Abstract:With the transformation of the global energy structure and the development of intelligent technology, smart microgrids, as an important part of the new electricity system, have attracted widespread attention for their development trends and key issues. First, the development history and connotation evolution of microgrids in various countries are discussed, then the development trend of smart microgrids in China is analyzed, including the main development scenarios, evolution trends, characteristic trends, etc. Finally, the key issues that needed to be paid attention to in smart microgrids are pointed out, and corresponding countermeasures and suggestions are proposed.

Abstract:The proposal of dual carbon target has significantly accelerated the integration of green production and low-carbon development.Addressing the challenges of energy sharing and low-carbon operations in commercial parks, an electricity-carbon trading model tailored for charging stations, commercial buildings, and photovoltaic energy storage power stations is proposed. It highlights the complex energy and information flow dynamics among park entities, power trading centers, microgrid operators, and carbon trading centers. Distributed robust optimization is utilized to address the non-convex chance constraint problem in the electricity-carbon scheduling model for commercial parks, reformulating it as a semidefinite programming problem. The proposed method’s effectiveness in cost reduction and carbon emissions mitigation is validated using three distinct schemes. Finally, the proposed method is compared with the Conditional Value at Risk（CVaR）based linear programming approach. Results demonstrate that the distributed robust approach achieves superior alignment with actual scheduling scenarios in terms of decision-making accuracy.

Abstract:Compared with a single virtual power plant, multi-virtual power plants optimization has significant advantages in energy interaction. By connecting multiple virtual power plants, a broader energy network can be formed to achieve more efficient energy allocation and utilization, and improve the flexibility and reliability of energy interaction. Therefore, a chance constrained optimal scheduling model for multi-virtual power plants considering cooperation satisfaction is proposed. Firstly, P2P transactions between virtual power plants are considered, and a cooperative game model for multi-virtual power plants is constructed with the goal of minimizing total operating costs. Then,chance constrained optimal programming method is used to address the impact of photovoltaic output uncertainty on the optimization results of virtual power plants, avoiding the risk of load loss caused by uncertainty in photovoltaic output. Then, a generalized Nash bargaining method based on cooperation satisfaction is proposed to achieve a reasonable allocation of cooperation surplus. Finally, an example is used to verify that this method can provide a multi-virtual power plant cooperative trading strategy that takes into account both economy and reliability, and improves the enthusiasm for cooperation among virtual power plants.

Abstract:To address the problem of optimal configuration of user-side energy storage under the latest tariff policy in the third regulatory cycle, a user side energy storage optimization configuration method considering multiple tariff modes is proposed for general industrial and commercial users with transformer capacity ranging from 100 kVA to 315 kVA and free choice of tariff modes. Firstly, the latest tariff policies analyzed and interpreted are as the basis for subsequent model construction. Secondly, optimization configuration models for energy storage are established under 3 modes：two-part tariff based on demand, two-part tariff based on capacity, and single tariff. The most economical energy storage configuration scheme is selected through the comparison of the three modes. Finally, the effectiveness of the proposed method is analyzed and verified through numerical simulation. The proposed method is also applicable to large industrial users and other general industrial and commercial users, which could be seen as a simplification of the proposed method.

Abstract:Multi-energy conversion and power demand response are effective ways to improve energy utilization, reduce system energy supply pressure, and balance supply and demand. Power demand response technology considers multi-energy conversion, aiming at the mutual conversion and substitution of multiple energy sources in the system, combined with price-based power demand response, to improve the energy supply capacity of the system, stabilize the load side fluctuation, and promote the system to eliminate wind and solar energy. First, linearize the modeling of related equipment in energy input, multi-energy conversion, multi-energy storage and other links a unified model for multi- energy conversion of electrical loads is proposed, and the resulting electrical energy storage characteristics are analyzed. Second,based on the peak-valley time-of-use electricity price, establish three load models that can be reduced, transferable, and replaceable. Finally, considering the four constraints of system operation, taking the lowest comprehensive cost of system operation as the objective function,the optimization effect of multi-energy conversion power demand response on system operation is studied, modeled by Yalmip modeling language, and Cplex solver is called to solve. Based on the optimal solution result of the objective function, the validity and feasibility of the model in this thesis are verified, which can further reduce the operating cost of the system and maximize the ability of the power system to absorb wind power.

