Demand Response System Based On Electrical Behavior Analysis

With the rapid increase of peak load,the peak-to-valley difference continues to increase,the load rate drops rapidly,and new energy fluctuations and environmental pollution problems make the application of demand response urgent.At present,the research on demand response at home and abroad mostly focuses on the response mechanism modeling and response effect evaluation,but rarely on the demand side of the actual system solution.Therefore,it is imperative to use the data collected on the demand side to fully exploit the users' electrical behavior and exert the role of demand response technology to develop a demand response system that can meet the individual needs of users.The specific work of this paper is as follows:(1)Analysis of electrical behavior.In view of the difference in grasping the characteristics of power load during the implementation of demand response,a typical power consumption mode is extracted by analyzing the behavior of electricity.The basic theory of electrical behavior analysis is discussed.From the perspective of electrical behavior analysis,a bilayer clustering algorithm for power load curve considering wavelet entropy dimension reduction is proposed.Considering the increasing load data dimension,the algorithm analyzes the volatility of the load curve from the time domain and the frequency domain by calculating the wavelet entropy,which reduces the dimensionality of the data and improves the efficiency of clustering.Considering the difference in the trend of user power consumption in the demand response,the algorithm adopts bilayer clustering.The outer layer adopts a spectral clustering algorithm based on cosine similarity to obtain morphological similarity load clusters.On the basis of the outer layer morphological similarity clustering,the inner layer uses the k-means clustering algorithm based on Euclidean distance to obtain the amplitude similarity load clusters.The simulation verifies that the algorithm is superior in terms of computation time,clustering validity and stability.(2)Demand response research based on electrical behavior analysis.Firstly,a short-term load forecasting model based on long short-term memory network algorithm is established to predict the user's next-day load,which improves the accuracy of the forecast.It is used to determine whether it is necessary to perform demand response,the time and the total amount of load required to perform demand response.Then the role of the bilayer clustering algorithm in the demand response is discussed.The morphological similarity load clusters and the amplitude similarity load clusters obtained by the bilayer clustering algorithm are combined with the fuzzy clustering algorithm based on peak-valley membership degree and the demand price elasticity theory.An effective demand response resource library can be established to implement accurate demand response by analyzing.Finally,the role of bilayer clustering results in demand response is verified.(3)The building of demand response system.Through the analysis of the overall demand of the demand response system,the system architecture is determined.Then the key technologies such as database,Web,and SVG are combined with the electrical behavior analysis to develop and implement the demand response system.System operation indicates that the system is of good interactivity and strengthens the interaction between the supply and demand sides,making the demand response more convenient and efficient.In summary,based on the in-depth discussion of electrical behavior analysis and demand response theory,this paper proposes a bilayer clustering algorithm for power load curve with wavelet entropy reduction dimension taken into consideration.The simulation proves the validity of the proposed algorithm.At the same time,a short-term load forecasting model based on long short-term memory network algorithm is established.The bilayer clustering algorithm is applied to the demand response analysis,and the demand response system based on the electrical behavior analysis is constructed,which realizes the interaction between power supply and demand sides,strengthens the management of demand response,and ensures the power system operates safely and steadily through demand response.