With the increasing penetration of renewable energy, ensuring reliability and security of power supply has become a significant challenge. In this paper, a novel photovoltaic (PV) intraday power supply guarantee capability forecasting method is proposed. Different from the conventional PV power forecasting methods, it can provide diverse forecasting information, including guarantee power, low output power period, and power supply guarantee probability. PV guarantee power which represents a conservative prediction of PV power is forecasted based on convolutional neural network-bidirectional gated recurrent unit (CNN-BiGRU) model with a compound loss function. PV low output power period represents the period when the power gap between the theoretical maximum PV power and the actual PV power is higher than a specific threshold. It is forecasted using hybrid gradient boosting decision tree and logistic regression (HGBDTLR) model based on an improved spatiotemporal feature encoding method. The power supply guarantee probability which represents the risk of PV power shortage is obtained using quantile regression model based on gradient boosted regression tree (GBRT) considering different PV power supply demands. Furthermore, new indexes, including guarantee rate, guarantee energy ratio, success index, forecasting gain index, and probability forecasting accuracy are proposed to evaluate the forecasting performance. The effectiveness of the proposed method is verified based on actual operation data in a province in Northwest China.

PV power forecasting is essential for the stable and efficient operation of power systems. However, forecasting accuracy is compromised by two main factors: the quality of solar irradiance predictions and the limited flexibility of forecasting methods. In this paper, an ultra-short-term PV power forecasting method for the SAT PV system is proposed based on irradiance transformation and the CNN-BiGRU model. First, a SAT-based irradiance transformation is adopted to convert the horizontal-plane NWP irradiance data into the effective irradiance received by the SAT PV panels. Then, the FCM clustering algorithm is used to classify weather types, improving the adaptability of forecasting models to diverse meteorological conditions. Finally, a hybrid CNN-BiGRU network is developed to capture spatiotemporal features for accurate power prediction. The proposed method is validated using two-year operational data from a 50 MW PV plant in Northwest China and a 60 MW plant in Southeast China. A high degree of forecasting accuracy is observed. For the Northwest plant, the MSE values are 2.46, 4.53 and 3.64 MW2 under sunny, cloudy and rainy conditions, respectively; corresponding results for the Southeast plant are 2.62, 4.84 and 3.80 MW2. Consistent improvements are also observed in MAE, RMSE, R2 and FS metrics. Comparative tests confirm the superiority of the proposed model over state-of-the-art models, demonstrating that the integration of irradiance transformation and CNN-BiGRU model significantly enhances forecasting performance for SAT PV systems.

Under the global demand for carbon emission reduction, distributed energy resources (DERs) in the power energy industry are booming, and virtual power plant (VPP) has become an effective solution for aggregation and management of DERs. Aiming at the deficiencies of VPP’s sources aggregation, such as single aggregation mode, insufficient preference value reflection of aggregation mechanism, and low efficiency of aggregation operation, a dynamic aggregation strategy for two-way selection between DERs and VPP is proposed based on evolutionary game in this study. Firstly, the operation framework of dynamic aggregation based on two-way selection of multi-type energy subjects is constructed, and the benefit model for aggregation operation of multi-type energy subjects is established. Secondly, the aggregate preference index system and preference selection model for each subject are established with a combined consideration of the market environment, the comprehensive benefit model for two-way selection between multi-type DERs and VPP is established based on potential game theory, and the nDERs-mVPPs evolutionary game decision model and calculation method are further constructed by introducing evolutionary game theory. Finally, cases are designed to verify the above models and methods. The simulation results reveal that the aggregation strategy for two-way selection can predict the market aggregation trend and fully reflect the individual preference and independent selectivity of the aggregation subjects. The dynamic aggregation mechanism can improve the overall operation benefit of the energy system by 17.61%. Furthermore, this aggregation strategy greatly improves the enthusiasm of aggregation participants and guarantees the stability of aggregation results, promoting VPPs to exert multi-energy complementary effects, thus demonstrating strong economic benefits and high promotional value.