Abstract:The integrated energy system of buildings（BIES）has a large load capacity and various types of energy. How to coordinate the operation of different types of energy sources and loads to realize the peak shaving and valley filling of the power grid is a research hotspot.Considering the factors such as light storage and soft environment comfort, BIES is optimized. First, a photovoltaic（PV）output model based on the uncertainty of solar radiation intensity distribution is constructed by using the Beta distribution method; second, considering the mobility characteristics, an electric vehicle（EV）storage model is proposed and a quantitative model of building environmental comfort that takes into account the influence of temperature on flexible loads are constructed; finally, based on the mixed-integer linear programming method to solve the optimal scheduling model of the integrated energy system in buildings is proposed. The correctness and validity of the proposed model are verified by analyzing the load optimization effect, cost, and related influencing factors of buildings in a certain region under different scenarios.

Abstract:In order to solve the insufficient heating capacity of heating stations caused by heat source shortage, and to avoid the high cost of a primary heating network transformation, an electric heat pump heating scheme is proposed by taking a heating station of a residential community as an example. Based on the TRNSYS, a simulation model of the electric heat pump heating system is established. The bi-level optimization model is used to optimize the capacity design and operation scheme of the main equipment with the goal of optimal life cycle economy. Finally, the marketing potentiality of the system is evaluated from the perspectives of economy and impact on the power grid. The results show that the present value of the total investment in the life cycle of the electric heat pump heating system is positive, indicating that the system has good economy. In addition, the seasonal imbalance coefficient of the power grid increased by 13.4% after the adoption of this system, which increased the power utilization rate.

Abstract:For smart building with photovoltaic installations, most of the time it is in the situation of unbalanced supply and demand, and there is the problem of large dependence on the power grid, such as“surplus power on the Grid, lack of power to buy”. In order to enhance the rate of local consumption of new energy and reduce the cost of power consumption for users. A power sharing scheduling strategy based on the optimal aggregation mode of building users is proposed, taking into account the complementary behavior of different users. First, An aggregation model based on the landscape theory energy function driven by an non- equilibrium cobweb pricing model is established to seek the aggregation combination with the best effect of power sharing among users, and the package price of power sharing within the combination is set by the matching degree of supply and demand. Then taking into account the uncertainty of generators‘output and users’behavior, the light robust optimization theory is used to solve the scheduling strategy of power sharing within the resulting aggregation portfolio, and the effectiveness of the proposed model is verified through example simulations.

Abstract:In order to popularize the load identification technology of smart meters and solve the problem of low identification accuracy of traditional non-intrusive load identification algorithm on unbalanced sampled data, a non-intrusive load identification method based on adaptive synthetic（ADASYN）and image analysis is proposed. 1D power data is converted into 2D MTF feature images by markov transition field（MTF）coding, which is used as the input of image recognition network. Based on the deep information mining capability of dense connectivity network（DenseNet）, 2D images are input into DenseNet121 network to extract feature information and realize the identification of load types. Based on ADASYN algorithm, the unbalanced data set is oversampled to eliminate the model learning bias caused by the unbalanced data distribution. The results show that ADASYN algorithm can solve the non-intrusive load monitoring data imbalance problem well, and its identification accuracy and F1 score are increased by 0.247 and 0.267, respectively. At the same time, MTF images have clear and easily distinguishable feature information. Combined with the powerful deep feature capture capability of DenseNet121 network, the identification accuracy and F1 score can both reach 0.952, which effectively improves the identification accuracy of non-intrusive load types on unbalanced sampled data.

