Assessing the influence of the temporal resolution on the electric vehicle charging load modeling accuracy

Abstract

In the scientific literature, various temporal resolutions have been used to model electric vehicle charging loads. However, in most studies, the used temporal resolution lacks a proper justification. To provide a strengthened theoretical background for all future studies related to electric vehicle charging load modeling, this paper investigates the influence of temporal resolution in different scenarios. To ensure reliable baselines for the comparisons, hardware-in-the-loop simulations with different commercial electric vehicles are carried out. The conducted hardware-in-the-loop simulations consists of 134 real charging sessions in total. In order to compare the influence of different temporal resolutions, a simulation model is developed. The simulation model utilizes comprehensive preliminary measurement-based charging profiles that can be used to model controlled charging in fine detail. The simulation results demonstrate that the simulation model provides sufficiently accurate results in most cases with a temporal resolution of one second. Conversely, a temporal resolution of 3600 s may lead to a modeling error of 50% or even higher. Additionally, the paper shows that the necessary resolution to achieve a modeling error of 5% or less vary between 1 and 900 s depending on the scenario. However, in most cases, resolution of 60 s is reasonably accurate.