Abstract:With the accelerated construction of new-type power system, the importance of diverse, widely distributed, and massive demand side resources for the power system continues to increase. The typical characteristics of three types of resources under the broad demand side resource category are explored: post metering resources, distribution network resources, and weakly connected self-balancing resources; the difficulties encountered in its development process and the challenges faced in governance are systematically analyzed, mainly manifested in institutional embedding, market identification, and system collaboration; Furthermore, the research proposed countermeasures and suggestions to improve the relevant governance mechanisms on the demand side of electricity and unleashed the potential of demand side resources from the four dimensions of system, market, technology, and credit under the unified national electricity market system.

Abstract:Against the backdrop of "dual-carbon" goals, growing installed capacity of new energy, changes in user load characteristics and increased load demand have intensified pressure on grid supply-demand balance. To maintain grid stability and fully tap the adjustable load potential of industrial users, an industrial user load potential assessment strategy based on a hybrid MPA-CNN-LSTM model combined with confidence interval correction is proposed. First, building on existing load characteristics, load reduction characteristics are introduced—describing the types and methods of load reduction among different users in the same industry—as inputs to the MPA-CNN-LSTM prediction model. Second, the MPA-optimized CNN-LSTM neural network is trained using actual adjustable potential data from responsive users to predict industrial users' adjustable potential. Finally, the confidence interval correction method is applied to refine the predicted adjustable potential, enhancing accuracy.

Abstract:Distributed generation (DG) in microgrid is generally connected through power electronic converter interface, which reduces the inertia of microgrid. The virtual synchronous generator (VSG) can provide inertia support for the microgrid. However, due to the inconsistent response speed to load changes, power frequency oscillation will occur between multiple DGs using VSG, and even cause the instability of the microgrid. To solve the above problems, a power frequency oscillation suppression strategy for microgrid with multiple VSGs is proposed considering the consistent second-order term. Firstly, the secondary control strategy of microgrid with multiple VSGs based on first-order consistency is established, and the mechanism of power oscillation between multiple VSGs is studied by using small signal analysis method. Then, the second-order term of consistency is introduced to provide damping for power frequency oscillation, and the adaptive gradient algorithm is used to optimize the second-order control coefficient to improve the suppression effect. Finally, the feasibility of the strategy is verified by a simulation example.

Abstract:Guided by the “dual carbon” goal, the AC/DC hybrid distribution network has gradually become an important direction for the future development of the distribution system. However, a certain contradiction exists between the two major demands of carbon reduction and improvement of power network resilience, and their efficient coordination is hampered. In view of this, a collaborative optimization method for AC/DC hybrid distribution networks based on approximate Nash equilibrium is proposed. Firstly, the impact of carbon reduction and resilience on the operation of AC/DC distribution networks under different conditions is explored. Subsequently, a collaborative regulation model considering carbon emissions reduction is proposed, and a measurement index for resilience evaluation is proposed. Based on this, an approximate Nash equilibrium strategy for optimized operation is designed. Ultimately, with the optimization goals of achieving balanced carbon reduction, economy, and resilience, an optimization and control model for the AC/DC hybrid distribution network is constructed. The effectiveness and feasibility of the proposed method in reducing carbon emissions, improving economic efficiency and resilience are verified through improved IEEE33-node examples and a certain park example in Xiong'an New Area.

Abstract:To achieve the strategic goal of carbon neutrality, optimizing the low-carbon operation of integrated energy systems (IES) through multi-energy coupling and synergy has emerged as a critical pathway for energy transition. Within the framework of multi-energy coupling and cooperative operation, a muti-dimensional optimal scheduling strategy incorporating multi-timescale analysis and load demand response is proposed. First, a multi-level coupled architecture for gas turbine-carbon capturing and storage-power to gas (GT-CCS-P2G) is constructed, and an integrated electricity-gas system (IEGS) considering comprehensive demand response is structured based on this architecture. Followed by the construction of the power grid and gas network as separate agents, and the conversion of the IEGS scheduling scheme into a Markov game process. Finally, the optimal scheduling strategy is obtained by the communication mechanism-enabled multi-agent soft actor-critic (CM-MASAC) method. Comparative analysis with multiple algorithms demonstrates both the superiority of the proposed method and the synergistic effectiveness of combining load demand response with GT-CCS-P2G technology. Experimental results indicate that this approach achieves optimal performance, reducing 10.29% operating costs and 16.07% carbon emissions compared to benchmark methods.

