English Version Journal Login
EN
最新刊期
卷 50 ， 期 2 ， 2026
- 摘要：With the advancement of the goals of carbon emission peak and carbon neutrality, the integration proportion of renewable energy continues to increase, making the need for flexible interaction between power enterprises and energy-intensive industrial users more urgent. However, the energy-intensive industrial sector encompasses diverse categories, with complex process technologies and energy supply-demand characteristics. Effectively exploiting their load regulation potential and achieving flexible interaction with the power grid still face many challenges. Focusing on the development trends of flexible interaction for energy-intensive industrial users under the goals of carbon emission peak and carbon neutrality, this paper classifies the implementation approaches for such interaction, reviews the current domestic and international status, and elaborates on the main technical challenges and key technologies. Through typical cases, the feasibility and necessity of energy-intensive industrial users participating in power grid interaction are discussed. Finally, the future development trends of flexible interaction for energy-intensive industrial users are prospected.关键词：carbon emission peak and carbon neutrality;energy-intensive industrial user;flexible interaction;energy supply-demand;load regulation;demand response3|3|0更新时间：2026-02-28
- 摘要：Industrial users possess significant potential to provide flexibility for the power system through demand-side regulation. Although industrial production processes involve complex energy consumption mechanisms, existing general models such as the state-task network (STN) can effectively model the energy constraints of certain types of industrial processes. However, the STN model involves a large number of discrete variables and requires extensive parameters, making it difficult to directly embed into economic dispatch for massive demand-side flexibility resources. This hinders the full utilization of industrial user flexibility. To address this, a modeling method for the feasible region based on the optimal adjustable load model (OALM) is proposed for industrial users whose production processes can be modeled using STN. First, the integer variables of the STN model are eliminated by leveraging the demand response timescale, thereby linearizing the model. Then, considering the electricity consumption characteristics of industrial users, the adjustable load model is employed to approximate the linearized STN model. Robust optimization method is used to solve its parameters to maximize the volume of the feasible region, and power-energy coupling constraints are added to resolve the unbounded issue in current feasible region parameter optimization. Case studies demonstrate that, compared to the high-dimensional and integer-variable-based original STN model, the low-dimensional and linear OALM maintains high accuracy and stability under different production plans, timescales, and electricity price scenarios. This provides a new approach to addressing the curse of dimensionality faced by large-scale integration of industrial users into load-side regulation.关键词：industrial user;optimal adjustable load model (OALM);state-task network (STN);Benders decomposition;feasible region;flexibility;demand response11|4|0更新时间：2026-02-28
- 摘要：The industrial loads of energy-intensive enterprises, represented by ceramic production, account for a large proportion of electricity consumption. Represented ceramic production, electricity consumption characteristics of their production processes are adjustable, which is an important resource to participate in demand side response. This paper proposes an optimization method for “load transfer” characteristic modeling and electricity consumption characteristics of ceramic production process oriented towards demand side response. Firstly, the characteristic model of electricity consumption is constructed according to the ceramic production process. The relationship between electricity consumption characteristics of the production process and storage capacity is portrayed. Secondly, from the perspective of electricity consumption characteristics, the equipment with load transfer characteristics and its storage capacity are portrayed. The load transfer characteristic model of the ceramic production process is established. Thirdly, the optimal dispatch method for ceramic enterprises to participate in demand side response is proposed by taking the comprehensive revenue as the optimization objective, the safety of the production process as the constraint, and the load transfer characteristics of the electricity consumption of equipment as the control variable. Finally, the production process of an architectural ceramic enterprise is used as a case to verify the effectiveness and feasibility of the proposed method.关键词：ceramic;production process;energy-intensive enterprise;optimal dispatch;load transfer;demand side response;electricity consumption;optimization3|2|0更新时间：2026-02-28
