English Version Journal Login
EN
最新刊期
卷 50 ， 期 12 ， 2026
- 摘要：With the intensification of global climate change, the green and low-carbon transition of energy is urgently needed. The green electricity trading market, as a key mechanism for promoting the development of renewable energy, is one of the important ways to achieve the goals of carbon emission peak and carbon neutrality. However, the current green electricity trading market is facing problems such as declining transaction prices and low market participation. At the same time, the lack of effective coordination between the green electricity trading market and other related mechanisms, such as the electricity spot market, the carbon market, and the renewable electricity consumption quota, has inhibited the vitality of green electricity trading. Therefore, this article summarizes the development history and current status of the green electricity trading market mechanism in China, analyzes the impact of the green electricity trading mechanism on multi-agent trading decisions, optimal scheduling, and carbon accounting, and summarizes the synergy mechanism between the green electricity trading market mechanism and key mechanisms such as the power spot market, green certificate trading market, carbon market, and the renewable energy electricity consumption quota mechanisms. Furthermore, it analyzes the bottlenecks and challenges in the development of the green electricity trading market and proposes a design scheme to improve the green electricity trading market mechanism, providing a decision-making reference for the long-term development of the green electricity trading market.关键词：green electricity trading;green certificate trading;power spot market;carbon market;renewable energy;electricity consumption;collaborative mechanism0|0|0更新时间：2026-08-21
- 摘要：In the context of new power systems with an increasing proportion of renewable energy, the renewable energy power forecasting has become one of the key technologies supporting high-proportion accommodation of renewable energy, which is widely used in such aspects as security analysis, optimal dispatching, stability control, and electricity market trading. Firstly, the renewable energy power forecasting technologies and evaluation indicators are outlined, and the characteristics and applicability of the existing evaluation indicators are analyzed. Secondly, the consistency problem between power forecasting and decision-making in the existing system is discussed, and the closed-loop forecasting-decision making-evaluation framework is constructed. Then, the application scenarios of renewable energy power forecasting are summarized; the quantitative evaluation system of forecasting values is discussed; and the value-oriented renewable energy power forecasting is constructed. Finally, future research directions are explored concerning forecasting value quantification, comprehensive evaluation indicators, value-oriented forecasting technologies, and forecasting-decision making integration, in order to provide practical ideas for the development of renewable energy power forecasting technologies in the context of new power systems.关键词：renewable energy;power forecasting;comprehensive evaluation indicator;value-oriented;decision-making;consistency;integration0|0|0更新时间：2026-08-21
Review·Perspective
- 摘要：Carbon emission reduction and carbon sink enhancement are key measures supporting the achievement of carbon neutrality goals. These two types of measures are coupled at the physical level but possess different economic characteristics. The carbon emission market and the carbon sink market are two key carbon market tools that incentivize carbon emission reduction and carbon sink enhancement activities. They operate according to their respective market rules while interacting with each other on multiple scales. To enhance the supporting role of both markets in achieving carbon neutrality goals, building upon the carbon market equivalent coordination mechanism, this paper proposes a regulatory decision optimization method based on closed-loop adaptive logic. This method targets the complex, time-varying, and non-linear problem of coordinated regulation for heterogeneous carbon markets, which involves multiple decision variables and tracking targets. It mitigates this complexity through the unified monetary valuation of multiple objectives, hierarchical decoupling of multiple decision variables, and adaptive adjustment of decision parameters. Taking a candidate path for China’s carbon neutrality goal as an example, the effectiveness of this method in enhancing target path tracking is validated through multi-scenario hybrid simulation. The results show that the coordinated use of regulatory and market instruments can effectively support the achievement of national long-term climate goals, providing decision support for building stable and efficient market mechanisms and regulatory strategies.关键词：cyber-physical-social system;carbon market;carbon emission reduction;carbon sink enhancement;equivalent coordination;decoupling scenario0|0|0更新时间：2026-08-21