Abstract:The estimation of state of health (SOH) for lithium-ion batteries is considered a critical task to ensure battery reliability and enable accurate lifetime prediction. To address the limitations of existing SOH estimation methods in handling outliers and selecting effective features, a novel health state feature processing approach is proposed. An improved Hampel filter with adaptive thresholding and multidimensional feature fusion optimization is integrated. The method is composed of three stages. In the initial feature selection stage, ten candidate features are extracted from three perspectives: the charging process, the discharging process, and the capacity increment curve. In the feature correction stage, an adaptive-threshold Hampel filtering algorithm is developed, which dynamically adjusts the window size and threshold to correct abnormal feature values. In the feature selection stage, dual-correlation analysis using both Pearson and Spearman coefficients is employed to identify key features. Finally, using the NASA battery dataset, a gated recurrent unit (GRU) network is adopted to evaluate the performance of different feature combinations for SOH estimation. Simulation results demonstrate that the selected three-feature combination significantly improves the accuracy and stability of SOH estimation.

Abstract:Current methods for configuring adjustable capacity transformers have several limitations. These include that the consideration of influencing factors is limited, the integration of comprehensive cost analysis is insufficient, and the assessment based on distribution transformer area operation and control results is absent. To address these issues, a configuration evaluation method for adjustable capacity transformers is proposed, which integrates operational control and economic performance. First, a power flow model is built using second-order cone programming. Based on the optimized operation of the distribution area, typical daily net load data are obtained to support capacity selection of adjustable capacity transformers. Then, key factors affecting configuration decisions are analyzed, and a comprehensive cost evaluation index is developed. Case studies using historical load data are conducted under both photovoltaic (PV) and non-PV scenarios. The effectiveness and advantages of the proposed method are demonstrated by comparing different planning periods and capacity margins with traditional configurations.

Abstract:Under the background of a tight power balance, improving energy efficiency on the user side is a crucial to alleviate grid supply pressure. For energy management of household air conditioning systems under time-of-use pricing, a coordinated optimization model for air conditioning and battery energy storage systems is proposed. This model limits the frequent switching of the battery storage device between charging and discharging states within a rolling time-domain control framework. Based on the laboratory setup, a combined air conditioning and energy storage system for cooperative operation is constructed based on the Internet of Thing, where a distributed integration scheme for the energy management monitoring platform and other core components is provided. Through simulation and experiments, the effectiveness of the air conditioning-battery storage economic operation model and the energy management system is validated. The air conditioning system and energy storage devices exhibit obvious pre-cooling and pre-charging behaviors before peak electricity prices. Compared to the independent operation of the air conditioning system, joint economic operation reduces electricity costs by 13%.

Abstract:By aggregating massive distributed resources in distribution networks, virtual power plants (VPPs) are formed to create flexible portfolios used in electricity market regulation and grid dispatching, thereby addressing power supply-demand imbalances. However, in existing VPP energy management strategies, the effective consideration of distribution network operational security remains a key challenge. To achieve both efficient coordination of distributed resources within VPPs and secure operation of distribution networks, the distribution network security-region theory is adopted, whereby a dynamic distributed energy management strategy is constructed. A security-region model is first established based on the power-flow topology, and a decentralized scheduling framework is then designed for internal VPP resources using a consensus algorithm. By introducing a security-consensus variable defined by the security-region boundary function, economic and security requirements are jointly satisfied, enabling distributed energy management under network security constraints. Case studies show that the security of distribution network power flows is effectively ensured, and internal VPP resources are regulated rapidly, accurately, and dynamically.