Abstract:Comparison and impact analysis of different risk dispatch strategies of new power system can help dispatchers get more detailed operation information, so as to promote their practical skills when making decisions about risk dispatch strategies. First, Copula theory is used to build the probabilistic distribution of forecast error given a specified forecast value, and multiple scenarios can be obtained by Monte-Carlo sampling. Second, three different risk dispatch strategies are built and compared, which are two-stage stochastic optimization modelling, chance-constrained approach and deterministic models, and then derives and analyses day-ahead electricity prices based on Lagrange multiplier method. Last, the example analysis shows that the results of other two models can be covered by setting different confidence levels in chance-constrained approach. The two-stage stochastic optimization model is rather conservative, while the deterministic model is rather aggressive. It also shows that it is prone to curtail new energy power during low-load periods in a high proportion new energy power system.

Abstract:Accurate prediction of regional electricity sales is crucial for the power sector’s effective energy management and planning. Existing forecasting models largely rely on historical electricity sales data and partially incorporate temperature effects, yet they inadequately consider a broad range of meteorological factors. In response, it introduces a novel forecasting method combining multi- head attention mechanisms with long short-term memory networks（MHAM-LSTM）for regional electricity sales. Initially, key variables are identified and redundant variables are eliminated through correlation analysis. Subsequently, the multi-head attention mechanism is used to focus on the key indicators that have an important impact on electricity sales. Finally, the LSTM network delves into the latent patterns of time-series data to forecast regional electricity sales. Experimental results show that the MHAM-LSTM model surpasses comparative models, including random forest, deep neural networks, long short-term memory networks, temporal convolutional networks, and transformer, in electricity sales forecasting accuracy, demonstrating significant performance advantages. Additionally, the analysis of meteorological factor importance reveals that incorporating multiple meteorological variables, particularly temperature, wind speed, and humidity, plays a crucial role in improving prediction accuracy.

Abstract:By aggregating the flexible resources on the demand side, the virtual power plant can effectively stabilize the power system fluctuations when the new energy generation is high. In addition, considering the scale and accuracy of new energy power generation, the consumption in the zone can no longer meet the requirements of new energy development. Moreover, modes such as centralized clearing cannot adapt to the characteristics of new energy that can be used immediately. Based on this, on the basis of the existing inter-provincial trading mechanism, the form of demand-side virtual power plant is introduced as a new type of market user, and the transaction strategy of virtual power plant participating in the intra-day listing of inter-provincial green power is proposed, and the effectiveness of the proposed strategy and model is verified by case analysis, which is helpful to realize the mutual aid of cross-regional resources.

Abstract:Virtual power plants（VPPs）can aggregate distributed energy resources to achieve economies of scale, but when different VPPs belong to different stakeholders, traditional centralized scheduling mechanisms may not be applicable. A master-slave game coordination strategy for multi-VPPs considering carbon trading is proposed for city-level distribution networks with multiple VPPs. Firstly, a carbon trading cost model is established, based on which a VPP optimization scheduling model incorporating the carbon trading mechanism is formulated. Secondly, a multi-VPP master-slave game model led by the VPP operation platform is constructed and solved using an improved particle swarm optimization algorithm. Finally, a simulation test system is set up to validate the effectiveness of the proposed method. Simulation results show that the proposed multi-VPP master-slave game strategy can effectively reduce operating costs, achieving coordinated optimal operation of multiple VPPs, and the introduction of carbon trading mechanism can bring additional benefits to virtual power plants and promote efficient consumption of renewable energy.

Abstract:A novel model based on adaptive recurrent neural network（RNN）is proposed to address the issues of low efficiency and poor performance in identifying abnormal electricity consumption behavior among residents. Design a SMOTE-ENN resampling method to increase the classification performance of imbalanced datasets. We have established an adaptive RNN detection model, using batch normalized RNN as the basic learner, and combining hyperparameter optimization and buffer to dynamically adjust the BNRNN model. In the experimental stage, after improved SMOTE-ENN resampling, the classification performance of the model was significantly improved. At the same time, experiments have verified that the proposed adaptive RNN model with buffering and hyperparameter optimization has the lowest MAE error, indicating that the proposed model has excellent generalization ability. The experimental results validate the practicality and excellent performance of the proposed model, which can provide some reference for the development of abnormal electricity consumption behavior detection.

