English Version Journal Login
EN
最新刊期
卷 50 ， 期 7 ， 2026
- 摘要：As a hub for offshore energy supply and conversion, the energy island can not only mitigate the fluctuation of renewable energy output but also achieve cross-temporal and spatial storage and clean utilization of energy. This paper introduces the basic concept of the energy island and its core features of multi-energy complementarity, spatial intensification and intelligent management. It also analyzes the policy support and research progress. In response to the complexity of the marine environment and the challenges of multi-energy system coupling, key technologies such as energy master planning, flexible electrical interconnection control, bidirectional efficient and flexible conversion of electricity-hydrogen-ammonia, energy management, and platform safety design need to be broken through in the future. This paper also discusses the demonstration project conception of “electricity-hydrogen-ammonia” synergy and engineering implementation risks. The research can provide innovative idea for the efficient utilization of renewable energy in the deep sea and is of great significance for achieving the goals of carbon emission peak and carbon neutrality.关键词：carbon emission peak and carbon neutrality;energy island;multi-energy system;renewable energy;electricity-hydrogen-ammonia;energy management;demonstration project2|19|0更新时间：2026-08-21
- 摘要：With the large-scale integration of renewable energy such as wind and solar, hydrogen production by water electrolysis has become an important approach to accommodate renewable energy. However, both alkaline and proton exchange membrane electrolysis systems operating under power fluctuation conditions face reliability challenges such as catalyst activity decay, membrane damage, thermal management failure, and gas crossover. This paper reviews relevant research progress. First, it summarizes typical failure mechanisms of various components in power-to-hydrogen systems under power fluctuations. Second, it categorizes fault diagnosis methods into four categories—rule-based, model-based, data-driven, and hybrid methods—and compares their differences in various aspects. Next, key supporting technologies for large-scale renewable-powered hydrogen production are discussed, including mechanistic modeling, state identification algorithms, and intelligent hardware optimization. Finally, future research directions toward enhanced reliability and fault diagnosis of power-to-hydrogen systems are prospected.关键词：power-to-hydrogen;renewable energy;power fluctuation;failure mechanism;fault diagnosis;reliability1|0|0更新时间：2026-08-21
- 摘要：To promote the green development of energy production and utilization, reduce societal carbon emission, and enhance the accommodation of clean energy, there is an urgent need to drive the low-carbon transformation of numerous existing park-level integrated energy systems (PIESs). This paper proposes a system integration method for the low-carbon transformation of electricity-hydrogen-gas-heat PIES, considering hydrogen energy utilization and carbon trading. First, an overall system integration framework is constructed and the system integration process is elaborated, which includes two stages: pre-transformation assessment and low-carbon transformation. Secondly, the models for hydrogen energy equipment including hydrogen production and seasonal hydrogen storage are developed, and a tiered carbon trading mechanism is introduced. Considering the volatility and temporal characteristics of renewable energy and multi-energy loads, an ordered clustering and scenario generation method based on Fisher optimal segmentation is employed for modeling. On this basis, a system evaluation model is established to analyze the weak links that constrain energy conservation and emission reduction in existing PIESs. Subsequently, a stochastic optimization model for low-carbon transformation is constructed, with the objective of minimizing the comprehensive annualized cost. Constraints for five types of transformation operations such as addition, replacement, and decommissioning, along with system operation constraints, are set to derive the optimal transformation scheme. Finally, the effectiveness of the proposed method is verified through a case study. The results show that the existing PIES after transformation achieves a 63.79% reduction in carbon emission and a 17.23% reduction in operation cost.关键词：integrated energy system;system integration;low-carbon transformation;hydrogen energy;tiered carbon trading;Fisher optimal segmentation7|1|0更新时间：2026-08-21
- 摘要：To address the challenges of high capacity allocation costs of storage and transportation equipment and the complex layout of energy transmission networks in the off-grid renewable energy based hydrogen production and ammonia synchesis integrated systems, a coordinated topology and capacity planning method is proposed. By establishing a mixed-integer second-order cone programming model covering the entire chain of the power networks, energy storage, hydrogen production, hydrogen transmission, hydrogen storage, and ammonia synthesis, with the objective of minimizing the annual comprehensive cost of the system, the equipment capacity allocation, electricity/hydrogen transmission network topology, and system operation strategies are collaboratively optimized, and a branch-and-bound based Benders decomposition algorithm is constructed for solution. Meanwhile, the information-gap decision theory (IGDT) is introduced to characterize the economic performance boundaries of various topologies under extreme conditions. A case study based on a practical system in Northern China demonstrates that the electricity-hydrogen hybrid transmission topology can significantly reduce total system costs, achieving markedly improved economic performance compared with the electricity transmission configuration and hydrogen transmission configuration, while also exhibiting favorable robustness and opportunity.关键词：hydrogen production;renewable energy;green ammonia;hydrogen pipeline;topology optimization;capacity optimization;coordinated planning2|1|0更新时间：2026-08-21