Abstract:In order to tap the potential of low-carbon scheduling of electric vehicles with dual characteristics of source and load, an IES optimal scheduling method considering the carbon integral reward mechanism of electric vehicles is proposed. Firstly, a carbon reward mechanism model of electric vehicles considering dynamic carbon emission factors is constructed to realize the coordinated low-carbon operation of electric vehicles and demand response resources on the load side. Secondly, a source-side flexible response model considering the coupling of two-stage power-to-gas, hydrogen fuel cell, hydrogen storage tank and carbon capture is established to enrich the flexible application of hydrogen energy and reduce carbon emissions. Finally, with the goal of minimizing the total operating cost of the system, the IES day-ahead low-carbon scheduling model is constructed. The simulation results show that the proposed method can effectively improve the enthusiasm and economic of electric vehicles participating in system scheduling.

Abstract:Virtual power plants (VPPs), by aggregating and collaboratively controlling flexible demand-side resources, have become an important means to mitigate the supply-demand imbalance in modern power systems. Due to unpredictable factors, resources such as air conditioners and electric vehicles may exhibit power response deviations when executing day-ahead dispatch plans as a result of forecast errors. To address this, a real-time rolling optimization correction method for VPPs with multiple flexible resources is proposed. First, control models for adjustable resources—including electric vehicle clusters, air conditioning loads, and battery storage—are established. Combined with real-time updated weather conditions and electric vehicle connection information, this approach accurately quantifies the potential power deviation of the VPP during the execution of the day-ahead dispatch plan. Second, by introducing the price-quantity relationship curves of various flexible resources, the necessary price incentives for further tapping into their intra-day flexibility are analyzed. Finally, a real-time multi-resource coordinated optimization correction model is constructed to determine the optimal allocation strategy for power deviation correction in VPPs, thereby economically eliminating the power deviations caused by day-ahead forecast errors in real time. Case study results show that, compared with traditional correction strategies relying solely on battery storage as a backup resource, the proposed method is more economical and fully exploits the regulation potential of multiple flexible resources, enabling efficient and economic operation of VPPs.

Abstract:Temperature controlled loads (TCLs) are of significant importance in demand response (DR) research due to their remarkable thermal inertia and regulation potential. Addressing the limitations of existing models in fully capturing the time-varying characteristics and regulation capacity differences of TCLs, a data-driven parameter identification and demand response capability evaluation method is proposed. Based on the second-order equivalent thermal parameter (ETP) model, a linear identification equation is constructed, and a multi-layer recursive least squares algorithm with an adaptive forgetting factor is introduced to achieve online identification of the dynamic equivalent parameter matrix. Furthermore, tailored demand response capability evaluation strategies are designed for switching loads and continuously adjustable loads, considering their distinct regulation characteristics. Simulation results demonstrate that the proposed method can accurately identify load parameters and evaluate demand response capabilities, providing reliable support for TCL modeling and optimization of grid ancillary services.

Abstract:A short-term power load forecasting model based on the beluga whale optimization (BWO) algorithm and long short-term memory network (LSTM) is proposed to address the high demand for accuracy in power load forecasting in smart grids, while also considering the instability of traditional optimized LSTM models in short-term power load forecasting. The model first uses the hyperparameters of LSTM as the location of the beluga whale, historical load, day type, and weather factors as the dataset, and the prediction error of LSTM on the training set as the fitness of BWO. Then, BWO is used to optimize the parameters and hyperparameters of the LSTM model that is most suitable for the training set, and predict and analyze the power load data of a certain region based on the optimal LSTM model. The research results indicate that the average absolute error percentage and root mean square error of BWO-LSTM prediction results are smaller, the prediction accuracy is higher, and the prediction results are more stable. It can be used as a reliable tool for short-term power load forecasting and provide strong support for the safe and stable operation of the power system.