- 摘要：Multi-energy coupling and efficient conversion can improve the utilization efficiency of various energy sources. This paper proposes an integrated energy system consisting of a reversible solid oxide cell (RSOC), a hybrid energy storage system, and combined cooling, heating and power (CCHP). Firstly, electric, heat and cold energy storage are introduced into the system as short-term energy storage units, and RSOC and CCHP are combined with hydrogen/methane storage tanks to realize long-term energy storage. Secondly, the output characteristics of renewable energy and load characteristics in different seasons are analyzed, and the operation strategy of the energy storage system is optimized. Then, technical and economic analysis is carried out for five typical scenarios, and the operation influences of the hybrid energy storage system, CCHP, and RSOC are discussed. Finally, the impact of carbon trading mechanism parameters on the planned capacity of energy storage and system cost is investigated. The simulation results show that the proposed model achieves high installed capacity of renewable energy and a lower energy abandonment rate, with both low-carbon performance and economic efficiency.关键词：energy storage;reversible solid oxide cell (RSOC);combined cooling, heating and power (CCHP);optimal allocation;carbon trading0|0|0更新时间：2026-08-21
- 摘要：In response to the issues of high carbon emissions and prominent carbon costs faced by combined heat and power (CHP) enterprises in Northern China due to the “heat for electricity” constraint on CHP units, an optimal operation strategy for CHP enterprises with carbon capture equipment considering dynamic carbon emission allowance (CEA) is proposed in the context of the electricity-carbon market. Firstly, given the unknown operating status of units for day-ahead optimization, Ebsilon software is used to obtain the heating supply ratio interval for two typical units, namely extraction condensing and high back-pressure units, through variable operating condition calculations, so as to establish dynamic CEA model considering changes in the load correction factors and heating supply correction factors of units, thereby solving the problem of estimation deviation inherent in traditional fixed quota factors. Secondly, from the perspective of enterprises putting carbon capture equipment into operation and by comprehensively considering the relationship between the electricity/heat demand of carbon capture equipment and the electricity-carbon market price, a model for the effective operating conditions of carbon capture equipment is established and a cooperative and optimal operation strategy for CHP enterprises with carbon capture equipment is further developed. Finally, based on the validation of actual CHP enterprises, it is demonstrated that the proposed model can accurately calculate the CEA and predict the operating conditions of carbon capture equipment, and related optimal strategy has significant low-carbon economic benefits in terms of increasing the profit of CHP enterprises and reducing carbon emissions.关键词：carbon emission;allowance;correction factor;carbon capture;carbon trading;low-carbon economy;optimal operation0|0|0更新时间：2026-08-21
Power Energy Transition for Carbon Emission Peak and Carbon Neutrality
- 摘要：To overcome the continuous impact of industrial risk propagation on the resource pool capacity and enhance the long-term stability of commercial operation of load-type virtual power plants, this paper proposes a structural optimization framework for resource pools considering capacity stability constraints from the perspective of system resilience. A production-electricity coupled network is firstly constructed, and a susceptible-infectious-recovered-susceptible (SIRS) model is introduced to simulate the risk propagation process within the resource network, thereby characterizing the impacts of economic disturbances on the resource pool capacity. On this basis, capacity stability constraints are introduced to construct structural optimization model for resource pools. Case study results show that the capacity fluctuation amplitude of the optimized resource pools under typical impacts decreases by about 30%, and the income increases by about 2.8%, while maintaining stability and adaptability under varying infection coefficient conditions. Further analysis indicates that the introduced constraints can enhance the capacity stability through improving the distribution of resource industries. The research results provide theoretical support for the transition of load-type virtual power plants from being merely “aggregatable” to achieving “sustainable” operation.关键词：virtual power plant;resource pool;capacity;stability;risk;propagation model;production network;stochastic optimization0|0|0更新时间：2026-08-21
- 摘要：With the continuous increase in installed capacity of renewable energy sources, there is an urgent need for new power systems to tap into adjustable resources on the user side for regulating grid parameters such as electric energy, voltage, and frequency. By employing commercial or technical means to aggregate adjustable resources on the user side, virtual power plants (VPPs) can effectively participate in grid regulation and reduce the costs associated with constructing reserve capacity in power systems. To address the high computational and scheduling pressure imposed on the platforms of VPP operators by traditional centralized direct-control methods, a consistency regulation strategy of distributed adjustable resources for VPPs based on a dynamic event-triggered mechanism is proposed. This strategy achieves distributed consistency regulation of user-side distributed adjustable resources aggregated by the VPP while avoiding communication redundancy caused by real-time communication of consistency algorithms. Additionally, the design of the dynamic event-triggered mechanism prevents a significant decline in the performance of conventional event-triggered algorithms as the number of aggregated resources in the VPP increases. Simulation results demonstrate that the proposed algorithm can extend the communication time interval between resources from 0.37 ms to over 16 ms in scenarios involving 8 and 16 distributed adjustable resources within the VPP, thereby rendering the algorithm feasible for wireless communication applications.关键词：new power system;virtual power plant (VPP);adjustable resource;consensus algorithm;event-triggered mechanism;Lyapunov function0|1|0更新时间：2026-08-21