- 摘要：The hydrogen energy microgrid achieves effective accommodation of renewable energy through hydrogen energy storage and can mitigate the impact of renewable energy fluctuations on the power grid, which is the key technical pathway to support the high-proportion integration of renewable energy into the grid. To improve the utilization efficiency of renewable energy and achieve economic operation of the system, it is necessary to optimize the capacity allocation of hydrogen energy microgrids, reasonably planning the power capacities of electrolyzers, hydrogen storage devices, electrical energy storage systems, and fuel cells based on the output characteristics of renewable energy. To this end, an optimization model for capacity allocation with economic objectives is constructed, and Benders decomposition is employed to divide it into two stages: the first stage involves capacity allocation, while the second stage performs optimal scheduling based on the given equipment capacities. To ensure the safe operation of electrolyzers, dynamic safety constraints for electrolyzers are introduced in the second-stage scheduling optimization, replacing the traditional constant load lower constraints. Simulation results demonstrate that, by incorporating dynamic safety constraints in advance during the planning stage, this method can generate capacity allocation schemes that are more conducive to the safe operation of microgrids, providing the necessary hardware foundation and capability guarantees for system safety. Additionally, sensitivity analysis results indicate that the nonlinear hydrogen production characteristics of electrolyzers have a significant impact on the overall profitability of hydrogen energy microgrids.关键词：microgrid;renewable energy;electrolyzer;green hydrogen;capacity allocation;dynamic security constraint3|1|0更新时间：2026-08-21
- 摘要：Large-scale power-to-hydrogen by water electrolysis is an important pathway for electro-hydrogen coupling development. The combined operation efficiency of multiple electrolyzers and the overall safety of hydrogen production are key issues in the control strategy design of the power-to-hydrogen system. Therefore, based on pressure-efficiency characteristics of the system, a system-level model is established by integrating the pressure coupling throughout the production process. The measures for operating pressure adjustment control are then analyzed. Furthermore, the influence of electrolyzer start-stop on the operating pressure characteristics is considered. Aiming at minimizing the operation cost, a model predictive control strategy is proposed, which incorporates electrolyzer start-stop dynamics to coordinate operating pressure among devices and balance the operation time across the electrolyzer array.The case study shows that operating pressure control can improve the overall system efficiency and increase the hydrogen production while ensuring pressure balance. Meanwhile, the operation economy is also significantly improved compared with existing strategies.关键词：renewable energy;power-to-hydrogen;pressure;electrolyzer;model predictive control;coordinated control1|2|0更新时间：2026-08-21
- 摘要：Addressing the issue that power distribution in wind-hydrogen-storage systems often fails to fully account for the coordinated operation of electrolyzer cold/hot start delays and energy storage systems, a refined power distribution strategy is proposed for system considering the start-stop characteristics of hybrid electrolyzers. First, a mathematical model of hybrid electrolyzers is developed by considering the operation characteristics of alkaline electrolyzer (AEL) and proton exchange membrane electrolyzer (PEMEL), along with a grouping and rotation control strategy designed to achieve rational intra-group and inter-group power distribution within the hybrid electrolyzer array. Next, a grid-forming energy storage system is introduced, enabling dynamic coupling with the start-stop and rotation phases of the electrolyzers while ensuring system dynamic response capability and hydrogen production efficiency. Furthermore, to enable precise power allocation, a fuzzy segmentation-based power distribution algorithm model is proposed to enhance the system coordinated control capability. Finally, simulation results demonstrate that, compared with a single PEMEL array under a simple time-averaged allocation strategy and a single AEL array without rotation control, the proposed power allocation strategy balances the operation time and start-stop frequency of each hybrid electrolyzer group, increases hydrogen production, and improves system energy efficiency.关键词：hydrogen production;power distribution;hybrid electrolyzer;start-stop characteristic;grid-forming energy storage1|1|0更新时间：2026-08-21
