Abstract:Aiming at the accommodation challenges of high-proportion new energy integration into power systems, an intelligent regulation method for new energy accommodation resources under uncertainty is proposed. A joint probability model of wind and photovoltaic output is constructed based on Monte Carlo simulation and Copula theory to quantify new energy uncertainty. A refined model of multiple adjustable resources such as energy storage and electric vehicle clusters is established, and their operational characteristics and dispatchable potential are analyzed. A three-layer optimization architecture is designed:in the day-ahead layer, two-stage stochastic programming is used to formulate the dispatch plan;in the intra-day layer, rolling correction is conducted based on model predictive control;and in the real-time layer, power balance is achieved through distributed feedback. Simulation results show that the proposed strategy significantly improves new energy accommodation, with wind power accommodation reaching 96.8% and photovoltaic accommodation reaching 94.2%, increased by 14.5 and 17.7 percentage points respectively;meanwhile, total operation costs are reduced by 8.5% compared to traditional strategies. The effectiveness of the proposed method in enhancing accommodation rates, improving operational economy, and increasing dispatch flexibility is verified, providing theoretical basis and technical support for optimal operation of high-proportion new energy power systems.

Abstract:To address the high transportation costs and severe carbon emissions caused by dam blockages at hydropower stations, a cross-dam energy-logistics coupled dispatching optimization method that integrates electricity-hydrogen-oil multi-energy complementarity is proposed. According to the flow diversion characteristics of cross-dam cargo between waterway and land routes, the operational characteristics of cross-dam logistics covering heterogeneous heavy-duty trucks and waterway handling facilities are analyzed, and energy consumption models for transportation equipment are established. Based on the operational power characteristics of facilities such as integrated hydrogen production-charging-storage stations, energy facility production models are constructed. Combined with the characterization of energy supply-demand coupling mechanisms, a cross-dam energy-logistics coupled dispatching optimization model is established. With the objective of minimizing operational costs while considering carbon emissions and load response capability, coupled optimization is performed on cross-dam energy and logistics systems to obtain cargo flow distribution and dispatching strategies for transportation and energy equipment. Taking a 6 400 MW hydropower station in Southwest China as a case study, simulation results demonstrate that the proposed method can effectively reduce cross-dam operational costs and carbon emissions.

Abstract:Aiming at the low carbon economy operation problem of integrated energy system, an optimization method of combined heat and power system considering augmented carbon emission flow and integrated demand response is proposed. Firstly, an augmented carbon emission flow model with carbon capture and heat storage devices is established. The carbon emission responsibility of load side is analyzed by combining carbon emission flow theory and Shapley value method, and its dynamic carbon emission factors are calculated. Secondly, a comprehensive demand response model based on load side thermal inertia is constructed. On this basis, a two-layer scheduling model of CHP system considering carbon trading mechanism is proposed. The upper layer is the power system model, and the lower layer is the energy hub model. The low-carbon economic scheduling is carried out with the minimum total system cost as the objective function, and the alternating direction multiplier method is used to solve the problem. Finally, the IEEE14-node power system is used for example analysis, and the simulation results verify the effectiveness of the proposed model.

Abstract:In order to achieve the low carbon operation and economic benefits of integrated energy system (IES) in electrolytic aluminum, a low carbon economy operation strategy considering the demand response (DR) and equipment capacity optimization is proposed. First, an IES topology integrating electricity and heat energies is established, and a DR model is also established based on the transferability of electricity and heat loads. Then, the IES carbon emission accounting model is established, and a ladder-type carbon trading mechanism is introduced to implement the restriction of carbon emission. Finally, the optimization operation model is transformed in the form of mixed integer linear problem, which considers the comprehensive costs of energy purchase, equipment operation and maintenance, carbon trading, equipment investment and wind and light abandonment, and is solved by CPLEX. Experimental results show that when DR is considered, the total operation cost and carbon emissions are reduced by 3.07% and 1.17% respectively. In addition, after further considering the optimization of equipment capacity, the total operating cost and carbon emission can be additionally reduced by 0.39% and 0.64% respectively, while the production of electrolytic aluminum is ensured. As discussed above, the method proposed can improve the economy and environmental performance of IES operation to a certain extent.

Abstract:A bi-level Stackelberg game model is established for coordinated trading among multiple virtual power plants and the dispatch of aggregated electric vehicles. At the upper level, trading prices are optimized to maximize the net profit of the virtual power plant operator. At the lower level, the operating cost of each virtual power plant is minimized by coordinating micro-turbines, energy storage, wind power, interruptible loads and aggregated electric vehicles. Electric vehicles are modeled only as time-varying connected resources in VPP4. Charging and discharging uncertainty is addressed by robust optimization, and the bi-level model is solved by Bayesian optimization. Case studies show that electric vehicle aggregation improves local flexibility, dynamic pricing promotes inter-VPP trading, and robust optimization enhances dispatch stability.