- 摘要：With the in-depth construction of the new power system, user demand for supply reliability is continuously increasing. The single-core asymmetric phase-shifting transformer (SAPST) provides a key technology for the uninterrupted power transfer of loads. However, the planning with phase-shifting transformers often ignores the threat posed by the inrush current of the loop-closing operation to system reliability. To address this issue, a collaborative planning method for SAPST in active distribution networks considering economy and reliability is proposed. Firstly, an equivalent model of SAPST is established, and a partitioned fault restoration strategy is introduced to improve power supply continuity. Secondly, loop-closing current constraints are incorporated into the reliability evaluation framework, and improved indices considering loop-closing current penalty are put forward, enabling the evaluation results to better reflect actual operation risks. On this basis, a bi-level optimization model is constructed, in which the upper layer aims to minimize the investment cost and the lower layer targets the minimization of operation cost and reliability cost. The second-order cone programming algorithm is adopted to solve the model. Case analysis based on the modified IEEE 33-bus system shows that the proposed method can significantly suppress loop-closing current and enhance power supply reliability while ensuring economy, which verifies the feasibility and effectiveness of the method in realizing the collaborative optimization of active distribution networks.关键词：active distribution network;phase-shifting transformer;reliability;economy;fault restoration;loop-closing current;planning0|0|0更新时间：2026-08-21
- 摘要：Grid-connected wind power hydrogen production is an important means to realize wind power smoothing and green power hydrogen production. The volatile wind power cannot meet the requirements of grid-connected stability, while the traditional electrolytic hydrogen production system suffers from unstable operation and low efficiency under fluctuating power input conditions. In this regard, a optimal capacity allocation method for a grid-connected hybrid electrolytic hydrogen production system is proposed to take into account the operation characteristics of the electrolyzers. Firstly, a power smoothing and power allocation mechanism based on wavelet packet and empirical modal double-layer decomposition is constructed by incorporating an optimal compensation strategy for negative power fluctuations. An operation strategy for optimizing the number of starting and stopping times of electrolysis tanks and the efficiency of hydrogen production is formulated by comprehensively taking into account the differences in the operation characteristics of alkaline and proton exchange membrane electrolysis tank devices. Based on this strategy, a capacity allocation model for a grid-connected hybrid electrolytic hydrogen production system is established with the goal of maximizing system revenue, and the results are analyzed by cases and compared with other schemes. The results show that the hybrid electrolyzer effectively improves the economy and flexibility of the electrolytic hydrogen production system, and the proposed capacity allocation scheme can significantly improve the safety and stability of the electrolytic hydrogen production system and realize the efficient and economic operation of the hybrid electrolytic hydrogen production system关键词：wind power hydrogen production;power smoothing;power allocation;hybrid electrolyzer;capacity allocation;stability;flexibility;economy0|0|0更新时间：2026-08-21
- 摘要：Oscillation problems in power systems with a high proportion of renewable energy integration is prominent, and accurately obtaining the control parameters of renewable energy units is the key to assessing oscillation stability of the system. This paper proposes a novel approach for identifying key control parameters of wind turbines using limited measured oscillation data, its core lies in: utilizing measured oscillation voltage/current phasors to establish an equivalent RLC circuit model of the system at the oscillation frequency, and calculating the real/imaginary parts of impedance for the wind turbine at the oscillation frequency. Simultaneously, the theoretical impedance model of the wind turbine to be identified is derived, the calculated RLC parameters of equivalent circuit and the real/imaginary parts of theoretical impedance for the unit are combined to establish and solve a system of equations containing the key control parameters to be identified. The advantages of the proposed identification method over existing methods are that it avoids the risks associated with external disturbance injection and fully exploits the potential of existing oscillation data for parameter identification. Moreover, its identification accuracy meets engineering requirements. Case study based on controller hardware-in-the-loop (CHIL) experiments and field-recorded waveform data demonstrate that the proposed method remains effective and accurate under the influence of noise or initial data deviations.关键词：wind turbine;parameter identification;sub-synchronous oscillation;super-synchronous oscillation;equivalent RLC circuit;stability;impedance0|2|0更新时间：2026-08-21