Abstract:In order to solve the problem of serious consequences caused by untimely processing of 95598 emergency work orders, a parallel hybrid neural network sentiment analysis model for power work order texts based on attention mechanism is proposed to meet the requirements of automatic intelligent reminders for emergency work orders. Firstly, a sentiment analysis dataset for electric power work order texts is generated by manually annotating the emotional urgency of work order texts through sentiment dictionaries and rule sets；Using text pre trained BERT model for text vectorization；Then, use text convolutional neural network（TextCNN）and bidirectional long short term memory（BiLSTM）to extract local and contextual features of the text, respectively, and perform feature fusion；Using attention mechanism to enhance the ability of the model to recognize key information in the fused text features. The calculation results show that the neural net?work model that integrates attention mechanism has better performance in sentiment classification of power work order texts compared toother deep learning models.

Abstract:With the extensive integration of distributed power sources and power electronic devices into distribution networks, new characteristics are manifesting in aspects of energy supply and load demand. A voltage sag source identification method combining complete ensemble empirical mode decomposition with adaptive noise（CEEMDAN）and improved hunter-prey optimizer cost sensitive support vector machine（IHPO-CSSVM）is proposed to address the difficulties in selecting hyperparameters for support vector machine（SVM）and the imbalance of voltage sag source signal data categories. By simulating circuits on the Matlab/Simulink simulation platform, different types of voltage sag sources are obtained. The CEEMDAN is used to extract the feature vectors of the three-phase voltage of the voltage sag source signal, and its approximate entropy is calculated. A new feature vector is constructed and input into the IHPO-CSSVM classifier for training. Compared with SVM, CSSVMand extreme learning machine, simulation results show that IHPO-CSSVM has the highest recognition accuracy. This method can accurately extract useful features from complex voltage signals and improve recognition accuracy by optimizing model parameters, providing an effective solution for voltage sag problems in power systems.

Abstract:Power generation enterprises are one of the important sources of carbon emission, and the refinement of carbon emission forecasts on the power generation side is of positive significance to the formulation of China’s carbon emission policy. In this context, a carbon emission forecast model based on variational modal decomposition（VMD）and temporal convolutional network（TCN）is proposed for the characteristics of irregularity, nonlinearity and temporal sequence of carbon emission on the power generation side. First, VMD is used to smooth the preprocessing of the carbon emission time series data, splitting the raw carbon emission data into several modal components to reduce irregularities and nonlinearities in the data series. Second, considering the performance degradation of existing machine learning algorithms during the network training process, each modal component is predicted separately based on TCN to maximize efficiency in the use of carbon emission time seriesdata. Finally, the forecast results are reconstructed to obtain the final forecast values of carbon emissions. The results show that compared with the traditional four forecast models, the method effectively improves the effectiveness and accuracy of the forecast model by innovatively combining VMD model and TCN.

Abstract:In order to meet the challenges of climate change and energy structure transformation, European and American governments have formulated a series of energy policies and power system development strategies to reduce greenhouse gas emissions, promote the sustainable development of energy systems and improve their international status in climate governance. In the context of domestic double carbon’goal and the construction of a new power system, the accumulated experience of energy transition in developed countries in Europe and America has a certain reference value for China’s energy strategy. The latest energy policies and power system development strategies of the European Union and the United States are sorted out, the similarities and differences of domestic and foreign power system construction are deeply analyzed in terms of policy guidance, technological innovation and market construction, and refine the elements in their strategic development plans that are worthy of reference for the construction of new power systems in China, so as to provide reference for the improvement of China’s energy policies and power system development routes.