Abstract:Nowadays, the degree of standardization of the power load management system needs to be further strengthened.The relevant domestic standards are mainly distributed in two business branches including power load management and power consumption information collection. The content of each standard doesn’t match the current information and communication technology, automation control technology, and the development of big data analysis technology. The standard functions and performance indicators lack advanced nature, and many key contents are difficult to meet the needs of accurate classification, real-time monitoring and real-time control of current loads. Firstly, the domestic standard architecture of the load management system is expounded, including four aspects：system class, terminal class, interface class and management class. Then, the current status of the load management system standard system is classified, sorted out and interpreted, and typical standards are selected for each aspect to introduce,and the development status of each link and the problems existing in practical application are discussed. Finally, combined with the new requirements of the power grid development for power load management system and new problems in actual operation, a proposal for improving the standard system of the new power load management system and a key action plan are put forward, which have certain reference value for revision of relevant standards of the new power load management system in the future.

Abstract:With the increasing proportion of new energy generation, the stochasticity of source and load makes the pressure on the balance of supply and demand in the power grid continuously increase and the cost persistently rise, so the new power load management system relying on the exploitation and utilization of demand- side flexible regulation resources is important, which can promote the power grid regulation mode from“source follows load”to“source, network, load and storage interaction”transformation.Significance, construction objectives and main contents of new power load management system are analyzed. The way of flexible regulation of load-side resources is proposed, which corresponds to station- line- transformer to achieve hierarchical zoning grouping,explains the hardware requirements of smart energy unit, software architecture and research status of multi-functional load management switch. Finally, combined with the actual application scenarios and adaptability, the professional capacity building requirements, support policy needs and hardware and software research direction of new load management system are proposed, which provide reference for the flexible regulation of load-side resources to participate in grid-friendly interaction.

Abstract:In the context of the development of new power systems, the continuous increase in the proportion of intermittent new energy sources puts forward higher requirements for the flexibility of the combined electric heating system. A multi-stage robust dispatch method for electric and heating combined systems that takes into account temperature control load demand response is proposed. Firstly, a model of electric- heat combined system that considers flexible resources such as electric-heat conversion equipment and temperature control load is established, and the concept of virtual energy storage is proposed to model the temperature control load. Then, the wind power output forecast error is modeled as a box uncertainty set, and a multi-stage robust scheduling model of the combined electric and heating system is proposed. In each scheduling period, the observation value of wind power can be used to minimize the total current and subsequent moments. The maximum cost is the goal, the subsequent scheduling plan is adjusted, and the model is solved by a robust dual dynamic programming algorithm. Finally, test cases are used to verify the effectiveness of the proposed method.

Abstract:It is essential for modern power system to forecast electricity price and load accurately. However, due to the strong correlation between the electricity price and the load, if the mutual influence is not taken into account, the accuracy of the forecast will be reduced. In order to improve the prediction accuracy of existing methods, price and load relationship are considered and a deep recurrent neural networks model is proposed for price and load forecasting, that is sparse autoencoder nonlinear autoregressive network with exogenous inputs comprising of feature engineering and forecasting. Firstly, an efficient sparse autoencoder is proposed to improve the effectiveness of feature extraction by improving the original method. Secondly, the nonlinear autoregressive network is used to forecast the load and price. The ISONE and PJM big datas of power market are simulated and verified. Compared with cascaded Elman networks, sparse autoencoder nonlinear autoregressive network reduces the mean absolute error by 16% in load forecasting and 7% in price forecasting.

Abstract:The large- scale grid connection of new energy brings huge challenges to the peak load regulation of power systems.The peak load regulation space of the traditional power supply side is gradually exhausted, an efficient market mechanism is urgently needed to guide the flexible loads on the user side to actively participate in the peak load regulation market to maintain the security and stability of the power grid. Combined with the actual situation of users, considering user satisfaction and load demand, the market mechanism of flexible load aggregation including electric vehicles,electric heating, and energy storage participating in day- ahead peak load regulation was proposed, and the bi- level optimization model of aggregator participating in day-ahead peak shaving regulation market clearing and flexible load scheduling on the user side was established. The upper-level model considers the influence of the reliability of load aggregators and conducts day- ahead market clearing with the objective of the lowest dispatching cost. The lower·level model considers user satisfaction and calls user-side resources in the form of signing a contract. The simulation results verify the feasibility of combining different loads for joint dispatch, indicating that the strategy proposed can significantly reduce the cost of dispatch, ensure the reliability of peak shaving in the market, and ensure the safe and reliable operation of the power grid.

