi-dimensional reward function is designed based on the structured prior knowledge from the knowledge graph, and the group relative policy optimization is introduced for reinforcement fine-tuning of the model. Case analyses demonstrate that the proposed method significantly improves the generation efficiency and physical consistency of dynamic reconfiguration strategies of distribution network, effectively improving the operational level of distribution network.关键词：distribution network;dynamic reconfiguration;large language model (LLM);artificial intelligence;reinforcement;fine-tuning;operational control;knowledge graph3|6|0更新时间：2026-08-21
- 摘要：Renewable energy can rapidly regulate the power in the new power system, demonstrating the potential of participating in overload emergency control for lines. However, when it is adopted, generation methods for the emergency control strategy based on deep reinforcement learning face the challenges of excessively large decision space and high solution complexity. To address this issue, a hybrid learning method for emergency control with dynamic iteration of decision space and strategy model is proposed. First, a dual-network model comprising a control location network and a control value network is constructed, and an iterative learning framework for both networks is designed. Second, the control location network and its learning objectives are introduced, and a sensitivity-based sample generation method is designed to learn the control location network. Then, a deep reinforcement learning method for the control value network is proposed, and a segmented exploration strategy is designed for efficient learning of the control value network. Next, a dynamic iteration implementation process between the control value network and control location network is designed. Finally, the effectiveness of the proposed method is validated in the IEEE 39-bus system, IEEE 300-bus system, and a provincial power grid of China.关键词：emergency control;renewable energy;deep reinforcement learning;line overload;hybrid learning;dynamic iteration;decision space;sample generation3|1|0更新时间：2026-08-21
- 摘要：To address the complex and varied source-load scenarios in the new power system, the scenario clustering method improves optimization efficiency by selecting typical scenarios. However, traditional scenario generation methods fail to adequately capture source-load correlations, making it difficult to generate representative typical scenarios. This paper proposes a wind-photovoltaic (PV)-load typical scenario generation method based on three-dimensional (3D) convolution and spatio-temporal attention mechanism. First, historical wind-PV-load data are preprocessed using Z-score normalization and cubic spline interpolation. Then, a deeply embedded autoencoder is constructed by integrating 3D convolution and spatio-temporal attention mechanism. The 3D convolution introduces the temporal dimension to extract spatio-temporal features of source-load scenarios, while the spatio-temporal attention mechanism enhances the extraction of key spatio-temporal features. Furthermore, a three-stage training strategy is proposed to train the reconstruction ability and the clustering performance of the model in stages, avoiding performance degradation caused by single-task training. Finally, a case study based on real wind-PV-load datasets is carried out. The results show that, compared with both traditional clustering methods and deep learning based clustering approaches, the proposed method demonstrates significant advantages in terms of clustering metrics, and can effectively meet the requirements of economic operation optimization of power system in terms of both accuracy and timeliness.关键词：scenario generation;uncertainty;three-dimensional convolution;autoencoder;spatio-temporal attention mechanism;clustering4|2|0更新时间：2026-08-21
- 摘要：The modular multilevel converter (MMC) based high voltage direct current (HVDC) transmission is an important approach to the large-scale integration of renewable energy. In areas where renewable energy is concentrated, the insufficient transient stability margin has become one of the key factors limiting the transmission section limits. To address the transient instability of MMC caused by large-signal disturbances, a virtual flux linkage feedback control is introduced without altering the original control structure, thereby enhancing the transient stability capability and prolonging the critical clearing time of the grid-following and grid-forming MMCs. By deriving the impedance characteristics of grid-following and grid-forming MMCs and the power angle characteristics of the grid-forming MMC before and after adding virtual flux linkage feedback control, it is found that the virtual flux linkage feedback control can reduce the resonant peak and enhance the phase angle stability margin of both grid-following and grid-forming MMCs, and shift the power angle characteristics of grid-forming MMC upward, endowing it with a natural characteristic of “large deceleration area and small acceleration area”. The effectiveness of the proposed control strategy and the correctness of the theoretical derivation results are verified through the two-terminal MMC-HVDC system and the 3-machine 9-bus system based on PSCAD/EMTDC and the hardware-in-the-loop (HIL) experiments.关键词：flexible direct current (DC) transmission;modular multilevel converter (MMC);transient stability;small-signal stability;virtual synchronous generator;grid-following control;grid-forming control;virtual flux linkage feedback control1|1|0更新时间：2026-08-21
