学术文献库

Electric Vehicles (EV) impact urban networks both when driving (e.g., noise and pollution reduction) and charging. For the electrical grid, the flexibility of EV charging makes it a significant actor in "Demand Response" mechanisms. Therefore, there is a need to design incentive mechanisms to foster customer engagement. A congestion game approach is adopted to evaluate the performance of such electrical transportation system with multiple classes of vehicles: EV and Gasoline Vehicles. Both temporal and energy operating costs are considered. The latter is nonseparable as it depends on the global charging need of all EV, which is scheduled in time by a centralized aggregator in function of nonflexible consumption at charging location. Thus, driving and charging decisions are coupled. An adaptation of Beckmann's method proves the existence of a Wardrop Equilibrium (WE) in the considered nonseparable congestion game; this WE is unique when the charging unit price is an increasing function of the global charging need. A condition on the nonflexible load is given to guarantee the monotonicity of this function. This condition is tested on real consumption data in France and in Texas, USA. Optimal tolls are used to control this electrical transportation system and then computed in order to minimize an environmental cost on a simple network topology.