- 摘要：Driven by the goals of carbon emission peak and carbon neutrality, the large-scale integration of power electronic-based renewable energy has significantly enhanced the dynamic response speed of the new power system, while concurrently increasing the risk of wideband oscillations. To effectively suppress the oscillations, there is an urgent need for fast, efficient, and precise detection of wideband oscillations. However, the existing detection algorithms struggle to simultaneously achieve rapidity, efficiency, and accuracy. Therefore, this paper proposes a fast detection method for wideband oscillations based on the hybrid sliding-window Fourier and frequency-locked algorithm, which incorporates the rapidity of the Fourier algorithm and the accuracy of the frequency-locked algorithm. Firstly, based on the instantaneous power theory, the voltage and current signals are converted into power signals, and the dual-frequency oscillation caused by frequency coupling is converted into a single-frequency oscillation. Secondly, the power signal is divided into low and high frequency bands to avoid mutual interference of signals during detection. For each frequency band, a single low-resolution sliding-window Fourier algorithm and a frequency-division sliding-window Fourier algorithm are employed to quickly obtain the approximate frequency and amplitude of the oscillation signal while significantly reducing the computational burden. Then, the approximate frequency is used as the center frequency of the phase-locked loop, and the oscillation amplitude is used as the trigger threshold of the phase-locked loop, so as to quickly lock the real frequency of the oscillation, and the detection time is consistently within 10 ms. Finally, the effectiveness of the method is verified by MATLAB/Simulink electromagnetic transient simulation and hardware-in-the-loop experiments.关键词：new power system;wideband oscillation;oscillation detection;frequency coupling;Fourier transform;phase-locked loop0|3|0更新时间：2026-08-21
- 摘要：Receiving-end power grids in China gradually present the trend of intensive multi-infeed high voltage direct current (HVDC) and power electronicization, resulting in the cascading failure risks significantly higher than the conventional power systems. Previous studies on cascading failures mostly emphasize the cascading line-overloading caused by power flow transfer, neglecting other cascading modes and the effects of system transients on cascading path, which cannot accurately reveal the evolving principle of cascading failure in multi-infeed HVDC systems. To adapt to the new characteristics of cascading events in the receiving-end system, this paper firstly constructs a multi-time scale and multi-level framework for the cascading evolution analysis, which is based on the typical events in the multi-infeed HVDC system. Secondly, by incorporating qualitative and quantitative methods, the coupling mechanism is studied on the event correlation between multiple DCs, between DC and nearby generators, and between generators and transmission lines. Meanwhile, the cascading failure mode of multi-infeed HVDC system is studied considering DCs and generators. Then, in view of the sensitivity of DC characteristics on control effect, the influence of voltage dependent current limit control on the evolution of cascading failures is studied. Finally, simulation is conducted on the customized IEEE 118-bus system and a regional power grid to validate the principles of cascading failures in multi-infeed HVDC system.关键词：multi-infeed HVDC system;cascading failure;commutation failure;generator transient;event correlation0|0|0更新时间：2026-08-21
- 摘要：Low-voltage flexibly interconnected technology enables power mutual support and flexible power flow control among distribution transformer stations, serving as an important means to improve the accommodation capability of distributed energy resources and the bearing capacity of loads. Transformerless flexibly interconnected technology has been widely applied in the field of low-voltage power distribution due to its high power density, high conversion efficiency, and low cost. However, the absence of an isolation transformer significantly raises the common-mode interference level of the system, resulting in large amplitude and high loss of common-mode current, which adversely affects the safe and stable operation of the system. Low-voltage power distribution systems adopt multiple grounding modes, whose differences directly influence the distribution characteristics of common-mode interference. Existing analysis models for common-mode interference usually ignore the influence of grounding modes, leading to inaccurate models and ambiguous characteristic analysis. In addition, topologies vary greatly among various grounding modes, making individual analysis complicated and cumbersome. Therefore, this paper proposes a construction method for a unified analysis model of flexibly interconnected systems based on the characteristic impedance equivalent method. It realizes the unified description of common-mode analysis models for various grounding systems and enhances the universality of the model. On this basis, the sub-band analysis method is adopted to investigate the distribution characteristics of common-mode components, so as to quantitatively evaluate the amplitude-frequency and resonance characteristics of system common-mode components in various grounding modes. Finally, the optimization selection principle of grounding modes for common-mode suppression is put forward, and the effectiveness of the proposed method is verified by simulation analysis.关键词：flexibly interconnected system;common-mode interference;grounding mode;distribution characteristic;stability0|0|0更新时间：2026-08-21