- 摘要：The grid-forming flexible direct current (DC) converter stations possess the capabilities of dynamic frequency and voltage regulation and active support, as well as favorable resonance stability under weak grid conditions. However, there is still a gap in the research on their small-signal stability under various non-fault and short-circuit fault conditions. This paper focuses on the analysis of linear periodic time-varying modeling and small-signal stability of grid-forming flexible DC transmission systems. Firstly, a nonlinear state space model of the grid-forming flexible DC transmission system is established, and a linear periodic model suitable for small-signal stability analysis is derived through linearization. Then, based on this model and using Floquet theorem, the small-signal stability of the system under various operating conditions with different short-circuit ratios is investigated. Aiming at the resonance risk of the system under low grid impedance, an additional damping strategy is proposed to improve the stability margin of the system. Finally, simulation verification is carried out on the PSCAD/EMTDC platform. The results show that short-circuit faults in the AC grid significantly affect the stability boundary of the differential-mode control parameters, and the impact of three-phase faults is more significant than that of single-phase faults. When the grid-forming flexible DC transmission system is connected to an AC grid with low impedance, there is an obvious resonance risk. Moreover, under fault conditions, the stability deteriorates as the fault severity increases.关键词：grid-forming control;flexible direct current (DC) transmission;Floquet theorem;fault condition;stability;small-signal stability;additional damping1|2|0更新时间：2026-08-21
- 摘要：Improper control in high-voltage direct current (HVDC) system induces subsequent commutation failure (SCF), leading to repeated severe power fluctuations. Under various operating conditions, the AC reactive power varies significantly. From the power perspective, the commutation security constraint law can be characterized jointly by DC transmitted active power and AC reactive power, providing a new insight for control optimization. The correlation between commutation security constraints characterized by extinction angle and DC active and AC reactive power is sorted out. It reveals that the increase in active power is conditional on the reduction of reactive power or the extinction angle, and the extinction angle security can be jointly characterized based on active and reactive power. Accordingly, a margin prediction and suppression method for SCF based on active and reactive power security constraints is proposed. The effect of control on power is quantified, and the active and reactive power during fault operation can be predicted. Considering the impact of the commutation process on power, the temporal characteristics of the valve-side current are used to detect the commutation angle, increasing the prediction accuracy. A critical extinction angle satisfying the predicted power level is derived, and the deviation between this angle and the extinction angle order is used to construct an SCF margin index and a compensation order, thereby suppressing SCF. Simulation verifies that the proposed method exhibits excellent performance under various operating conditions across multiple control systems.关键词：high-voltage direct current system (HVDC);power security constraint;commutation failure;power prediction;margin prediction;extinction angle compensation2|1|0更新时间：2026-08-21
- 摘要：The penetration of distributed energy resources in the power system is constantly increasing. Establishing local electricity markets will help promote local accommodation and ensure the safe, stable, and economical operation of power systems with a high proportion of distributed energy resources. This paper proposes a local electricity market mechanism based on a combined clock auction, comprising three phases: the clock round, the complementary round, and settlement. In the clock round stage, an iterative clearing mechanism of “pricing by the market operator and decentralized optimal quantity bidding by market participants” is designed. A local price adjustment model incorporating the idea of fusion momentum method is introduced to transmit market dynamic information to aggregators through local price iteration, thereby reducing clearing uncertainty and trading risks. In the complementary round stage, a sealed-bid model considering behavioral constraints on bidding is developed to lower the bidding difficulty for aggregators. It is proven that the pre-cleared results of the clock round represent the optimal allocation with sealed bids in the complementary round. In the settlement stage, a settlement mechanism based on VCG-nearest payment is adopted to ensure that settlements satisfy core payment constraints, enhance market stability, and incentivize truthful bidding by market participants to a certain extent. Case study results demonstrate that the proposed market mechanism can effectively guide the local balancing of distributed energy resources and improve the economic efficiency of system operation.关键词：electricity market;combinatorial clock auction;VCG-nearest payment;uncertainty;aggregator;bidding;distributed energy resource2|1|0更新时间：2026-08-21
