Research On Demand Response Strategy Based On Load Characteristics Under Energy Internet Environment

Energy Internet is not only an important development trend of energy system,but also an important means to promote the revolution of energy production and consumption.However,under the environment of Energy Internet,the power supply and demand situation are more complex.On the supply side,the proportion of renewable energy on the supply side is increasing,the distributed and centralized energy supply are developing in coordination,and the cold,heat,electricity,gas,and other energy systems are coupled with each other.On the demand side,the types of power terminals are increasingly diverse,the power consumption patterns are various,the power load is increasing rapidly,and the load characteristics are complex and changeable.Therefore,under the environment of Energy Internet,the problem of power supply and demand mismatch is very prominent.Demand response,as an important method of demand side management(DSM),can promote the balance of power supply and demand and ensure the security of energy system.However,the traditional demand response strategies are relatively simple,which cannot meet the personalized and diversified energy needs of users.Under the environment of Energy Internet,massive,high-dimensional,and heterogeneous power consumption data can be collected,which provides an important support for more accurate and effective mining of users' power load patterns and more comprehensive and in-depth analysis of users' power load characteristics,thus helping to design accurate and diversified demand response strategies.Therefore,this dissertation focuses on the problem of demand response strategy based on load characteristics under the environment of Energy Internet.Based on the effective identification of users' load characteristics,the incentive-based demand response model,price-based demand response model and integrated demand response model are constructed to design flexible and effective demand response strategies for uses.This provides theoretical support for the safe,reliable,economic,and efficient operation of energy system.The specific research contents and innovations of this dissertation are as follows:(1)Data driven load characteristics recognition method.In order to effectively identify the power load pattern and its characteristics,a load curve clustering algorithm based on improved k-means and a load curve clustering method based on dynamic time warping are proposed,and the corresponding process and steps of load curve clustering are given.Secondly,in order to accurately grasp the change trend and characteristics of power load,a short-term load forecasting model based on deep learning is constructed,and the relevant modeling steps are given.At the same time,a correction method for load forecasting results is proposed,which further improves the accuracy of load forecasting and lays the foundation for the accurate design of diversified demand response strategy.(2)Incentive-based demand response strategy based on load characteristics.Based on the analysis of the trend and characteristics of the user's power load,an incentive-based demand response model based on the load characteristics is constructed according to the difference of users' load characteristics.The model takes the maximization of the total profit of the participants in demand response as the objective function and considers the uncertainty of both sides of power supply and demand.The model is solved by the method of reinforcement learning,thereby providing personalized incentive-based demand response strategies for users.In addition,through experiments,the implementation effect of incentive-based demand response strategies under two different scenarios is compared and analyzed,and the effectiveness of the proposed model is verified.(3)Price-based demand response strategy based on load characteristics.Based on the identification of users' load characteristics and user's types,a price-based demand response model is constructed for different types of users.In this model,game theory is used to model the dynamic interaction among participants in demand response,and the existence of Nash equilibrium solution is proved by reverse induction method,and then the optimal dynamic electricity price strategy is obtained.Furthermore,the impact of the uncertainty of renewable energy on the price-based demand response strategy is considered,and the proposed price-based demand response model is verified by real data sets.(4)Integrated demand response strategy based on load characteristics.In the environment of Energy Internet,the multi-energy resources such as cold,heat,electricity and gas can be coupled and transformed with each other.In order to improve the integrated energy efficiency,a multi-energy complement-based integrated demand response mechanism is proposed.Based on the changing trend and characteristics of the power and thermal loads of users,a load optimization model of integrated energy system considering the integrated demand response is constructed.The model is further transformed into a mixed integer linear programming model,and the corresponding integrated demand response strategy is obtained.Finally,the economic performance and reliability of the integrated energy system under different scenarios is analyzed and compared by simulation experiments.Through the above research,this dissertation enriches and develops the theoretical system of DSM under the environment of Energy Internet,as well as provides important theoretical support for promoting the balance of power supply and demand,improving energy efficiency,and promoting the realization of peak carbon dioxide emissions and carbon neutrality.With the rapid development and construction of the Energy Internet,users' requirements for the quality of energy services continue to improve,and integrated energy services will become an important research direction in the future.