With the rapid development of smart cities, coordinating diverse distributed energy resources through storage-centric shared management has become a critical challenge. This paper proposes a bi-level energy management framework to support peer-to-peer energy trading among multiple virtual power plants (VPPs) under multidimensional uncertainties. The interaction is modeled as a Stackelberg-Nash equilibrium framework, in which OK, we will make the necessary revisions as per the requirements.a public energy storage operator and a natural gas company act as leaders to maximize social welfare and design differentiated trading strategies for VPPs. The VPPs act as followers and participate in cooperative energy trading based on a generalized Nash equilibrium scheme, sharing surplus energy and allocating cooperative benefits according to their contributions. To address uncertainty, Conditional Value at Risk (CVaR) is adopted to quantify the expected loss of the upper-level decision makers. The lower-level VPP problem is formulated as a three-stage stochastic robust optimization model considering renewable generation uncertainty. To solve the resulting nonlinear bi-level problem, a two-stage solution approach combining particle swarm optimization and KKT-based reformulation is developed to transform it into a tractable mixed-integer linear programming model. Numerical case studies verify the effectiveness of the proposed framework.

Extreme disasters, such as typhoons, can cause serious damage to lines, which poses a significant threat to the power supply of critical loads. An effective approach to enhance supply security is the flexible adjustment of the power system topology to redirect power flows and optimize load transfer. However, many studies have addressed optimal transmission switching (OTS) and distribution network reconfiguration (DNR) in isolation, with limited research on coordinated transmission and distribution topology optimization for improving resilience. This gap may result in overly conservative preventive dispatch strategies and an underestimation of resilience. To overcome these limitations, this paper incorporates coordinated topology optimization measures, including OTS, DNR, and post‑fault line repair, into a unified preventive scheduling framework to withstand disasters. An improved alternating direction method of multipliers (ADMM) algorithm, with a penalty multiplier updating strategy based on objective function deviation, is employed to solve the proposed preventive scheduling model. Finally, the proposed method is validated by two modified test systems. In the test system consisting of one IEEE-118 and five IEEE-33 bus systems, the results indicate an 8.91% reduction in load shedding and a 2.89% decline in total cost.

To enhance the flexibility and economic performance of integrated thermoelectric energy systems, this study develops an optimal operation strategy for supplementary electric heating in heat-exchange stations by explicitly exploiting the virtual thermal energy storage (VTS) potential of district heating networks (DHNs) and buildings. The VTS of the pipe network is modeled based on adjustable supply water temperature, while that of buildings is modeled based on allowable indoor temperature variation, enabling unified scheduling within the heat–electric coupling framework. A coordinated operation model is established for the combined heat and power (CHP) system with supplementary electric heating under time-of-use electricity pricing, aiming to minimize the total operating cost while satisfying thermal comfort and operational constraints. The model is expressed as a mixed-integer linear programming (MILP) problem and solved with the CPLEX optimizer. Case studies demonstrate that the proposed strategy achieves significant cost and carbon-emission reductions: compared with the scenario without VTS, the energy purchase cost decreases by ∼14%, and carbon-emission-related costs are also reduced. Further analysis reveals that the DHN provides a more substantial storage effect (reducing electricity purchases by 1.1%) than buildings (0.1%), highlighting the need to prioritize network heat storage in practical applications. These results confirm the effectiveness, economic advantages, and low-carbon potential of the proposed approach, providing a viable pathway for clean heating systems to reduce dependence on fossil fuels.

Following the proposal of "carbon neutrality and peak carbon emissions" goals, electrical energy substitution has emerged as a key strategy for sustainable development across various industries. As a crucial component of agricultural development globally, this paper introduces a particle swarm optimization vector regression method to predict the potential for electrical energy substitution in the fisheries sector. Initially, this article analyzes the key factors influencing the electrical energy substitution in fisheries from technological, economic, and policy perspectives. To explore how these various factors influence the potential for energy substitution, a Support Vector Regression (SVR) algorithm is applied. This SVR model's performance is subsequently improved by optimizing its parameters using Particle Swarm Optimization (PSO).Compared to a BP neural algorithm, the enhanced particle swarm optimization (PSO)-SVR model demonstrated markedly superior prediction accuracy and goodness-of-fit. Consequently, this model was effectively utilized to generate a forecast of the electrical energy substitution potential within China's fisheries in recent years. This study provides theoretical and data support for promoting electrical energy substitution in the fisheries sector and offers guidance for analyzing the potential for energy substitution.

The inherent intermittency and fluctuation of renewable energy generation introduce uncertainty in electricity carbon emission intensity, posing challenges to the low-carbon and economic scheduling of integrated energy systems. To this end, this paper proposes an optimization model for integrated electricity-heat-gas energy system operation considering uncertain indirect carbon emission intensity. First, considering the tiered carbon trading mechanism, an optimization framework for a park-level model is developed. Then, polyhedral uncertainty sets are introduced to capture the uncertainties in renewable energy generation and carbon emission intensity, offering a flexible and intuitive approach to handling diverse uncertainty types. These uncertainty sets are integrated into the optimization problem to enhance robustness. Finally, the uncertainty set-based stochastic optimization method is introduced to improve the model's robustness, addressing both cost minimization and carbon reduction under uncertain conditions. Case studies using data from a Mediterranean-region hospital demonstrate the model's effectiveness in addressing uncertainties at park-level systems and providing robust solutions that balance cost efficiency and environmental impact.

