A Statistical modelling and analysis of residential electric vehicles' charging demand in smart grids

Electric vehicles (EVs) add significant load on the power grid as they become widespread. The characteristics of this extra load follow the patterns of people's driving behaviours. In particular, random parameters such as arrival time and charging time of the vehicles determine their expected charging demand profile from the power grid. In this paper, we first develop a model for uncoordinated charging power demand of an EV based on a stochastic process in order to characterize its expected daily power demand profile. Next, we illustrate it for different charging time distributions through simulations. This gives us useful insights into the long-term planning for upgrading the power systems' infrastructure to accommodate EVs. Then, we incorporate departure time as another random variable into this modelling and introduce an autonomous demand response (DR) technique to manage the EVs' charging demand. Our results show that, it is possible to accommodate a large number of EVs and achieve the same peak-to-average ratio (PAR) in daily aggregated power consumption of the grid as when there is no EV in the system without any change in the users' commuting behaviours. We also show that this peak value can be further decreased significantly when we add vehicle-to-grid (V2G) capability in the system.