A Spatio-Temporal Simulation Model for Electric Vehicle Charging Demands Considering User and Battery Behaviors

Accurate spatio-temporal estimation of electric vehicle (EV) charging demands is crucial for the operation and construction of charging networks. The complex behaviors of EV users (travel and charging) and batteries (discharge and charging) have a significant impact on charging demands. This study proposes a bottom-up simulation model (STEP-TV) for EV charging demands with the Monte Carlo method, considering the influence of user behavior, ambient temperature, and road congestion. 1) Firstly, vehicle trajectories are simulated based on multi-source data; 2) Secondly, the probability of users' destination charging is obtained using the Bayes formula; 3) Finally, an open-source tool, PyChargeModel, is introduced to simulate the non-uniform charging process and obtain the spatio-temporal distribution of charging demands. In a case study of Sioux Falls, we compared the simulation results of the proposed STEP-TV model and the EVI-Pro developed by the National Renewable Energy Laboratory (NREL). We found that user behavior and ambient temperature significantly impacted the charging demand estimation.