The rapid development of power system has significantly increased the number and proportion of distributed energy resources (DERs) and flexible loads (FLs), presenting substantial challenges to the safe and stable operation and economic dispatch of distribution network (DN). Virtual Power Plant (VPP) technology addresses these challenges by integrating DERs and FLs for unified control and dispatch. However, existing VPP aggregation methods suffer from issues including ignoring power loss, using fixed aggregation ranges and quantities, and relying on a single evaluation indicator. This paper presents a novel VPP aggregation method based on Virtual Contribution Theory (VCT) and Community Detection Algorithm (CDA). First, a graph network-based electrical intensity model is developed using VCT, which fully considers the impact of electrical intensity on the power transmission paths between nodes, resolving the issue of insufficient power loss management during aggregation and transactions. Next, an improved CDA algorithm is introduced to dynamically identify the optimal number and configuration of VPPs, overcoming the limitations of traditional methods in reflecting electrical dynamics, parameter selection, and local optimization. To guide and evaluate VPP aggregation, a comprehensive index system is established, focusing on modularity based on electrical intensity, power balance capability, and aggregation stability-addressing gaps in current aggregation performance evaluation. Finally, simulation results using the IEEE 69-bus system demonstrate that: 1) The proposed method reduces overall power loss by 17.07% by allowing VPP to bear only a small portion of branch losses in the power market. 2) By optimizing modularity, this method uniquely determines the aggregation range and the number of VPPs, with stable and optimal results. 3)The proposed method performs consistently across various operating scenarios in terms of modularity, power balance, and aggregation stability. 4) The proposed method exhibits lower computational complexity than conventional aggregation methods, significantly improving the speed of VPP regulation and enhancing operational efficiency. This method improves resource utilization efficiency for VPP operators and provides effective regulatory tools for market regulators, supporting stable operation of the DN and ensures fairness and transparency in transactions.

The time-varying characteristics of distributed generation (DG) output upper limit and load demand bring new challenges to the safe and economic operation of active distribution network (ADN). Based on the dual uncertainty of DG output and load demand, a novel cooperative dynamic optimization method of network reconfiguration (NR) and DG scheduling is proposed considering the security and economic objectives of ADN. Firstly, the ADN operation is classified into normal and risky states according to the node voltage and branch load rate. Secondly, the mathematical models of cooperative dynamic optimization for NR and DG scheduling in different states are developed, respectively. Thirdly, a co-evolutionary algorithm of Improved non-dominated sorting genetic algorithm II (INSGA-II) and improved particle swarm optimization algorithm (IPSOA) is designed to solve the proposed model. Finally, the effectiveness of the proposed model and method is verified in the IEEE 33-bus distribution system.

Electromagnetic Transients Programs (EMTP) have been widely employed for power system electromagnetic transient simulation. With deepening interdependencies between electrical systems and natural gas (NG) systems, cascading failure risks escalate significantly. It is beneficial to extend the application of the EMTP-type to multi-physical transients in integrated electricity and gas systems (IEGS). This study proposes a dynamic modeling method for NG systems based on circuit analogy. Through the analogy between electrical and pneumatic quantities, an equivalent circuit model of NG pipelines is established, which accounts for time-varying fluid resistance and has constant impedance characteristics. The modeling method is extended to fault scenarios, and equivalent circuit models for leakage and blockage faults are constructed to achieve accurate description of key parameters such as pressure and mass flow under fault conditions. The key technologies of energy conversion device integration in IEGS are systematically summarized, including the detailed dynamic models of the power-to-gas (P2G) and the gas turbine. A novel energy conversion device-based interface model is proposed to characterize the interactions between electrical and NG networks. Case studies are conducted to investigate the dynamic characteristics of IEGS under multiple scenarios and the effects of NG network failures on IEGS performance.