- 摘要：The output curves of wind power and photovoltaic are characterized by strong fluctuation and randomness, and the medium- and long-time bilateral negotiation trading faces the significant risks of electricity volume deviations and contract performance. To this end, an optimal medium- and long-term bilateral negotiation trading strategy and the optimal decision-making method of the contracted volume and price for the integrated participation of multiple entities from both renewable energy and users are proposed. Firstly, a matching model that integrates the fast dynamic time warping alternating based on slope adjustment with K-center point clustering is introduced to identify the best trading user cluster for renewable energy generation entities, thus enabling the optimal trading-object matching for the integrated trading participation of multiple entities. Secondly, based on the principle of the dual-interactive Stackelberg game, a method for formulating the optimal total trading power and contracted electricity price curves of the medium- and long-time trading contracts among renewable energy and users is proposed. Finally, based on the predicted power ratio among power generation and consumption entities, a decomposition model of the total trading power curve for renewable energy and users is constructed to determine the self-contracted power decomposition curves of each power generation and consumption entity participating in the integrated trading. Simulation calculations are carried out using an actual 436-bus system to verify the effectiveness of the proposed optimal medium- and long-term trading strategy. In comparison with the traditional bilateral negotiation trading strategy, the proposed strategy shows the obvious advantages in terms of the small electricity volume deviation, the increased medium- and long-term revenues of renewable energy, and the standardization of the trading process.关键词：renewable energy;electricity market;contracted volume and price;bilateral negotiation;Stackelberg game;trading2|1|0更新时间：2026-08-21
- 摘要：Multiple uncertainties in renewable energy, loads, and market transactions in spot markets pose significant challenges to the operation of industrial parks. To address this, this paper proposes a coordinated optimization strategy for bidding and resource scheduling considering dual-risk control. First, a day-ahead and real-time optimization model for industrial parks is constructed. The day-ahead optimization model employs a bi-level optimization method for solution, in which the upper-level model optimizes the bidding and scheduling strategies of the industrial park, and the lower-level model simulates the market clearing process. The real-time optimization model uses a rolling optimization method for solution, and dynamically adjusts scheduling strategies. Second, to address the uncertainties in renewable energy and non-adjustable loads, a probability distribution ambiguity set is constructed based on the Wasserstein distance, and a distributionally robust optimization approach is employed to derive the strategy with the worst-case probability distribution. And for the uncertainties associated with bidding prices of competitors and real-time electricity prices, a limited set of scenarios for typical markets is constructed, and the minimax regret value criterion is adopted to minimize the maximum regret value across all scenarios. Simulation results demonstrate that the proposed bidding and scheduling strategy can significantly improve the revenues of industrial parks and reduce the bidding risks.关键词：industrial park;electricity market;spot market;bidding;resource scheduling;robust optimization;uncertainty5|1|0更新时间：2026-08-21
Invited Paper
- 摘要：To address the issues of high complexity and cost in large-scale heterogeneous distributed resource aggregation models, a quantitative evaluation method for distributed resource aggregation capability based on safe reinforcement learning is proposed. Each scheduling cycle of the virtual power plant (VPP) agent is divided into two stages. In the first stage, the output characteristics of current distributed resources are preliminarily evaluated based on the optimal power flow model. In the second stage, the agent performs corrections to ensure the voltage stability and power flow safety amid fluctuations in distributed resources, and reports to the upper-level grid the active power range required for power exchange, which serves as a quantitative indicator of flexibility for the current cycle. In a test case based on actual charging data from Shenzhen, China, the VPP agent trained by soft actor critic (SAC) algorithm fully ensures the network security, with the average fluctuation variance of the reported range below 0.062 and the violation rate as low as 6.25%. The average satisfaction rates for charging loads at two electric vehicle charging stations reach as high as 94.81% and 83.82%, respectively. The results demonstrate that the proposed data-driven method can automatically capture complex nonlinear relationship and high-dimensional features through its multi-layer neural network architecture in VPP scheduling scenarios, thereby offering higher reliability.关键词：flexible resource;distributed resources;virtual power plant (VPP);safe reinforcement learning;agent;aggregation model;quantitative evaluation1|3|0更新时间：2026-08-21