- 摘要：Under the goals of carbon emission peak and carbon neutrality, playing the synergistic role of the mechanism of green certificate trading and carbon trading, and tapping the joint response capability of different types of industrial loads will have a positive effect on the green transformation and low-carbon development of industrial parks. Therefore, this paper proposes a dynamic pricing and energy management strategy for green power industrial parks considering the synergy of the electricity-certificate-carbon mechanisms. Firstly, the structure and operation mechanism of green power industrial parks is analyzed and an energy management framework of “one-master-multi-slave” is established. Secondly, a two-layer game model is developed with the objective of optimizing the benefits of the park operator and differential industrial loads. The upper-layer model optimizes the dynamic pricing strategy of the operator by considering the linkage between the mechanism of the green certificate trading and the carbon emission trading. The lower-layer model makes full use of the differential characteristics among users to optimize the energy management strategy of the loads. Then, a high-efficiency hierarchical iterative solution algorithm based on the improved alternating direction multiplier method is proposed for the two-layer game model. Finally, a case study is conducted on a typical system to verify the effectiveness of the proposed strategy.关键词：industrial park;energy management;industrial load;green power;green certificate trading;carbon emission trading;shared energy storage5|0|0更新时间：2026-02-28
Power Supply and Demand Interaction for Re-electrification in Industrial Production Processes
- 摘要：Following a wildfire disaster, quickly grasping the disaster situation and reasonably dispatching emergency resources can help shorten the restoration time of the distribution system and achieve the rapid restoration of critical loads. To expedite the post-disaster restoration process of the distribution system, an integrated satellite-terrestrial network (ISTN) architecture based on low earth orbit (LEO) satellite internet is introduced to build a three-layer restoration framework, encompassing information perception, fire response, and load restoration. With the objective of minimizing load loss, perception time and satellite terminal deployment cost, a cyber-physical collaborative restoration model of the distribution system under wildfire disasters is proposed. Firstly, a multi-source coupled fire detection model is constructed to evaluate fire levels and propagation trends in the disaster area. Secondly, the aerial and ground firefighting teams are coordinated to establish a multi-constraint firefighting dispatching model for optimizing the fire extinguishing paths and time. Then, a load restoration model is proposed that considers node state knowability, power topology accessibility, and LEO satellite communication terminal deployment strategies. Finally, a linearized two-stage optimization method is used to improve the solving efficiency. Simulations are conducted on the IEEE 33-bus system and a 10-bus sensing network, and the results validate the effectiveness and superiority of the proposed method in terms of shortening restoration time, enhancing total restoration capacity, and optimizing resource allocation.关键词：wildfire disaster;emergency resource;load restoration;integrated satellite-terrestrial network (ISTN);resilience;cyber-physical system;distribution system4|2|0更新时间：2026-02-28
- 摘要：The mechanism of fault evolution in distribution networks under extreme ice and snow weather is not accurately characterized. Additionally, existing fault restoration methods in distribution network cannot effectively represent the control methods of distributed power sources. This paper proposes a fault restoration method for active distribution networks considering fault evolution under extreme ice and snow weather. Firstly, this paper clarifies the impact of distribution line faults and uncertainty in distribution network power flow under extreme ice and snow weather. The probability of distribution line faults and the probability of system state transitions are quantified. A cascading fault evolution model for distribution network based on probabilistic power flow is constructed. Secondly, based on the region partitioning concept of point-based localization, the hierarchical relationships among master control nodes, subordinate control nodes, and load nodes within islanded regions are comprehensively considered. By integrating the master-slave control logic of islanded operation, a fault restoration model of distribution network is proposed, incorporating the coordination of network reconfiguration and island partition. Finally, the effectiveness of the proposed strategy is verified using the PG&E 69-bus distribution system and a certain 185-bus actual distribution system.关键词：probabilistic power flow;extreme ice and snow weather;cascading fault;master-slave control;fault restoration;distribution network reconfiguration;island partition;active distribution network7|3|0更新时间：2026-02-28