Abstract:A day-ahead dispatch model for microgrids under source-load uncertainty conditions is established to enhance their economical effectiveness. In view of the source-load uncertainty composed of photovoltaic power and demand, a novel adaptive optimization decision-based distributionally robust optimization is proposed to overcome the conservativeness of traditional robust optimization. Firstly, a conservativeness-adjustable ambiguous set of source-load uncertainty is constructed, which can flexibly adjust the conservatism through distribution constraints. Meanwhile, auxiliary uncertainty variables are introduced to transform the ambiguous set into a manageable form. Secondly, leveraging the duality principle of infinite programming and fixing the dual variables, the distribution uncertainty is reduced to scenario uncertainty. Finally, employing an extreme dual variable generation method, the proposed model is converted into a deterministic programming in a two-stage solving framework. Numerical simulations verify the superiority of the proposed model in balancing the economical effectiveness and robustness of microgrid operations.

Abstract:To fully exploit the dispatchable potential of electric-vehicle (EV) clusters, a real-time EV charging and discharging scheduling strategy is proposed that takes into account users' willingness to participate in vehicle-to-grid (V2G) discharging. First, considering the battery degradation incurred during participation in scheduling, an additional economic subsidy is provided, and based on consumer psychology, a response-rate function is introduced to quantify the uncertainty in how users' discharge willingness varies with the subsidy level. Second, using the grid dispatch center's real-time load forecasts together with the online status of EVs, charging and discharging power commands are rapidly re-optimized in each scheduling cycle via a sliding-window mechanism to achieve adaptive real-time control. Finally, case-study simulations verify that the proposed strategy can respond in real time to random EV-cluster access and grid load fluctuations while achieving dual optimization on both the distribution-grid side and the user side.

Abstract:Load identification is one of the key technologies in power system planning, operation, and management, playing a crucial role in the efficient scheduling and stable operation of smart grids. Traditional load identification methods typically rely on the closed-set assumption. However, in practical applications, the presence of unknown appliances makes it difficult for algorithms based on this assumption to achieve accurate recognition. To address this issue, an open-set load identification algorithm, OpenAppliance, based on threshold adjustment is proposed. The proposed algorithm integrates deep learning and probabilistic models, calibrating the neural network outputs to enhance the detection capability for unknown categories while maintaining recognition accuracy for known categories. First, load data is transformed into an image format suitable for deep learning, and a CNN-based load identification model is constructed. Then, the OpenAppliance algorithm is applied for post-processing to adjust classification thresholds and optimize recognition results. Finally, the method is validated on the BLUED load dataset and compared with existing load identification algorithms. Experimental results demonstrate that the OpenAppliance algorithm enhances the generalization ability of load identification and significantly improves the accuracy and robustness of the load identification system.

Abstract:As a crucial platform for integrating distributed renewable energy, the efficient and optimal scheduling of microgrid clusters is key to enhancing the economic efficiency and reliability of regional power supply. Addressing the high-dimensional, non-convex, and multi-constraint nature of the scheduling model, alongside the tendency of traditional optimization algorithms to trap in local optima, a collaborative optimization scheduling method for microgrid clusters based on an improved simulated annealing dragonfly algorithm (SADA) is proposed. First, a scheduling optimization model is established with the objective of minimizing the combined operational and environmental costs. Then, the logistic chaos mapping is employed to enhance initial population diversity, the nonlinear adaptive inertia weights are introduced to balance global and local search capabilities, and a simulated annealing mechanism is embedded to improve the ability to escape local optima by leveraging its probabilistic jump property. Finally, compared with other benchmark algorithms, the proposed SADA achieves reductions in total cost by 7.33%, 5.09%, and 4.24%, respectively, effectively validating its effectiveness and superiority in complex energy scheduling scenarios.