- 摘要：Due to the limited control degrees of freedom, the low-voltage ride-through (LVRT) control strategy for two-stage grid-connected photovoltaic system has difficulties in simultaneously meeting the multi-objective control requirements such as positive sequence voltage support at the point of common coupling (PCC), negative sequence voltage suppression, inverter overcurrent limitation, and rapid stabilization of the DC bus voltage. Moreover, according to the latest grid-connected photovoltaic technical regulations, most traditional LVRT control strategies lack the dynamic negative sequence reactive current support capability, and the voltage unbalance degree at the PCC is relatively high. Therefore, this paper proposes a multi-objective LVRT control strategy for two-stage grid-connected photovoltaic system. This strategy takes into account the dynamic positive and negative sequence reactive current support capabilities required by grid-connected photovoltaic technical regulations. During asymmetric faults, the LVRT mode is selected based on the state of charge of the energy storage. Current feedforward compensation control is introduced on the super-capacitor converter to rapidly stabilize the DC bus voltage. On the grid-connected inverter, the line impedance is taken into account and the grid-connected positive sequence voltage support, negative sequence voltage suppression, and inverter overcurrent limitation are integrated to control LVRT. Finally, on the MATLAB simulation and a semi-physical simulation experiment platform, the proposed control strategy is comparatively analyzed against existing LVRT control strategies, and the effectiveness of the strategy is verified.关键词：grid-connected photovoltaic system;low-voltage ride-through (LVRT);reactive current;super-capacitor;point of common coupling (PCC) line impedance;feedforward compensation1|0|0更新时间：2026-08-21
- 摘要：Due to limitations such as the sparse distribution of meteorological observation networks, low data resolution, and unstable communications, existing photovoltaic power forecasting methods suffer from insufficient information and delayed perception under rapid meteorological disturbances, making it difficult to support high-precision forecasting. To address this issue, this paper proposes a short-term photovoltaic power forecasting method based on information gain from satellite-terrestrial integrated micro observation stations. First, mutual information is used to evaluate the marginal contribution of potential deployment locations, optimizing the spatial configuration of micro observation stations. Then, a multi-timescale collaborative forecasting framework integrating hour-level trends and minute-level disturbances is constructed: for the trend component, multi-source numerical weather forecasting data is integrated, combining an XGBoost model with physical models to generate hourly forecasting results; for the disturbance component, a feature-aware enhanced temporal convolutional network based on a multi-head attention mechanism is proposed to model non-stationary local disturbances and achieve high-frequency compensation. Case studies demonstrate that the proposed method outperforms mainstream models in typical seasons, offering advantages in accuracy, responsiveness, and stability, thereby effectively improving short-term photovoltaic forecasting performance.关键词：photovoltaic;power forecasting;micro observation station;satellite-terrestrial integrated network;multi-timescale forecasting;spatio-temporal information gain;numerical weather forecasting;multi-head attention mechanism0|0|0更新时间：2026-08-21
- 摘要：To address the issues of insufficient historical data for spatial power load forecasting in urban distribution networks and the lack of explainability of existing artificial intelligence models, this paper proposes a spatial power load forecasting method based on explainable artificial intelligence. First, a power geographic information system is established to generate TypeⅠ cells and Type Ⅱ cells, and a time series generative adversarial network (TimeGAN) is employed to construct a data augmentation model for few-shot scenarios. Second, a spatio-temporal information graph of Type Ⅱ cells is built, and the spatial attention mechanism of a graph attention network (GAT) is used to extract the spatial features of each Type Ⅱ cell load. A heatmap of attention weights is plotted to visualize the degree of attention each TypeⅡcell receives during information aggregation, thereby providing explainability of the forecasting model in the spatial dimension. Then, the temporal self-attention mechanism of iTransformer is utilized to capture the temporal features in the load sequences of the cells. The temporal attention weights are extracted to identify the time steps that play key roles in the forecasting task, offering explainable support for the model’s decision-making basis in the temporal modeling process. Finally, the spatial features of Type Ⅱ cell loads output by the GAT and the temporal features output by the iTransformer are fed into the fully connected layer of the iTransformer. Meanwhile, a parallel one-dimensional convolutional neural network (1D-CNN) module is introduced, and skip connections are used to enhance the model’s ability to represent and utilize spatio-temporal features at the output stage. Engineering case studies demonstrate that the proposed forecasting method achieves higher accuracy than traditional forecasting methods.This work is supported byFoundationNational Key R&D Program of China (No. 2017YFB0902205) and Industrial Innovation Foundation of Jilin Province (No. 2019C058-7)关键词：distribution network;spatial load forecasting;time series generative adversarial network (TimeGAN);graph attention network (GAT);convolutional neural network (CNN);artificial intelligence0|0|0更新时间：2026-08-21