- 摘要：In recent years, extreme weather disasters occur frequently, leading to an increased probability of power grid equipment failure and exacerbating the operation risks of the power system. Based on deep reinforcement learning algorithms, an expansion planning method for transmission networks is proposed, considering both the voltage support strength of power system with renewable energy and the economic efficiency of expansion under extreme scenarios. Firstly, considering the impact of extreme disasters on the power grid, typical extreme scenarios for the power grid have been constructed. Secondly, the network planning problem is transformed into a sequential decision-making process by using Markov chains. The short-circuit ratio margin index and the comprehensive cost index of line expansion are taken as the objective functions, resulting in a network expansion planning model. Furthermore, a solving method for expansion planning model based on the multi-gate mixture-of-expert model and twin-delayed deep deterministic policy gradient reinforcement learning algorithm is proposed. Finally, in a modified IEEE RTS-24 system with integrated wind and photovoltaic power, as well as a DC transmission system case from a certain region in Northwest China, extreme operation scenarios of the system are simulated, and network expansion planning schemes considering various extreme scenarios are solved to verify the effectiveness and robustness of the proposed method.关键词：renewable energy;transmission network;reinforcement learning;voltage support strength;extreme disaster;short-circuit ratio margin;network expansion planning4|1|0更新时间：2026-02-28
- 摘要：Photovoltaic (PV) projects in the Gobi Desert and other arid areas serve as important supports for new power systems. However, such project sites are often characterized by complex climatic conditions, with frequent sandstorms, strong winds, extremely high temperatures, and other meteorological phenomena. These meteorological factors exhibit strong coupling and time-varying characteristics, posing significant challenges to PV power prediction. To address this issue, this paper proposes an ultra-short-term PV power prediction method that considers the impact of complex meteorological factors in Gobi Desert and other arid areas. First, a dynamic graph modeling method is introduced to represent the temporal correlations among multiple meteorological factors. The correlations are expressed as edge features to model coupling information. Second, an edge convolution improved graph attention network (EC-IGAT) prediction model is proposed. The proposed model captures both local geometric information and global context to improve prediction accuracy. Finally, the superiority of the proposed method is validated using actual data from a PV power station in the Gobi Desert and other arid areas of Northwest China关键词：new power system;photovoltaic power prediction;Gobi Desert and other arid areas;meteorological factor;dynamic graph;graph attention network7|1|0更新时间：2026-02-28
- 摘要：Distributed photovoltaic (PV) systems lack on-site meteorological measurements, while the cost of acquiring large-scale and high-resolution numerical weather prediction (NWP) data is very high, which makes it difficult for existing power forecasting methods for centralized photovoltaics to be applicable. Satellite cloud images, offering wide-area, high-resolution cloud observations at a low cost, which is expected to become an important source of meteorological information to solve this problem. Therefore, this paper proposes a probabilistic power forecasting method for distributed PV clusters based on similar cloud condition matching. First, an optical-flow-enhanced 3-dimensional convolutional neural network (3D-CNN) is constructed to forecast future satellite cloud images. By incorporating a physical consistency loss, the model is guided to focus on the dynamic evolution of cloud structures, thereby improving the physical plausibility of the forecasting results. Secondly, a cloud condition matching method is introduced, by searching for the cloud conditions that are most similar to the forecasting period in history, the corresponding photovoltaic power is extracted as input features for subsequent power forecasting. Finally, the common information such as similar historical power and measured PV power at adjacent times will be combined as inputs for the spline quantile regression model to achieve accurate probabilistic forecasting of PV power. Case studies are conducted on a large-scale real dataset containing over 600 distributed PV sites, and the results show that the proposed method shows significant improvement compared to classical kernel density estimation methods in terms of the continuous ranked probability score index.关键词：distributed photovoltaic;satellite cloud image;numerical weather prediction (NWP);meteorological information;probabilistic power forecasting;convolutional neural network (CNN);cloud condition matching5|3|0更新时间：2026-02-28
- 摘要：To address the issue of insufficient accommodation capacity of distribution networks caused by high penetration of distributed photovoltaic (PV) integration, an enhancement method for PV accommodation capacity of distribution networks based on self-attention diffusion reinforcement learning is proposed. Aiming to enhance the PV accommodation capacity of the distribution network, a multi-objective optimization mathematical model is established from the perspective of collaborative optimization of distributed PV inverters on the source side, static var compensators on the grid side, flexible loads, and battery energy storage controllers. The proposed model takes into account both voltage deviation and system network loss indicators, with the goals of maximizing the distributed PV allowable capacity and minimizing the total operation cost. While satisfying the operation constraints, the method uses a diffusion mechanism to rapidly adapt to complex and uncertain environments, and leverages a self-attention mechanism to enable parallel reading of batch data, thereby improving training speed and solution efficiency. Finally, simulation analysis of a case study verifies the effectiveness and economic feasibility of the proposed method.关键词：distribution network;distributed photovoltaic;accommodation capacity;collaborative optimization;reinforcement learning;diffusion;self-attention4|4|0更新时间：2026-02-28