Abstract:The peakvalley difference is increasing in the power system, which requires more capacity to keep the power balance. As a significant demand response resource, thermostatically controlled loads can provide peak shaving capacity.In this context,an energy efficiency ratio considering consumer satisfaction and a peak shaving strategy are proposed, which helps the system operators regulate the thermostatically controlled loads based on energy efficiency management. Firstly, based on the fuzzy set method, a model is proposed to calculate the energy efficiency ratio considering consumer satisfaction of thermostatically controlled loads. This model helps the system operators to estimate the energy efficiency rate of mass thermostatically controlled loads without installing hardware. Then, a fitting model is built to estimate the peak shaving capacity. Based on the above models, the peak shaving strategy is developed. Before the peak shaving, the system operatorscheck whether the thermostatically controlled loads can participate in the peak shaving and calculate energy efficiency ratio considering consumer satisfaction and capacity. The system operators may regulate-thermostatically controlled loads whose energy efficiency ratio considering consumer satisfaction is low. Therefore, the satisfaction of society and the comprehensive energy efficiency will improve when providing the peak shaving capacity. The case uses realistic data,which contains consumers’electricity consumption data in an eastern city in China and its temperature data.The results indicate that the comprehensive energy efficiency improves obviously after adopting the proposed strategy.

Abstract:With the continuous upgrading of the energy structure, the new parks with new energy power generation will play an important role in the future new power system. Uncertainties such as the randomness of demand, intermittency of wind and solar output, and volatility of electricity prices in electricity market are coupled together, making it difficult to achieve the reasonable operation between wind and solar energy and battery energy storage system. Considering the limitations of traditional optimization methods, a deep reinforcement learning method based on the PPO algorithm is proposed to solve the problem of interactive operation of wind-solar-storage in parks under uncertain environments. Based on the theoretical framework of reinforcement learning, a Markov decision model with continuous state space and continuous action space and unknown transition probability is constructed for the interactive operation of the park. The new load control system controls the battery energy storage system and flexible resources in the microgrid of the park to realize the economic operation, fully considering battery degradation.

Abstract:For short-term load forecasting of different types of users, support vector machine and deep learning model are widely used at present. A hybrid model is proposed to solve the problems of the least squares support vector machine model, such as the difficulty in determining the super parameters, the high data quality requirements of the model, and the slow optimization speed and easy to fall into the local optimization of the integrated conventional optimization algorithm. Firstly, the original feature data is clustered by hierarchical clustering and then the corresponding least squares support vector machine model is established for the same prediction day. Then, the super parameters in least squares support vector machine are heuristic searched by the improved simulated annealing algorithm. Finally, by comparing the performance of the load forecasting model with that of various load forecasting models, the results show that the proposed model can effectively improve the ac?curacy of load forecasting and shorten the forecasting time.

Abstract:A new power load forecasting method based on optimized Prophet algorithm is presented. A power load time series can be disassembled into trend term, seasonal term and random term by Prophet algorithm. In the predicted time series, trend and seasonal term are predicted by Prophet, while the random term is replaced by the prediction output by XGBoost algorithm. The model proposed is easy to understand and requires only power load data, whose parameters can be adjusted by the analysts intuitively.The effectiveness of the forecasting method is proved by conducting experiments on the power load data obtained from power consumption information acquisition system. The result shows that the mean absolute percentage error of the forecasting result is reduced by up to 2.5% comparing with the original Prophet algorithm. At the same time, the root mean square error is reduced by up to 30.19%, which show its excellent improvement.

Abstract:In order to improve the accuracy and generalization ability of load forecasting, a GRU-BiLSTM-Self-attention model based on Bagging integrated algorithm is proposed. In order to fully extract multiple features of high-dimensional input data, the model uses BiLSTM- Self- attention model to extract local features and GRU model to extract temporal features, independently extract sample subsets from the same training set and train them, integrate the output results and obtain the final prediction results. The real data of a power supply company in Nanjing are selected for the experiment and compared with the prediction models such as LSTM neural network, GRU neural network and BiSTM neural network.The experimental data show that the root mean square error of the model is 50.770 3, and the accuracy is 97.36% . Compared with other models used for comparison, the results show that this model has certain advantages in prediction effect, which shows that the proposed model has better generalization ability and prediction accuracy.

Abstract:Firstly, based on the combined prediction model of grey prediction model, linear regression model and BP neural network model, the traditional car ownership prediction curve is calculated, and three parameter values of traditional car ownership based on Bass model are fitted by nonlinear square method. Then, through the Delphi method based on AHP, the relationship between parameters of traditional vehicles and electric vehicles is constructed, and the Bass model that can predict the number of electric vehicles is obtained. On the basis of the prediction results of the inventory, the Monte Carlo algorithm is used to simulate the electric private cars,electric buses and electric vehicles in the city by combining the initial charging time, daily mileage, battery parameters, charging efficiency and other influencing factors of the user’s use of electric vehicles. The load prediction of electric vehicles are completed. The application of this method for electric vehicle load prediction has higher accuracy and better effect.