- 摘要：Driven by the goals of carbon emission peak and carbon neutrality, hydrogen energy, as a zero-carbon renewable energy, is playing an increasingly vital role in the energy transition. Utilizing renewable energy such as wind and solar to directly drive alkaline electrolyzers for hydrogen production can, on one hand, improve the utilization rate of renewable energy, and on the other hand, the generated hydrogen can replace traditional fossil fuels, effectively reducing environmental pollution. However, in renewable-powered alkaline electrolysis hydrogen production systems, time-varying renewable energy can affect the temperature of alkaline water electrolyzers, thereby altering their structural parameters and rated power. Under these conditions, if constant droop coefficients are still employed, it is easy to lead to uneven probability distribution. Additionally, changes in the parameters of alkaline electrolyzers can deteriorate the control performance of traditional proportional-integral (PI) controllers. To address this issue, an adaptive droop control strategy based on virtual thermistors and an adaptive fuzzy PI controller is proposed. Compared with the constant droop coefficients in traditional droop control, the droop coefficients in the proposed adaptive droop control are adjusted according to the real-time measured electrolyzer temperature. The core of the adaptive fuzzy PI controller involves modifying the fuzzy universe of discourse based on the electrolyzer temperature. Hydrogen production electrolysis experiments show that this control strategy can not only achieve reasonable power distribution among electrolyzers but also enhance the dynamic performance of electrolyzers.关键词：hydrogen production;power allocation;stability;renewable energy;adaptive fuzzy proportional-integral controller;adaptive droop control;virtual thermistor1|2|0更新时间：2026-08-21
- 摘要：Off-grid photovoltaic (PV) hydrogen production enables the on-site conversion of clean energy by regulating the power of water electrolysis equipment, offering both economic and environmental advantages due to its grid-independent nature. To enhance the operation stability and economy of DC-coupled off-grid PV hydrogen production systems without energy storage, this paper proposes a system capacity optimization method and a corresponding control strategy. Firstly, an optimal model is established based on the operation characteristics of the PV array and the electrolyzer, taking the levelized cost of hydrogen (LCOH) and the PV curtailment rate as the objectives for optimizing the PV capacity allocation. Secondly, considering the operation safety limits of the hydrogen production equipment, a PV control strategy is proposed that integrates the maximum power point tracking functions with the power security constraints. The control parameters are then optimized through a multi-objective design. Finally, a case study demonstrates that, compared with a non-optimized hydrogen production system, the optimized hydrogen production system reduces the LCOH by 6.96% and increases the annual hydrogen yield by 23.2%. Hardware-in-the-loop experiments validate the effectiveness of the proposed control method, ensuring the safe, stable, and efficient operation of the hydrogen production system.关键词：photovoltaic;hydrogen production;elecctrolyzer;off-grid;capacity allocation;maximum power point tracking;power security constraint;control parameter design0|2|0更新时间：2026-08-21
- 摘要：The green energy transition in rural areas is a crucial pathway to achieving the goals of carbon emission peak and carbon neutrality and the vision of beautiful countryside. Addressing the challenges of diverse energy demands and the spatial-temporal mismatches of seasonal multi-energy loads, this paper proposes a time-series production simulation method for a rural distributed energy system (RDES) with electricity-hydrogen-ammonia-heat multi-energy coupling. This method leverages the complementary characteristics of heterogeneous energy flows across long- and short-term timescales to enhance the reliability of multi-energy supply and improve the comprehensive energy utilization efficiency. By analyzing the energy conversion pathways of the RDES from renewable energy power generation input to multi-energy load output, a hub model is developed to quantify the production, storage, conversion, and utilization processes of various energy flows, including electricity, hydrogen, ammonia, and heat, under steady-state operation conditions. To further improve the computational efficiency, a conjugate distributed parallel solving algorithm is designed, which balances temporal characteristics with solving speed. This algorithm decomposes and iteratively optimizes the high-dimensional sequential production simulation problem across 8 760 hours of the whole year through layer-by-layer decomposition and progressive iterative optimization. Taking the rural energy system of a Southern Province, China, as a case study, the proposed method is validated for its advantages in enhancing the operation economy and comprehensive energy utilization efficiency of the RDES.关键词：rural distributed energy system (RDES);electricity;hydrogen;ammonia;heat;time-series production simulation;multi-energy coupling;renewable energy0|1|0更新时间：2026-08-21