Basic Research
- 摘要：In the large-scale photovoltaic (PV) power transmission system, the control response of PV inverters after the clearance of AC system faults may lead to DC-side voltage sags, representing a new type of power quality issue induced by control response. To address this issue, this paper first investigates the internal power characteristics of PV generation units and the impact of inverter control states on the DC-side voltage. Based on the mechanistic analysis, the evolution process of the DC-side voltage sag is analyzed. Aiming to prevent the operational state evolution during this process, a fault detection method based on DC-side voltage and a mitigation strategy for the DC-side voltage sag based on active power output limitation are proposed. After detecting a fault, the constant active current control is used to limit the active power output of the inverter, thereby achieving the effect of mitigating the DC-side voltage sag. Finally, the optimal parameters of the mitigation strategy are obtained based on PSCAD/EMTDC simulation, and the effectiveness of the proposed mitigation strategy is verified by comparing the fault simulation of the system before and after the application of the strategy.关键词：new power system;power quality;control response;photovoltaic inverter;voltage sag;fault detection;mitigation strategy0|0|0更新时间：2026-08-21
- 摘要：The AC collection lines of offshore wind power systems utilizing flexible DC transmission are affected by the fault ride-through controls of two-ended converters, which significantly weaken fault characteristics. Consequently, conventional power-frequency-based protection schemes are prone to failure to operate. To address this, a Tanimoto similarity-based pilot protection scheme is proposed for offshore wind power AC collection lines, integrating adaptive negative-sequence current control. Firstly, the asymmetric fault characteristics of offshore AC collection lines under the negative-sequence control of two-ended converters are analyzed, and the correlation between fault types and fault severity is established. Based on the differences in current sequence components between ground faults and phase-to-phase faults, a fast fault type identification method is proposed using the ratio of integrated zero- and negative-sequence currents. On this basis, the correlation between fault types and negative-sequence current control is further explored. The offshore flexible DC converter adaptively injects negative-sequence current to amplify the current waveform differences between both ends of the line. The Tanimoto similarity is introduced to quantify fault severity, enabling the identification of both internal and external faults in the offshore AC collection lines. Finally, a simulation model of the offshore wind farm is developed on MATLAB/Simulink platform for verification. The results demonstrate that the proposed scheme is unaffected by the distribution capacitance of submarine cables or the output characteristics of wind turbines. It can quickly and accurately identify faults, providing higher reliability and sensitivity compared with traditional protection.关键词：offshore wind power;collection line;modular multilevel converter (MMC);negative-sequence current;asymmetric fault;Tanimoto similarity;control and protection coordination0|0|0更新时间：2026-08-21
- 摘要：Redundant characteristics of wind turbine supervisory control and data acquisition (SCADA) data can easily obscure effective information, affecting the accuracy of fault diagnosis. Therefore, an fault diagnosis method for wind turbines combining improved sequential attention feature selection and spatio-temporal feature fusion is proposed. First, a random sparsification strategy is introduced into the sequential attention feature selection to enhance the model’s focus on the global information interaction among variables. Secondly, a matrix sample set incorporating local spatio-temporal features is constructed to improve the mining effectiveness of spatio-temporal features. Next, an adaptive spatio-temporal information fusion model combining a gate-convolution neural network and a weighted bidirectional sequential network is proposed. This model achieves multi-scale global perception of the spatio-temporal matrix through hierarchical receptive field design and adaptively regulates information flow between layers using a convolutional forget gate, thereby enhancing the rapid response capability to abrupt electrical fault features. Simultaneously, the weighted bidirectional sequential network fully models forward and backward temporal dependencies, effectively capturing gradual mechanical fault features. On this basis, the spatio-temporal features extracted by various networks are fused to further improve the model’s comprehensive representation capability for multiple fault modes. Finally, validation results based on actual wind turbine SCADA data show that the proposed method can accurately identify various typical wind turbine faults.关键词：wind turbine;supervisory control and data acquisition (SCADA);fault diagnosis;feature variable;sequential attention;spatio-temporal feature fusion0|0|0更新时间：2026-08-21
Application 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