Predicting Electric Vehicle Charging Demand using Mixed Generalized Extreme Value Models with Panel Effects
In the past 5 years Electric Car use has grown rapidly, almost doubling each year. To provide adequate charging infrastructure it is necessary to model the demand. In this paper we model the distribution of charging demand in the city of Amsterdam using a Cross-Nested Logit Model with socio-demographic statistics of neighborhoods and charging history of vehicles. Models are obtained for three user-types: regular users, electric car-share participants and taxis. Regular users are later split into three subgroups based on their charging behaviour throughout the day: Visitors, Commuters and Residents.
|, , ,|
|International Conference on Ambient Systems, Networks and Technologies|
|Organisation||Centrum Wiskunde & Informatica, Amsterdam, The Netherlands|
Berkelmans, G.A, Berkelmans, W.M, Piersma, N, van der Mei, R.D, & Dugundji, E.R. (2018). Predicting Electric Vehicle Charging Demand using Mixed Generalized Extreme Value Models with Panel Effects. In Procedia Computer Science (pp. 549–556). doi:10.1016/j.procs.2018.04.080