Abstract:Aiming at the problems of new energy curtailment, high carbon emissions and economic benefits in the distribution network with high wind and solar penetration rate, considering the joint operation of hydrogen energy utilization, fossil energy unit fuel cleaning and carbon trading mechanism, an optimal scheduling method of active distribution network considering hydrogen energy utilization, unit co-combustion and stepped carbon trading is proposed. Firstly, the mechanism analysis and model construction of hydrogen energy utilization unit composed of electrolytic cell, hydrogen storage tank, hydrogen fuel cell, methane reactor and ammonia production device are carried out, and the scheduling strategy of active distribution network considering multi-energy flow is analyzed. Then, considering the participation of ammonia and hydrogen in the combustion of fossil energy units, a mixed combustion model of fossil energy units is constructed, and then a carbon trading cost model under the stepped carbon trading mechanism is constructed. Finally, a low-carbon economic dispatch model of active distribution network is established with the goal of minimizing the sum of operating costs such as coal consumption, gas purchase and wind power curtailment. By setting up multiple operation schemes, the effectiveness of the proposed optimal scheduling method in improving the operation economy and energy cleanliness of the distribution network is analyzed and verified.

Abstract:Load forecasting is an important aspect of ensuring power supply-demand balance in distribution station and has significant guiding significance for the safety warning and stable operation of the power system. Affected by various other factors, the direct prediction model of the load in the distribution area usually has poor generalization ability and is difficult to meet the load forecasting requirements of complex distribution substations. To improve the generalization ability of load forecasting methods in distribution areas, a load prediction method based on multi-source environmental feature fusion is proposed. Firstly, the external environmental features are analyzed by principal component analysis, and the input variables are reduced and corrected to extract the environmental feature components that affect the load changes in distribution area. Based on the proposed load identification model, the substation load is identified by feature fusion. Then, based on the comprehensive analysis of the identified substation load and environmental feature components, the proposed load forecasting model is used to predict the substation's individual load by feature fusion. Finally, the linear superposition of the predicted results of individual loads is obtained, and the load prediction result of the distribution substation is obtained. Selecting the load data of a low-voltage distribution substation in China for two years as an example, compared with other direct prediction methods, the proposed method's load prediction curve is closer to the true load data curve, effectively improving the accuracy of load forecasting.

Abstract:Addressing the issues of insufficient accuracy in physical models and lack of physical constraints in data-driven models for real-time wind power estimation, a hybrid estimation method that integrates physical modeling with machine learning is proposed. A three-layer architecture comprising physical modeling, machine learning residual correction, and physics-informed regularization is constructed to unify estimation accuracy and physical credibility. A physics-informed regularization strategy based on sample weight adjustment is designed, which defines three physical constraints including monotonicity, density consistency, and power coefficient. Through iterative optimization, physical constraint embedding is achieved without modifying the kernel of ensemble learning algorithms. Monte Carlo simulation is employed to process power expectation values under turbulence effects, and combined with physical factors such as air density correction, yaw efficiency, and wake effects, a baseline physical model is established. A multi-level feature system is constructed, utilizing random forest and gradient boosting trees ensemble learning framework to capture systematic biases. Experimental results demonstrate that the proposed method achieves significant improvements in estimation accuracy compared with both pure physical models and pure ensemble learning model. Meanwhile, estimation results that violate physical laws are effectively suppressed, providing a new technical approach for real-time wind power estimation.

Abstract:To achieve green and low-carbon electricity consumption and fully activate the user-side carbon emission potential as well as flexibility resources, an electricity-carbon collaborative trading model for multi-prosumers is constructed. The model aims to promote the utilization of distributed renewable energy by sharing energy storage capacity, thereby reduce dependence on the upper-level distribution network and fossil fuel consumption, and indirectly cutting carbon emissions. In terms of methodology, a shared operation mechanism between multi-prosumers and centralized energy storage is proposed, and a corresponding trading model is established. Its rationality and effectiveness are verified through case studies. The results show that integrating the user side into the carbon market can significantly improve the overall revenue of the electricity-carbon coupled market. Under the shared energy storage operation mode, the proportion of carbon cost drops sharply from 63.6% in the traditional mode to 0.09%. Further analysis indicates that when users' free carbon emission allowance exceeds 60%, the system can operate in a mode dominated by electricity revenue with carbon cost approaching zero. The result provides a feasible business model reference for user-side participation in the carbon market under the carbon neutrality goal.

Abstract:In recent years, the scale of renewable energy is characterized by rapid growth, while issues such as the disorderly development of distributed PV and decentralized grid integration are gradually revealed, posing significant challenges to the safe operation and sustainable development of distribution networks. A planning model for rooftop distributed PV site selection and capacity determination that accounts for spatiotemporal differences is proposed, based on the spatiotemporal variability in rooftop distributed PV output and regional economic indicators of PV systems. First, considering the characteristics of spatiotemporal uncertainty of multiple sources and loads in new smart distribution networks, a PV-load spatiotemporal difference model is constructed. Next, the model is solved using a multi-objective sparrow search algorithm, taking into account the economic benefits on the user side, node voltage deviations, and system network losses. Finally, simulation verification is conducted using the IEEE 33-node distribution system as an example. It is demonstrated that the proposed model has practical significance for improving the economic value of rooftop distributed PV and the hosting capacity of distribution networks, while also providing valuable insights for the large-scale development of new smart distribution networks.