- 摘要：The integration of a large number of three-phase unbalanced devices has led to three-phase current imbalance in station areas. Utilizing converters in the distribution network station areas to compensate for unbalanced currents is an effective method to suppress negative-sequence and zero-sequence currents. Three-phase photovoltaic converters are widely used in distribution networks, and their remaining capacity can be employed to address power quality issues. This paper proposes a distributed control strategy for mitigating unbalanced currents in station areas using three-phase photovoltaic converters. First, the three-phase unbalanced power flowing through the station area transformer is measured. Next, based on the remaining capacity of the photovoltaic converters in the station area at the current time, a distributed controller is used to calculate the power command for each converter tasked with compensation. This power command is then issued and executed via low-bandwidth unidirectional communication. Subsequently, the applicability of the proposed control strategy in scenarios involving communication failures and dual-end power supply networks is discussed. Finally, hardware-in-the-loop simulation is conducted in RT-LAB. The results demonstrate that the proposed distributed unbalanced current compensation control can effectively reduce the unbalanced current in the station area under the constraints of photovoltaic converter current limits and is widely applicable to various scenarios in distribution networks.关键词：unbalanced current;power quality;distributed photovoltaic;converter;current limit;station area;distribution network7|2|0更新时间：2026-02-28
- 摘要：With the widespread popularization of unbalanced equipment such as distributed generator and electric vehicle, a series of problems such as increased network loss and three-phase imbalance degree appear in the operation of distribution network. In order to reduce the voltage imbalance degree of distribution network with mixed phases, an optimal day-ahead commutation model of distribution load for three-phase imbalance mitigation is proposed. Firstly, a mixed-integer semidefinite optimization model is established with the objective of minimizing the voltage imbalance degree, incorporating constraints such as system commutation time, three-phase power flow in distribution networks, and system security operation. Secondly, discrete variables are transformed into equation-based quadratic constraints, yielding an optimal day-ahead commutation model for three-phase unbalanced distribution networks. Then, a Gaussian randomized iterative method is employed to recover the rank-one constraint, enabling the solution of an optimal load commutation strategy. Finally, by using standard IEEE test cases, the proposed method is compared against the three-phase power flow linearization, validating the correctness and effectiveness of both the model and the solution approach.关键词：three-phase imbalance;distribution network;optimal power flow;commutation;semidefinite programming;Gaussian randomization3|1|0更新时间：2026-02-28
- 摘要：With the increasing number of AC and DC interconnection modes in the distribution network, the issue of harmonics in the network has become increasingly prominent, making the demand for real-time monitoring and dynamic tracking of harmonics ever more urgent. However, the traditional dynamic state estimation algorithms, represented by the extended Kalman filter (EKF), are usually based on the empirically set state transition model and the measurement noise assumption of Gaussian distribution. In practical AC/DC distribution networks, these algorithms are prone to system model mismatch, leading to a decline or even failure in state estimation accuracy. To address this, a KalmanNet-based dynamic state estimation method for harmonics in AC/DC distribution networks is proposed. This method integrates EKF with recurrent neural network (RNN). Based on the framework of traditional EKF model, it eliminates the need for calculation and storage of high-dimensional covariance matrices and replaces the explicit calculation of state estimates with RNN. Testing and analysis are conducted on an expanded AC/DC hybrid distribution network derived from an improved three-phase unbalanced 33-bus case study. The results demonstrate that, compared with traditional algorithms, the proposed method exhibits higher accuracy and computational efficiency in harmonic state estimation, along with stronger robustness in the scenarios involving bad data.关键词：AC/DC distribution network;harmonic;state estimation;KalmanNet;extended Kalman filter (EKF);recurrent neural network3|1|0更新时间：2026-02-28