Abstract:With the rapid development of carbon and electricity markets, virtual power plants (VPPs) park have emerged as a novel energy management platform, playing a critical role in achieving low-carbon economic operations. A low-carbon economic optimization scheduling method is proposed for VPPs park under the uncertainties of electricity and carbon markets. A double-layer bidirectional long short-term memory (DBLSTM)-based model is developed to evaluate electricity and carbon price uncertainties. By constructing an electricity-carbon market-coupled optimization model, the study explores the classification modeling and differentiated control strategies of electric vehicle (EV) clusters, proposing a multi-energy system optimization scheme aimed at maximizing revenue. Simulation results demonstrate that the proposed method significantly enhances the system 's economic efficiency and carbon reduction capabilities, improves the flexibility of EV market participation, and achieves maximum comprehensive revenue under more realistic electricity-carbon market conditions, offering a novel approach for the efficient operation of VPPs.

Abstract:For power selling companies with controllable loads, the active participation in grid demand response is an important means to obtain excess profits. Based on the mechanism of demand response, the load-type power selling company is taken as the research object. It considers both the incentive-based and the price-based demand responses, and models of the response costs and benefits for the power selling company as well as residential and industrial users participating in demand response are established. Building on the costs incurred by users participating in demand response, a calculation method for the user cost allocation coefficient is proposed. Additionally, a dynamic compensation pricing strategy for the power selling company is introduced, which formulates different compensation prices based on user types to maximize the regulation potential. Finally, a two-layer dynamic pricing model for power selling companies based on demand response is established. Through case simulation, the feasibility and rationality of the model are verified.

Abstract:As electricity market reform deepens, the participation of end users in long term electricity transactions is intensifying, and their price perception and demand response behavior have become critical determinants of resource allocation efficiency. To address the limited capability of conventional bidding models to represent dynamic user behavior and to handle continuous decision spaces, a bilateral bidding strategy optimization framework is developed that integrates a multi period price perception mechanism with the deep deterministic policy gradient (DDPG) algorithm. A nonlinear price perception model with a Sigmoid structure is first established, in which cycle adjustment factors and stochastic perturbations are introduced to characterize response thresholds and behavioral heterogeneity under different price regimes. On this basis, a DDPG based policy learning architecture is constructed, and a differentiated reward function is designed to jointly account for profit maximization, social welfare improvement, and market fairness enhancement, thereby enabling adaptive optimization in a continuous action space. The proposed framework is evaluated through numerical simulation studies on representative market scenarios. The results indicate that market revenues can be increased, load modulation and temporal shifting can be promoted, and a favorable trade off among multiple operational objectives can be achieved. Moreover, user behavior is captured with higher fidelity, generator bidding strategies are more efficiently optimized, and the stability and social welfare of the electricity market are enhanced.

Abstract:Time series analysis is considered extremely important in the industrial field, anomalies in time series can be accurately identified to effectively improve production efficiency and reduce costs, and for electricity usage, operating costs of power companies are reduced and service quality is improved through anomaly detection in users' electricity consumption time series. With the rapid development of large language models, they are being applied to more fields, including time series prediction. A method for anomaly detection of users' electricity consumption time series based on large language models is proposed, communication with large language models is achieved via prompt engineering, the time series is preprocessed so that input data can be understood by the models and the anomaly detection task can be correctly executed, and experimental analysis on real user electricity consumption data shows that abnormal data can be effectively detected by the proposed method compared with existing models.

Abstract:To address the capacity allocation issue in frequency modulation auxiliary services provided by shared energy storage for wind farm clusters, a dynamic energy storage capacity allocation method integrating power generation prediction and optimal scheduling is proposed. First, a long short-term memory neural network (LSTM) is used to achieve high-precision prediction of the real-time power generation of wind farm, and then constructs a capacity optimization model with the goal of maximizing the overall revenue of shared energy storage and wind farm clusters. A dynamic balance constraint on the state of charge (SOC) of shared energy storage in wind farm clusters is introduced into the model, and a particle swarm optimization (PSO) algorithm is adopted to globally optimize the capacity allocation scheme, realizing the on-demand dynamic allocation of shared energy storage resources among wind farm clusters. Typical case studies verify that the method has significant advantages in improving the utilization efficiency of energy storage systems and maintaining the stability of the SOC of shared energy storage. Compared with the traditional equal distribution strategy, the proposed strategy can significantly enhance the response capability and economic benefits of wind farm cluster in frequency modulation services.