Design And Implementation Of EV Charging System Based On Electric Load Regulation

Aiming at the problems of line overload,voltage sag and poor power quality caused by ev charging,the current charging management systems mostly focus on the peak-valley difference of theload curve,and seldom consider the real-time state and predictive state of the grid,and sometimes do not meet the protection requirements of relay protection system.Therefore,the ev charging system has important research significance,which can regulate load according to the state of the grid and meet the protection requirements of the grid.Based on the tree structure of charging system,the node state is divided into current protection state,voltage protection state,power improvement state and normal state.Considering the historical state,real-time state and predictive state of the grid,the criterion of different node states is designed.According to the node state,the method of distinguishing system load state is designed.On this basis,the combination forecasting method of wavelet transform and BP neural network is chosen to forecast the non-charging load using the historical data of the grid to obtain the forecasting status of each node,and the real-time state of each node is acquired by real-time sampling.According to tree structure and node state,load regulation constraints and load regulation solving strategies are designed.Finally,according to the time requirement of the grid protection,the protection time margin under different conditions is analyzed.The power line carrier communication is selected as the load regulation instruction transmission mode,and the networking mode is determined.The feasibility of networking is verified by calculating the time loss of information transmission.In order to verify the feasibility of the system,the system hardware and software are designed and experiments are carried out.In the aspect of hardware,the industrial computer and STM32F4 series chips are chosen,the key circuits such as power supply,sampling,communication and load regulation are designed.The software has completed the key functions of historical data processing,power prediction,power grid information collection,load regulation decision,charging power regulation and so on.Finally,the system is verified experimentally.The feasibility of the system is verified from three aspects: power prediction,information collection and load regulation,and the system meets the expected requirements.