- 摘要：While grid-forming doubly-fed induction generators (DFIGs) actively support power grid frequency regulation, their power-frequency dynamic interactions with the power grid may also impact low-frequency or even ultra-low frequency oscillations in power systems. Using a wind power and hydropower combined system as the study case, this paper investigates the influence mechanism of grid-forming DFIGs with primary frequency regulation and virtual inertia control on ultra-low frequency oscillation modes in the power system. First, the electromechanical-scale components of wind turbines are linearized, accounting for various wind speed operating regions, and a simplified power-frequency response model at the wind turbine terminal is derived. Based on this model, a unified frequency response model for the wind power and hydropower combined system is then established. By using the complex torque coefficient method, analytical expressions of the additional synchronization and damping coefficients of the wind turbine are derived. The influence rules of control parameters, including the virtual inertia coefficient, the pitch angle control, and the primary frequency regulation droop coefficient, on the frequency and damping characteristics of ultra-low frequency oscillation modes under various operating conditions are systematically analyzed. Results demonstrate that increasing the virtual inertia coefficient and primary frequency regulation droop coefficient significantly reduces damping of the ultra-low frequency oscillation mode of the system, potentially triggering oscillation instability. Finally, the time-domain simulation on a modified 3-machine 9-bus system validates the effectiveness of the theoretical analysis.关键词：grid-forming doubly-fed induction generator;ultra-low frequency oscillation;frequency regulation;virtual inertia;complex torque coefficient method;pitch angle control;frequency response4|1|0更新时间：2026-02-28
- 摘要：The increase in the penetration rate of renewable energy has exacerbated the differences in the spatial distribution of inertia in the power system, and easily leads to the stability issues such as excessive node frequency change rate and maximum frequency difference. Therefore, a method for calculating and compensating the equivalent inertia of nodes in power system with renewable energy, taking into account power oscillations, is proposed. Firstly, the effective inertia source in the power system with renewable energy is analyzed, and the energy interaction characteristics in the frequency spatio-temporal dynamic process are elucidated. Secondly, considering the low increment in power oscillation between units, a distributed node equivalent inertia analytical model is established. On this basis, a method for calculating node correlation degree has been proposed, which utilizes the analytic hierarchy process method to construct a node dissimilarity matrix and implements cluster partitioning based on the differential characteristics between nodes. Then, by comparing the sum of weight coefficients, the optimal compensation node for the weak area of equivalent inertia of the node is accurately located, and the lowest compensation inertia is determined to achieve the optimal compensation of regional inertia at the system planning level. Finally, simulation verification is conducted on the modified IEEE 10-machine 39-bus system, and the simulation results verify the correctness and effectiveness of the proposed method.关键词：equivalent inertia;renewable energy;inertia compensation;analytic hierarchy process;power oscillation;cluster partitioning4|2|0更新时间：2026-02-28
- 摘要：With the large-scale grid-integration of renewable energy, the installed capacity proportion of conventional units is continuously decreasing. The inertia level and frequency regulation capability of the power system are weakened, increasing the risk of dynamic frequency security. To address this issue, an adaptive coordinated planning method of energy storage for low-inertia power systems is proposed based on dynamic frequency security assessment. Firstly, the dynamic frequency security indices considering time accumulation effects are proposed to assess the entire primary frequency response process of the system. Secondly, based on the dynamic frequency security indices, a zone-based energy storage siting strategy considering capacity sensitivity is developed. Finally, an adaptive coordinated planning model of energy storage based on numerical-electromechanical transient joint simulation is established. The economic, frequency security and adaptability objectives are trade-off optimized through alternating iterations of the numerical-electromechanical transient model, and the non-dominated sorting genetic algorithm is used to obtain the optimal storage capacity configuration. Case studies on a modified IEEE 39-bus system show that the proposed method improves the dynamic frequency security indices in the siting stage, and in the sizing stage, simultaneously guarantees economy and frequency security while enhancing the adaptability of the planning scheme to various uncertain scenarios.关键词：inertia;energy storage;planning;dynamic frequency security;assessment;electromechanical transient simulation;siting and sizing4|2|0更新时间：2026-02-28
