Modelling the Intensity of Electric Vehicle Arrivals at Charging Points

Abstract

With Electric Vehicles’ (EV) market adoption surging in recent years, the smart grid paradigm requires accurate forecasts of EV arrivals at charging points. One efficient way to model these arrivals is to use Point Processes. This study introduces an additive model using both spline and wavelet effects for fitting the intensity of a non-homogeneous Poisson process applied to EV arrivals at charging points. The key contribution of this work is a novel estimation procedure inspired from backfitting which is illustrated by a case study on real-world EV arrivals at charging points. The results obtained show that this approach can help better capturing EV arrival peaks.