- 摘要：Hydrogen energy, characterized by high energy density and cleanliness, plays a crucial role in promoting energy structure transformation when introduced into integrated energy systems. To fully leverage the advantages of hydrogen energy in reducing carbon emissions and enhance system economic operation, a two-stage low-carbon economic dispatch model for gas-hydrogen-blended integrated energy systems based on the extreme scenario method is proposed. Firstly, the energy conversion forms and equipment coupling characteristics of power to gas (P2G) and gas to power (G2P) processes are analyzed. A variable-efficiency dynamic operation model for electrolyzers and a refined mixed combustion output model for hydrogen-blended gas turbines that thoroughly considers the energy conversion process are established. Secondly, based on the extreme scenario method, a data-driven uncertainty set for wind and photovoltaic output is constructed. A day-ahead and intraday two-stage low-carbon economic dispatch model for the integrated energy system is proposed. In the day-ahead stage, the scheduling plan minimizes unit costs, spare capacity costs, energy acquisition costs, and curtailment costs for the wind and photovoltaic power. In the intraday stage, real-time corrections to the day-ahead plan are made with the objective of minimizing operation costs of each equipment in the system, carbon trading costs, and adjustment costs. The model is then solved using a rolling optimization approach. Finally, simulation results demonstrate that the proposed model effectively promotes the low-carbon economic operation of the system and improves the model solving speed.关键词：integrated energy system;hydrogen energy;gas-hydrogen-blended;low-carbon economic dispatch;energy conversion;extreme scenario method;refined modeling;data-driven3|1|0更新时间：2026-08-21
- 摘要：Distributed power-hydrogen coupled systems (DPHCS) can significantly enhance the renewable energy capacity and flexible control capabilities of distribution networks through cross-temporal energy conversion and dynamic power regulation. However, limited research has explored medium- and long-term operational optimization of DPHCS coordinated distribution networks. To address this gap, a medium- and long-term operational optimization method for distribution networks with coordination of DPHCS based on safe reinforcement learning is proposed. First, a medium- and long-term optimization model for distribution networks incorporating flexible regulation characteristics of DPHCS is established and formulated as a Markov decision process, enabling efficient cross-temporal energy transfer while optimizing both operational costs and long-term system performance. Next, to address the issue of operational constraints being violated due to the randomness of distributed resources, a DistFlow-based proximal policy optimization (DFPPO) algorithm is proposed. In this algorithm, a domain-knowledge safety layer is designed to map infeasible actions into the security region, ensuring the strict satisfaction of distribution network operational constraints. Furthermore, to solve the problem of high computational cost and low efficiency of power flow calculation during long-horizon training, a Laurent series-based linearized power flow method is introduced, significantly improving the computational efficiency of reinforcement learning in long-term and high-frequency interactions. Finally, the simulation on the IEEE 33-bus and IEEE 118-bus systems demonstrates that the proposed method can effectively enable the cross-seasonal energy transfer. The average operation cost can be reduced by approximately 27% while ensuring the operational safety over medium- to long-term scales, demonstrating significant advantages in both safety and economic efficiency.关键词：power-hydrogen coupled system;distribution network;operational optimization;safe reinforcement learning;domain-knowledge safety layer;security region;power flow0|1|0更新时间：2026-08-21
Key Technologies of Renewable Energy for Coupling of Electricity\, Hydrogen\, and Ammonia Towards Carbon Neutrality
-
Impact Assessment and Optimization Decision Method for Multi-timescale Transient Frequency Stability
摘要：High-penetration renewable energy power systems lack control performance indicators for various types of frequency regulation devices in the context of transient frequency stability, making it difficult to effectively achieve the coordinated and optimization decision and control of multiple measures under the spatio-temporal distribution characteristics of frequency after large power grid disturbances. This paper proposes a calculation method for control performance indicators that meets the transient frequency stability control requirements of multi-timescale devices. The method requires only a single time-domain simulation to determine the impact degree of all control nodes in the power grid during different periods of the transient process. The calculation period of the impact degree is determined according to the timescale of the active power-frequency dynamic response of the devices, and the dominant time point set of each node is determined based on the frequency of each monitoring node within the period. On this basis, the active power sensitivity of frequency regulation devices to the active power inflow of connected branches during the dynamic simulation process is used as an approximate frequency variation index to further calculate the transient frequency stability impact degree of control nodes on dominant nodes within the period. Finally, the above calculation method of impact degree is applied to the optimization decision of system transient frequency stability control, and its effectiveness and accuracy are verified through an electromagnetic transient simulation platform of a high-penetration renewable energy power system constructed based on an actual power grid.