To address the challenges of distributional discrepancies and partially inconsistent class labels between source domain and target domain data in real-world engineering applications, this study proposes an open-domain adaptation method based on a Dynamic Convolutional Graph Network and bi-classifier adversarial learning (DCGNDAT). First, a dynamic convolution module is introduced to replace conventional convolution layers, enhancing the model's ability to extract fault-related features from vibration signals. Second, a graph network is employed to encode structural information, thereby improving the model's representation of complex data. Finally, a bi-classifier adversarial training mechanism is designed, incorporating entropy maximization and minimization strategies for the source and target domains, along with a binary cross-entropy scheme for target domain outputs, to accurately delineate the boundaries between known and unknown classes. Fault diagnosis experiments conducted on bearing and self-priming centrifugal pump datasets demonstrate that the proposed method achieves significantly higher recognition rates for unknown fault categories compared to existing algorithms, validating its superior diagnostic performance and robust open-domain adaptability.

With the large-scale integration of high-penetration distributed energy resources and the transformation of terminal loads into prosumer models, the operation of active distribution networks (ADNs) has become increasingly flexible, often leading to frequent bidirectional power flows. These dynamics pose significant challenges for accurate loss calculation in ADNs and the correction of transaction outcomes between generators and consumers under power market environments. Moreover, the distinct operating structures of the spot and bilateral contract markets further increase the complexity of ADN operation and impose significant challenges on the adaptability of loss allocation (LA) mechanisms. To address these issues, this paper proposes an innovative LA method for ADNs based on Virtual Contribution Theory (VCT), applicable to both spot and bilateral contract market scenarios. By constructing virtual contribution matrices—branch-based for the spot market and transaction-based for the bilateral market—the method accurately traces and quantifies the actual contributions of each market participant under different trading models, thereby enabling reasonable, transparent, and fair loss allocation. Additionally, the method eliminates the need for complex matrix inversion required in existing approaches by leveraging basic operations on the virtual contribution matrix, thereby reducing computational complexity and significantly improving computational efficiency. Case studies on IEEE 33 and IEEE 69 systems demonstrate the effectiveness, applicability, and fairness of the proposed approach.

With the continuous advancement of demand response (DR) research, significant progress has been made in its application in conventional electricity usage scenarios. However, studies focusing on special areas, such as offshore fishing farms, are relatively scarce. This paper focuses on the electricity consumption characteristics of offshore fishing farms and proposes a method for calculating DR potential. A load classification framework tailored to the specific features of fishing farms is established, and a DR model is constructed to maximize the benefits of fishing farm participation in DR while maintaining electricity comfort. A method for calculating DR potential is designed for different load characteristics. The case study analysis results show that DR can achieve a total load reduction of 34.6% for offshore fishing farms, with electric boats (EBs), feed mills, and lighting equipment contributing 73.8%, 19.2%, and 7%, respectively, to the overall DR potential. The findings provide theoretical support for offshore fishing farms' participation in DR and have significant implications for improving their power supply reliability.

The novel energy storage serves as a crucial infrastructure and a key enabling technology in shaping a new power system and advancing the green and low-carbon energy transformation. It stands as a vital support for achieving the dual-carbon strategy. In response to the high cost of deploying distributed energy storage and the potential for fraudulent behaviors in transactions, this research proposes an optimized operational strategy for fraudulent balancing in electricity trading considering the neighborhood scene public energy storage, taking into account potential fraudulent behaviors by market participants. Firstly, this research establishes a comprehensive framework for the joint operation of neighborhood scene public energy storage and multiple microgrids (MMG), developing the operational cost models for both MMG and public energy storage unit (PESU). Secondly, drawing upon the Nash bargaining theory, a cost minimization model is formulated to guide the collaborative operation of MMG and PESU. In addition, a transaction model is constructed to maximize profit distribution while considering fraudulent behaviors among market members. Finally, design a numerical example and solve it using the alternating direction method of multipliers (ADMM), the calculations suggest that the implementation of peer-to-peer (P2P) energy trading can increase the overall benefit by 65.45%, while the incorporation of PESU can boost the overall benefit by 49.93%, the fraudulent balancing model guarantees a sense of fairness in energy transactions. Addressing the potential obstacles to model expansion, further strategies involving the method of secondary distribution and setting an initial quotation range have been proposed to balance the fraudulent benefits among various entities and enhance participant enthusiasm. The fraudulent balancing strategy presented in this paper facilitates the achievement of fraud-balanced trading and maximization of profits among multi-energy entities, demonstrating considerable economic efficacy and high potential for widespread adoption.