- 摘要：Fast and accurate assessment of transient voltage stability after power grid disturbance is of great significance for the safe and stable operation of the system. The time-domain simulation method is time-consuming and difficult to meet the requirements of real-time assessment. Artificial intelligence algorithms, due to the low efficiency of model training and transfer, are difficult to meet the requirements of practical applications. To this end, an assessment method for transient voltage stability of power grids based on combined active learning and self-training is proposed. Firstly, a contrastive loss metric is introduced into the loss function to enhance the performance of the transient voltage stability classification model. Then, an active learning strategy is adopted to screen a small number of samples in the target domain for long-timescale simulation and precisely labels the voltage stability tags, and the initial fine-tuning of the model parameters is carried out. Secondly, the optimal fine-tuned model is applied to pseudo-label the remaining unlabeled samples in the target domain, and high-confidence samples are selected for the secondary fine-tuning of the model parameters, to fully explore the value of the unlabeled samples in the target domain, thus improving the prediction accuracy and transfer efficiency of the model. Finally, the CEPRI-VC standard cases are applied to verify that the proposed method is superior to the conventional methods in terms of accuracy and timeliness.关键词：transient voltage stability;artificial intelligence;transfer learning;active learning;self-training5|3|0更新时间：2026-02-28
Basic Research
- 摘要：In the wind-photovoltaic-hydro-thermal-storage integrated energy base, the key to enhancing the controllability of wind and photovoltaic power generation and providing high-quality auxiliary services for the power system lies in the coordinated control of various energy sources. The existing control strategies have the problems of low efficiency of decomposition calculation, slow data processing, poor reliability of results, and weak capability in exceeding-limit warning for the regulation margin of pumped storage and energy storage. Therefore, this paper applies an efficient online time series decomposition method, called OneShotSTL, to the analysis and processing of time series data for active power regulation. This method only updates the local calculation related to the new data through an incremental calculation strategy, which significantly improves the processing efficiency and real-time performance. Firstly, in the mode of tracking power generation plan and auxiliary frequency regulation, the OneShotSTL method is used to extract the trend, periodic and residual components from the time series data of active power regulation, to realize the accurate pairing of thermal power, variable speed pumped storage, and energy storage with trend components, periodic components and residual components, respectively. Secondly, to ensure the regulation margin of variable speed pumped storage and energy storage, the storage capacity of the pumped storage reservoir and the state of charge (SOC) recovery strategy of energy storage based on stochastic indices are proposed. On this basis, a real-time active power regulation and control strategy in the energy base is formed. Finally, the simulation analysis is conducted on a certain wind-photovoltaic-hydro-thermal-storage integrated energy base in the southern region of China to verify the effectiveness of the proposed strategy.关键词：energy base;real-time control;coordinated control;active power regulation;frequency regulation;wind power;photovoltaic5|2|0更新时间：2026-02-28
- 摘要：Currently, there are numerous wind turbine manufacturers with different control logics, and most of wind turbines are packaged models with unknown control structures. The application of these models has significant limitations. However, the stability study of grid-connected wind power system requires electromagnetic transient models that can accurately reflect the response characteristics of wind turbines during various typical fault ride-through processes. To address this issue, this paper tests the fault ride-through characteristics of a large number of packaged models from various wind turbine manufacturers, analyzes the test data, and proposes whole-process universal response curves of low and high voltage ride-through for both symmetric and asymmetric faults in multiple stages. By designing reference expressions for the outer-loop power control at the rotor and grid sides for each stage, and modifying the reference values for active and reactive power along with the control parameters, the entire fault response and recovery process of various wind turbines for manufacturers can be conveniently simulated. By comparing the test data of two packaged models with mainstream manufacturers and analyzing the error calculation, the adaptability and universality of the proposed modeling method for all fault conditions are verified.关键词：doubly-fed wind turbine;electromagnetic transient model;power outer-loop control;fault ride-through;modeling3|2|0更新时间：2026-02-28
Engineering Application
- 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