A Prediction Method for EV Charging Load Considering Driving Energy Consumption under Speed Forecasting

In the previous research of electric vehicles (EVs) charging load prediction, the energy consumption during EVs driving are not considered reasonably. Aiming at this problem, a prediction method for EV charging load considering driving energy consumption under speed forecasting is proposed. First, according to the National Household Travel Survey (NHTS) data, the trip chain is introduced, and the space-time variables in the trip chain are modeled. Then, considering that the speed of EVs is not constant in the process of driving, the BP neural network is used to predict the speed of EVs, and then the energy consumption model of EVs is established according to the integral of battery power output. Next, a fuzzy inference system(FIS) is established considering three input variables, including battery SOC, driving time of the next trip and parking time, to generate charging probability describing the user's charging intention. Finally, the load distribution curves of EV in different charging regions are simulated by Monte Carlo method. Simulation results verify the feasibility and accuracy of this prediction method.