Charge/Discharge Event Model for Electric Vehicles with Predictive Control Application

Electric vehicles (EV)s are acknowledged as key technologies to decarbonize the transportation sector and counteract variable power generation. However, in order for EVs to have an impact on the power system, a large number of them must be managed and aggregated, and this requires an efficient modeling of their charging/discharging operations. In this paper, EVs charge/discharge operations are modeled as charging/discharging events. The modeling approach is based on historical utilization of charging stations and allows individual EV owners to opt in/out of bidirectional energy exchange once they connect their vehicles to the energy network. Binary decision variables are used to monitor vehicles' presence in the network and owner's charging preference. The energy network is modeled using a control oriented modeling framework derived from the energy hub concept. This novel EV modeling approach is integrated into the modeling framework and used in the formulation of an optimization problem, leading directly to the design of a model predictive controller. Using this scheme, EV presence and state-of-charge can be continuously monitored and assessed, and predictions of future EV charging/discharging events can be efficiently incorporated.