Abstract:With the increasing penetration of renewable energy, the diversification of load types, and the advancement of power electronics technology, traditional AC distribution networks are gradually evolving into hybrid AC/DC distribution networks. To address the challenge of power flow convergence in complex hybrid networks with multiple converters-particularly the instability arising from improper initial value selection-a unified power flow calculation method for hybrid AC/DC distribution networks incorporating incomplete LU decomposition preconditioning is proposed. Initially, the voltage and power equations for both the AC and DC sides of the voltage source converter (VSC) are established. Subsequently, power balance equations are formulated, taking into account various node types and the active/reactive power control strategies of the VSC. Jacobian matrix is then preconditioned using inoomplete LU decomposition, and the power flow of the hybrid AC/DC distribution network is solved via the preconditioned biconjugate gradient stabilized (BiCGSTAB) method. Finally, the effectiveness of the proposed preconditioning approach is verified through a comparison of non-zero element fill-ins and the number of iterations required for convergence. Furthermore, the robustness of the algorithm is validated by comparing its convergence behavior against alternative numerical methods.

Abstract:Under the dual-carbon goal and the ongoing development of a new-type power system, greater flexibility challenges are imposed on power system operation by the large-scale integration of electric vehicles (EVs), and the adjustable potential of EV resources therefore needs to be further exploited. To address the coordinated optimization problem of charging point operators (CPO) participating in coupled energy and frequency regulation markets, a bilevel scheduling model is established. In the upper level, CPO profit maximization is pursued, with electricity trading revenue, frequency regulation capacity compensation, mileage compensation, and operating cost all being considered. In the lower level, total system operating cost minimization is achieved through the joint clearing of the energy market and the frequency regulation market. The bilevel model is then transformed into a mixed-integer linear programming problem by introducing karush-kuhn-tucker conditions and linearization techniques, and case studies are conducted on the IEEE 30-bus system. The results show that only a limited proportion of bids is cleared in the energy market, whereas more than 70% of the regulation tasks are undertaken by the CPO in the frequency regulation market. As a result, operating costs are effectively covered and positive net benefits are obtained. Effective coordination between resource allocation and price formation in the coupled markets is also achieved by the proposed strategy. Both the economic performance of the CPO and the flexibility support capability of the power system are thereby improved. A feasible theoretical basis and methodological support are thus provided for the cross-market optimal operation of aggregated EV resources.

Abstract:The rapid charging demand of electric vehicles is required to be satisfied by supercharging stations. However, owing to the high-power and concentrated nature of charging, significant load peaks are liable to be generated over short-time intervals, whereby operational stability is adversely affected. Under time-of-use electricity pricing and grid capacity constraints, an intelligent scheduling approach integrating a hybrid genetic algorithm (HGA) with reinforcement learning (RL), denoted as HGA-RL, is developed, within which the supercharging station operator is defined as the decision-making entity. The charging sequence of users is globally optimised by means of HGA, whereby a rational basis for power allocation is established. Subsequently, charging power and scheduling intervals are dynamically regulated through RL, such that load shifting is achieved, with charging demand being increased during low-price periods and reduced during high-price periods.It is demonstrated by simulation results that charging load peaks are effectively mitigated and that the overall load profile is significantly smoothed. Meanwhile, the electricity procurement cost is reduced whilst user satisfaction is maintained, and the coordination efficiency between the supercharging station and the power grid is consequently improved.

Abstract:With the integration of distributed energy resources, the load characteristics of power grids have changed. In particular, under conditions where renewable energy sources such as photovoltaic and wind power exhibit large fluctuations, the grid may experience light-load conditions. To address the misjudgment of wiring errors in electric energy meters under scenarios of light load and reactive power over-compensation caused by a high penetration of distributed energy resources, an intelligent identification method based on XGBoost is proposed. By analyzing features such as current and power factor in light-load and reactive power overcompensation scenarios, and by combining actual collected data with generated data to build a classification model, the method effectively distinguishes normal operating states from wiring errors. In validation tests, the model achieved an accuracy of over 98% in detecting wiring errors, as demonstrated by multiple evaluation metrics, including accuracy, precision, recall, and F1-score, significantly reducing the false judgment rate. The results show that this method can effectively handle complex power load scenarios caused by the high penetration of distributed energy resources, providing an efficient and reliable intelligent solution for identifying wiring errors in power systems.