- 摘要：Virtual power plants can participate in the ancillary service market by aggregating the distributed resources of prosumers, but the existing aggregation modes suffer from insufficient privacy protection, dependence on accurate mathematical models, and a lack of dynamic adaptation mechanisms. To address the dynamic evolutionary requirements of prosumers participating in aggregation and device configuration update, a peer-to-peer (P2P) trading-driven aggregation method for virtual power plants based on multi-agent reinforcement learning is proposed. Firstly, an observable Markov game framework is constructed, with each prosumer and virtual power plant as an independent decision-making agent. Secondly, the attention mechanism is used to extract the characteristics of multi-energy flow coupling between devices, capture the complex interaction relationship, and construct an adaptive model to achieve dynamic adjustment. Then, the asynchronous multi-agent proximal policy optimization algorithm is used to solve the equilibrium strategy, and the P2P price signals are used to guide the prosumer’s aggregation to realize the multi-market arbitrage of the virtual power plant. Finally, case studies verify that the proposed method outperforms traditional algorithms in terms of both convergence speed and economic performance.关键词：virtual power plant;prosumer;aggregation;peer-to-peer (P2P) trading;multi-agent reinforcement learning;multi-energy flow coupling;Transformer;adaptive multi-layer perceptron (AdaptMLP)1|1|0更新时间：2026-08-21
- 摘要：As a crucial mechanism connecting distributed energy resources and end users, the local energy market can effectively promote the coordinated operation and economic development of integrated energy systems. To address this issue, an optimal operation model of integrated energy systems for electricity-hydrogen local energy markets is proposed. First, a system operation framework is constructed, including electricity-hydrogen energy service providers, integrated electricity-hydrogen energy consumers, electricity consumers, and hydrogen consumers, where the electricity-hydrogen energy service provider acts as the system operator participating in the market and exchanges energy with the power grid. Then, based on the proposed framework, a bi-level optimization model is developed, with the objective of minimizing the cost of service providers at the upper-level and minimizing the cost of each consumer at the lower-level. Finally, the lower-level problem is transformed into Karush-Kuhn-Tucker (KKT) conditions and incorporated into the upper-level, after which the duality theory and the big-M method are applied to transform the model into a mixed-integer linear programming (MILP) problem. Case study results show that the proposed model can effectively coordinate electricity-hydrogen energy flows, optimize system operation strategies, and promote deep integration between local energy markets and integrated energy systems, thereby significantly improving the operation economy and flexibility.关键词：electricity-hydrogen coupling;energy market;duality theory;big-M method;integrated energy system;optimal operation;flexibility4|1|0更新时间：2026-08-21
- 摘要：As a critical carrier for the low-carbon transition, the development of hydrogen energy requires not only technological breakthroughs but also robust support from market mechanisms. To address the challenges of the lack of accounting rules for hydrogen-enabled carbon reduction and inadequate market coordination mechanisms, this paper proposes a hydrogen credit market mechanism to enhance the carbon reduction value of hydrogen energy. First, a hydrogen credit certification system based on equivalent carbon emission reduction is established from the perspective of the full life cycle of hydrogen energy, enabling accurate accounting of hydrogen decarbonization equivalents. Second, a hydrogen credit market mechanism is designed based on offsetting, trading and settlement. The offsetting mechanism allows hydrogen credits to be converted into carbon allowances, expanding compliance pathways in the carbon market. The trading mechanism realizes cross-regional transactions through a peer-to-peer mode, facilitating the circulation of carbon reduction data and mutual recognition of value. The settlement mechanism entrusts regulators to settle surplus hydrogen credits and realize the monetization of environmental benefits via subsidies. Third, mobile hydrogen storage resources such as hydrogen-fueled ships are used as hubs for the conversion of hydrogen credit value. Combined with the energy-transportation coupling characteristics, a cooperative game model for a multi-port integrated energy system is constructed, and the alternating direction method of multipliers is adopted for distributed solution, which effectively protects the privacy of carbon reduction data. Finally, simulation results based on three ports verify that the proposed market mechanism can effectively achieve the environmental and economic synergy of hydrogen energy.关键词：hydrogen energy;hydrogen credit;market mechanism;carbon emission reduction;peer-to-peer (P2P) trading;port;integrated energy system;energy-transportation coupling6|1|0更新时间：2026-08-21