Current methods for bearing fault diagnosis often fall short in addressing data privacy concerns and typically rely on one-to-one transfer strategies, which are inadequate for achieving knowledge transfer in distributed environments. To address this issue, a distributed fault diagnosis method for rolling bearings based on federated transfer learning is proposed. This method ensures data privacy while integrating fault knowledge from multiple domains, thereby enabling more efficient knowledge transfer. Specifically, a Domain Adversarial Neural Network (DANN) is introduced as the base model within the federated learning framework. Additionally, Maximum Mean Discrepancy (MMD) is incorporated into the DANN to enhance the transfer of fault knowledge. Finally, a dynamic weighting parameter update method based on MMD is designed to evaluate the feature discrepancies between source and target domains, thereby updating the parameters of the federated framework and achieving global model aggregation. Experimental results on two bearing datasets demonstrate that the proposed method excels in both distribution alignment and fault diagnosis.

To maximize the effectiveness of smart grids and integrated energy systems, it is essential to acquire accurate data on electricity consumption loads. However, current smart meters are still unable to collect specific power consumption data from individual electrical devices. Therefore, to address the issue, a low-cost, real-time remote monitoring load monitoring system based on Bluetooth mesh wireless network and IoT technology has been developed in this work. The Bluetooth mesh network consists of multiple client nodes, one server node and relay nodes (if necessary). The server node comprises a power metering chip and an ARM architecture Bluetooth microcontroller. It is responsible for measuring data, conducting initial filtering and processing on voltage and current signals, and transmitting the processed data to the service port via the Bluetooth mesh network. The client node, composed of a Bluetooth microcontroller and an ESP32 module, consolidates diverse data and transmits it to the cloud platform for storage via WIFI. Simultaneously, the coverage of the wireless power data collection system can be extended further through Bluetooth mesh relay nodes. A simple WeChat mini program display interface was developed to enable users to view the power detection status at any time during actual use, providing underlying data support for the deployment of users or power departments.

Under the background of dual carbon goals and rural revitalization strategy, a design problem of energy supply and use mode of circular economy in courtyards represented by aquaculture is proposed, which takes into account the reuse of manure in organic composting system. At the same time, the economic and environmental optimization model is used to verify its advantages. Firstly, the energy demand and electricity and heat load characteristics of circular economy in aquaculture courtyards are studied, and the load model of circular economy in aquaculture courtyards is established; secondly, the energy supply and demand relationship of circular economy in aquaculture courtyards is studied, the energy flow mode is divided and sorted out, and the energy supply and use mode is planned according to local conditions; then, the optimization operation model of circular economy in courtyards is established, and the optimization operation strategy of electricity and heat energy utilization and supply and demand matching is studied with the goal of economy and environmental protection; finally, based on the field data of Shengdeku Village, Fuyuan City, Heilongjiang Province, the optimization operation simulation of circular economy energy supply and use mode of aquaculture courtyards considering the reuse of manure in organic composting system is carried out based on the planning. The simulation results show that the designed mode can effectively reduce energy operation cost and carbon emission under the premise of meeting the demand of electricity and heat load.

The continuous development of informatization and digitization in active distribution networks has endowed them with the characteristic of multi -flow convergence, incorporating energy flow, information flow, control flow, and business flow. The combination of multi -flow fusion and business flow modeling in active distribution networks are examined in this work, and a method for constructing business flow model that considers the process of multi -flow convergence is described. The viability of modeling business flow that incorporates the fusion process is examined by studying the interactive process of multi -flow fusion. The operating states of the network and the characteristics of business events are defined by using the finite state machine approach to create the business flow model of active distribution networks. The experimental findings show that the suggested strategy successfully captures the business flow process in active distribution networks and satisfies the requirements for practical use. This study serves as a reference for business flow modeling and the use of finite -state machine models, and it offers useful insights for optimizing and managing the functioning of active distribution networks.

In order to efficiently deal with the problem of wind curtailment and carbon emission caused by the contradiction between heat and power supply and demand in integrated energy system (IES), this paper constructs an optimal scheduling model considering carbon capture and storage (CCS) technology and concentrating solar power (CSP) station. Firstly, CCS is used to transform the combined heat and power (CHP) unit into a low-carbon unit, and a mathematical model of the CHP unit with carbon capture is established. Then, on this basis, the CSP power station is introduced to form the CSP-CHP-CCS collaborative framework, and the IES low-carbon economic dispatch model with CSP-CHP-CCS is established. In addition, for the source and load uncertainties in the system, the information gap decision theory is used for simulation analysis, construct the risk aversion robust model and the risk-seeking opportunity model, respectively. Finally, the effectiveness of the proposed model in promoting new energy consumption and reducing carbon emissions is verified by simulation comparison.