Abstract:As the peak load of electricity in China continues to rise, the importance of household load participation in demand response is increasing. Aiming at the lack of common interoperability standards for household appliances and the difficulty in interacting with the grid load, a double-layer game method based on clustering results is used to conduct the aggregate test of household appliance loads. After the household appliance loads is clustered by the clustering method based on fuzzy C-means, the application scenarios of automatic control of residents’environment and temperature, smart kitchen, smart residents’home theater, remote control of smart appliances and so on are established. According to the requirements of grid demand response, the optimal energy consumption strategy of residential household appliances is solved through the energy game on the lower game model. Based on the application scenarios of residential household appliance loads established after clustering, the optimal household appliance interoperability test method is solved through the aggregate value demand game on the upper game model. The calculation example has been applied in a certain urban community, and the test accuracy rate is 97.2%.The application results show that the method improves the interoperability of household appliance loads and the interaction level between household appliances and power grid loads.

Abstract:With the rapid increase of China’s total power demand and the implementation of the dual carbon target, the contradiction between power supply and demand is becoming more and more serious. Orderly use of electricity is an important mean to alleviate the contradiction between supply and demand in peak hours of power supply system. In order to fully tap the user’s ordered power consumption potential, a limitable load model based on the inner box approximation model is proposed. The user aggregation model is obtained by Minkowski Sum, which effectively simplifies the computational complexity of large-scale user planning and reduces the risk of user electricity information exposure. Furthermore, the power company establishes a multi-objective optimal scheduling model considering the value of emergency error peak avoidance, fairness and social impact of users when implementing orderly power consumption plan, in order to minimize the impact of lack of electricity on social economy and residents’lives. Finally, the effectiveness of the proposed model is verified by simulation, and its advantages in computing efficiency and privacy security are analyzed with an example.

Abstract:With the growth of wind power and photovoltaic generation, and the increasing of electric vehicle application, the balance characteristics have been furtherly affected between renewable energy and electric load, several resources are used as flexible regulation for the safety and stability of power system, such as energy storage system（ESS）and flexible interconnection（FI）. It is focus on the typical topology and application scenarios wind generation （WG）- photovoltaic（PV）-ESS- electric vehicle（EV）FI, combined with the randomness and disorder of electric vehicle load and the constraints such as energy balance, operation balance, reliable power supply, power fluctuations, the configuration method of the ESS is proposed, meanwhile, the proposed control optimization method leads to the configuration of the energy storage system can be greatly reduced by optimizing the control strategy. Under the same control and operational effect, the optimal configuration of the ESS is established.

Abstract:In order to effectively improve the economy of multi-station integration construction, based on ensuring the power supply reliability and new energy consumption of the system, a multi-station integration capacity optimal allocation model with levelized energy cost as the optimization objective is established, and a multi-station integration capacity optimal allocation method considering multiple uncertainties to solve the stochastic optimization model is proposed. The simulation results of multi-station fusion data show the effectiveness of the proposed method and model.

Abstract:Achieving the goal of“carbon peak and neutrality”has led to a rapid increase in the penetration rate of clean energy in the smart grid, which also requires the public buildings in the smart grid to have the ability to respond to demand, and the load of heating, ventilation and air-conditioning（HVAC）systems cannot be underestimated in the total load of public buildings. For the sake of reducing the load of the public building and achieving the effect of “supply on demand”, first of all, for the public building, its energy systems, heat balance of the building and the HVAC system are modeled, and based on the model a two-stage scheduling scheme is proposed. The proposed scheduling scheme aims at minimizing the cost of building electricity consumption while maintaining indoor environmental comfort, and controls the HVAC system to get the best day-ahead scheduling plan. Then, in the real-time scheduling stage, the optimal scheduling scheme is completed with the goal of reducing the carbon emissions during the normal operation of the public building. The effectiveness and economy of the proposed new optimal scheduling scheme are verified by the simulation of an actual public building.

Abstract:In order to promote cross-provincial mid-and-long term market transactions, and furtherly realize the optimal allocation of energy resources in a wider range, it is necessary to improve the cross-provincial mid-and-long term trading mechanism. Firstly,current situation of Chinese cross- provincial mid- and- long term market is analyzed from the perspective of market characteristics and existing problems. Secondly, the development, trading varieties, interregional coordination and other market information of PJM,Iberia, Northern Europe and the United Kingdom are combed, and then the differences between European and American markets and Chinese market are compared and analyzed in terms of market types, mid-to-long term market and spot coordination methods, electricity financial markets, promotion of new energy consumption mechanisms and deal mode. Finally, combined with European and American market-related mechanisms to Chinese cross- provincial mid- and- long term market development, the experiences on Chinese market construction are analyzed.