- 摘要：Existing graph attention networks (GATs) rely solely on the topological connections, making it difficult to capture the hierarchical partitioned structure characteristics of power grids. Moreover, traditional self-attention mechanisms have extremely high computational complexity and are difficult to directly apply to ultra-large-scale power systems. Therefore, a hierarchical partitioned attention (HPA) mechanism is proposed for the pre-trained models in the power flow analysis of power systems, aiming to enhance the generalization capability of pre-trained models to power grid topology variations. First, based on the indirect connection relationship between nodes at various voltage levels, a dynamic allocation mechanism for attention weights at cross-level nodes is designed to capture cross-level structural characteristics. Then, a joint pre-trained framework for the heterogeneous power grid is designed, in which a zero-padding strategy is proposed to enhance the node-scale adaptability of the pre-trained model and a parameter sharing mechanism is used to achieve distributed learning of heterogeneous power grid features. Finally, based on the actual distribution network data, a pre-trained model for power flow analysis capable of handling up to 50 000 nodes is constructed, and its computation performance in power flow analysis is validated. The results demonstrate that the proposed HPA mechanism can capture cross-level node characteristics and significantly enhance the generalization capability under topology variations.关键词：power flow analysis;hierarchical partitioning;Attention mechanism;pre-trained model;dynamic allocation;distributed learning0|1|0更新时间：2026-08-21
- 摘要：To address issues such as signal delay, impedance mismatch, and stability challenges at the interfaces of hybrid simulation systems, a power interface modeling method for the hybrid simulation system based on predictive loop active disturbance rejection control (ADRC) is proposed. Firstly, by analyzing the principles and stability conditions of the voltage-type ideal transformer method, it is proposed to applying the ADRC in the real-time estimation and compensation of interactive errors and unmodeled dynamics between the digital and physical sides, thereby improving the anti-interference capability of the hybrid simulation system. Secondly, a Smith predictor is designed to obtain a delay-free system output, which is then fed back into the extended state observer to enhance the control system performance. Then, a control parameter design method is proposed based on the bandwidth-parameter method from the controller perspective. This method designs a control parameter tuning process suitable for digital-analog hybrid simulation system by setting system bandwidth and tracking performance indicators. Finally, the correctness of the theoretical analysis and the effectiveness of the proposed method are verified based on digital simulation and power hardware-in-the-loop experiments.关键词：hybrid simulation system;power interface;active disturbance rejection control;Smith predictor;parameter design;power hardware-in-the-loop4|1|0更新时间：2026-08-21
-
Secure Packet Offloading and Scheduling for IoT Devices in Smart Substations Based on Edge Computing
摘要：To address the issues of balancing security and real-time performance of communication network in industrial Internet of Things (IoT) scenarios such as smart substations, this paper proposes a secure communication strategy based on edge computing. First, the communication security protection boundary and implementation mechanism for IoT devices in substation based on national cryptography algorithms are defined in detail. Then, the advantages of edge computing applying to the communication security of IoT devices are elaborated, and a collaborative processing model incorporating security protection, computing services, and message task models is established. The key technologies for secure packet offloading and scheduling are clarified. Finally, taking the communication services between typical IoT devices in a D2-1-type substation as an example, the node processing delay is precisely quantified through theoretical analysis and formula derivation. Subsequently, the network communication delay is obtained via OPNET simulation, and the end-to-end communication delays of the proposed strategy under multiple scenarios are compared with those of conventional strategies. The results show that the proposed strategy effectively improves the real-time response speed while meeting the security requirements of IoT device communication services in smart substations.关键词：industrial Internet of Things;smart substation;network security;communication;edge computing;task offloading2|1|0更新时间：2026-08-21
Basic Research
- 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