Time-of-Use Electricity Price Strategy And Application Considering New Energy Consumption And Demand Response

New energy is one of the strategic emerging industries that China is accelerating incubation and development.With the continuous deepening of energy structure reform,the grid-connected capacity of wind power continues to increase,and the scale of electric vehicles is gradually expanding.New energy has become an important part of the smart grid.However,this also has an impact on the stable operation of the power grid.How to eliminate wind abandonment at night and guide users to orderly use electricity has become a hot issue.Based on the load forecasting results of the deep learning method,this article aims to absorb wind power,proposes a time-of-use electricity price strategy that takes into account user demand response and applies it to scenarios to achieve optimal scheduling of electric vehicle charging.First,study the power load forecasting method based on deep learning to prepare for the formulation of time-of-use electricity prices.Use dilated convolution,residual connection and skip connection to construct a convolutional neural network.In order to improve the efficiency and accuracy,the Tent mapping and simulated annealing mechanism are introduced to improve the moth fire suppression algorithm.Based on the convolutional neural network structure and optimization algorithm,an improved moth-fighting algorithm optimized convolutional neural network model(IMFO-CNN)is established,which is used for load forecasting,adding meteorological conditions and other data to the network input to predict load and new energy output,Which effectively improves the prediction accuracy.Then,combined with the electricity price elasticity theory,the k-means clustering algorithm is used to classify users into several categories with different elastic coefficients,and the FCM clustering algorithm is used to reasonably divide a day into peak-to-valley periods,so as to formulate different time-of-use electricity price plans.In the two scenarios of the park microgrid and the distribution network,different user groups are individually formulated time-of-use electricity pricing strategies.The overall pricing plan and the classified pricing plan effectively achieve the goal of reducing the peak-to-valley difference and absorbing wind power,and discussedFinally,based on the historical data of automobile travel,a model of electric vehicle travel behavior is established to determine the availability of each electric vehicle and group of electric vehicles.Considering the uncertainty of available vehicles and the uncertainty of battery SOC,develop an electric vehicle charging optimization method and build a simulation framework.In the time-of-use electricity price scenario,compared to orderly charging and disorderly charging considering uncertainty,the developed charging optimization method significantly reduces user charging costs and load peaks.