关键词：renewable energy;transient frequency stability;multi-timescale;frequency regulation;impact degree1|1|0更新时间：2026-08-21 - 摘要：In bipolar DC distribution networks, the asymmetric integration and stochastic characteristics of DC sources and loads can lead to various voltage issues, such as inter-pole voltage unbalance and voltage violation resulting from uneven power distribution. To address these challenges, a multi-dimensional flexibly interconnected device with multi-dimensional power regulation capabilities is developed, based on which a comprehensive voltage regulation strategy for bipolar DC distribution networks is proposed. Firstly, the topological structure and power transmission model of the flexibly interconnected device are designed and analyzed, with a focus on its multi-dimensional regulatory characteristics. Secondly, the power transmission models for both mode-switchable direct current transformers and energy storage systems are established. Next, a comprehensive voltage regulation model is developed to minimize the operation costs and penalties for voltage violations, taking into account the coordination between bipolar flexibly interconnected devices and other regulatory devices. Thirdly, the linearization of this model is achieved through techniques such as convex relaxation and piecewise linearization. Finally, the effectiveness and adaptability of the proposed strategy are validated through case studies in 33-bus and 123-bus DC distribution networks.关键词：DC distribution network;voltage unbalance;flexibly interconnected device;voltage regulation;convex relaxation0|1|0更新时间：2026-08-21
- 摘要：Under the strategy of carbon emission peak and carbon neutrality, the high penetration of renewable energy imposes higher requirements on the secure and stable operation of power systems. How to enhance economic performance under a secure operational framework has become a significant issue. To address this, this paper proposes an optimal dispatch method integrating improved convex inner approximation with feasible solution recovery. First, a dynamic boundary mechanism is proposed to refine the convex inner approximation method and construct an optimization model. This approach strictly ensures security while significantly expanding the feasible region, thereby establishing a safety foundation for better economic performance. Second, by incorporating a feasible solution recovery mechanism, the initial feasible solution undergoes gradient-guided, refined and coordinated adjustment of active and reactive power based on sensitivity information. After being fully validated by AC power flow analysis, a more secure and economically favorable dispatch plan is ultimately obtained. Finally, the simulation conducted on IEEE 33-bus, PG&E 69-bus, and IEEE 118-bus distribution systems using the YALMIP-CPLEX platform demonstrates that the proposed method achieves efficient dispatch while rigorously ensuring security.关键词：active distribution network;power flow;convex inner approximation;feasible solution;feasible region;optimal dispatch4|1|0更新时间：2026-08-21
- 摘要：Aiming at the challenge of how to fairly allocate the costs of electricity and carbon markets to electric vehicle (EV) users through pricing strategies while incentivizing their collaboration in promoting low-carbon operation of the system under the environment of dual electricity and carbon markets, this paper proposes a low-carbon economic dispatch strategy for the multi-energy microgrid with EVs based on the electricity-carbon pricing mechanism using Stackelberg game. First, based on an extended carbon emission flow model for the multi-energy microgrid with EVs, a carbon emission accounting model for charging and discharging of EV clusters is constructed. Next, a Stackelberg game market framework is designed for the multi-energy microgrid operator and EV users, and a dynamic electricity-carbon pricing model for EVs is developed based on the carbon emission accounting model. Finally, a particle swarm optimization (PSO) algorithm is employed in conjunction with the CPLEX solver to solve this bi-level Stackelberg game model, yielding a low-carbon economic dispatch strategy for the multi-energy microgrid with EVs. Simulation results validate that the proposed strategy ensures economic benefits for both parties while effectively utilizing the nodal carbon emission factors combined with the carbon price. Employing the dynamic electricity-carbon price as an economic incentive, it guides EV users toward low-carbon energy consumption behaviors and enhances the renewable energy accommodation and carbon emission reduction of the multi-energy microgrid.关键词：microgrid;low-carbon economic dispatch;electric vehicle (EV);electric vehicle carbon storage factor;Stackelberg game;electricity-carbon price;extended carbon emission flow3|1|0更新时间：2026-08-21
- 摘要：Driven by the goals of carbon emission peak and carbon neutrality and various policies, installation capacities of rural distributed photovoltaic (PV) have grown rapidly. The integration of large-scale PV has led to severe reverse overload in numerous aquaculture-PV distribution station areas across Southern China. Meanwhile, the fish pond aquacultural loads demonstrate significant flexibility potential due to their predictability and adjustability. Therefore, this paper proposes a reverse overload mitigation strategy for the high-penetration aquaculture-PV distribution station areas based on the coordinated control of flexible aquacultural loads and energy storage. Firstly, the most critical PV output scenario is extracted from the annual operation data of an actual distribution station area, and the corresponding net-load curve is constructed to reveal the typical characteristics of reverse overload. Secondly, a differential dynamic model describing the oxygen content evolution of a single aerator is established based on the oxygen demand mechanisms of fish and the coupling characteristics of aerator. This model is then used to derive the non-adjustable nighttime aeration load and to quantify the shiftable and transferable capabilities and their demand-response proportions of feeding-related aeration and sunny-day supplemental aeration load. On this basis, a coordinated optimization model combining flexible loads and energy storage is developed to prevent reverse power violations and enhance PV accommodation capability. Finally, case studies are conducted on a modified IEEE 33-bus aquaculture-PV distribution station area and other topological systems. Results demonstrate that the proposed strategy can effectively eliminate the reverse overload, significantly reduce operation costs, and maintain good applicability and economy across distribution station areas with varying scales.关键词：photovoltaic (PV);distribution station area;reverse overload;aerator;aquaculture-photovoltaic complementation;aquacultural load;energy storage;demand response1|1|0更新时间：2026-08-21
- 摘要：China’s wind-power sector has entered a cluster-based development stage. However, the few-shot problems caused by cross-regional spatio-temporal feature mismatches and limited operational data from newly built farms seriously restrict the accuracy and generalization ability of forecasting models. This paper proposes an adaptive dual-predictor domain adversarial neural network (ADP-DANN) that effectively enables few-shot forecasting problems through multi-level cross-domain knowledge transfer. Firstly, a dynamic adaptation layer is designed to flexibly handle heterogeneous inputs with varying numbers of wind farms between the source and target domains, ensuring the integrity of input information. Secondly, by integrating sequence memory and shared feature extractor with context mining, combined with domain adversarial training, efficient extraction and alignment of domain invariant spatio-temporal features have been achieved. Finally, a dual-predictor structure is established to adapt independent forecasting functions to the target domain based on shared features, achieving fine-grained transfer from feature space to task space. The comprehensive experiment based on measured data from multiple stations shows that the proposed ADP-DANN outperforms the baseline model in various forecasting time-scales. The ablation experiment further validates the indispensability and synergistic effects of core components such as the dynamic adaptation layer, dual-predictor, and domain adversarial mechanism. In addition, with the help of t-distribution stochastic neighbor embedding feature visualization and SHAP interpretability analysis, the effective domain alignment ability of the model in the feature space and the physical rationality of its forecasting decisions have been confirmed from both data and mechanism perspectives. This provides a high-performance and clear-mechanism transferable modeling framework for few-shot power forecasting of wind power clusters, which is of great value for intelligent power grid scheduling and efficient accommodation of renewable energy.关键词：wind power cluster;power forecasting;adaptive dual-predictor;domain adversarial neural network;few-shot;transfer learning1|1|0更新时间：2026-08-21
- 摘要：With the advancement of the “Thousands of Households Embracing Sunlight Initiative”, an increasing number of distributed photovoltaic (PV) installations are being placed behind smart meters. Their unobservable nature poses significant challenges to distribution network planning and operation, highlighting an urgent need for accurate and reliable methods for behind-the-meter PV disaggregation. However, existing disaggregation methods remain inadequate in extracting and deeply integrating periodic temporal features and spatial correlation features. To address this issue, this paper proposes a behind-the-meter PV disaggregation method based on adaptive extraction and state transition fusion of spatiotemporal features. First, an adaptive adjacency matrix is constructed to dynamically characterize the spatial correlations among users, mapping net load data into a dynamic spatiotemporal graph structure. Second, the net load data is periodically decoupled by integrating time-domain and frequency-domain information, enabling the extraction and aggregation of multi-scale periodic features. Subsequently, graph diffusion convolutional neural networks are employed in the spatial dimension to extract wide-area spatial features. Finally, a selective spatiotemporal state model based on dynamic spatiotemporal graph embedding is used to deeply integrate the spatiotemporal features and generate behind-the-meter PV disaggregation results. Experimental results on a real-world dataset comprising 268 users demonstrate that the proposed method reduces the mean absolute error by 58.8% compared to existing disaggregation methods. Even under a metering data missing rate of up to 15%, the error is reduced by at least 61.7%.关键词：photovoltaic disaggregation;dynamic spatiotemporal graph;spatiotemporal feature;state transition;graph diffusion convolutional neural network;selective spatiotemporal state model1|0|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