A game-theory analysis of charging stations selection by EV drivers

Abstract

We address the problem of Electric Vehicle (EV) drivers' assistance through Intelligent Transportation System (ITS). Drivers of EVs that are low in battery may ask a navigation service for advice on which charging station to use and which route to take. A rational driver will follow the received advice, provided there is no better choice i.e., in game-theory terms, if such advice corresponds to a Nash-equilibrium strategy. Thus, we model the problem as a game: first we propose a congestion game, then a game with congestion-averse utilities, both admitting at least one pure-strategy Nash equilibrium. The former represents a practical scenario with a high level of realism, although at a high computational price. The latter neglects some features of the real-world scenario but it exhibits very low complexity, and is shown to provide results that, on average, differ by 16% from those obtained with the former approach. Furthermore, when drivers value the trip time most, the average per-EV performance yielded by the Nash equilibria and the one attained by solving a centralized optimization problem that minimizes the EV trip time differ by 15% at most. This is an important result, as minimizing this quantity implies reduced road traffic congestion and energy consumption, as well as higher user satisfaction.

A Comprehensive Game-Theoretic Model for Electric Vehicle Charging Station Competition

“…This is similar to our setting where queuing time increases with the number of EVs that travel to the same station. However, congestion games assume that players have incentive to follow a pure-strategy Nash equilibrium, which requires some coordination mechanism, as in [22], but in general it is more realistic to assume a mixed strategy for drivers, as is the case in other work [4], [6], [7].…”

Toward Anycasting-Driven Reservation System for Electric Vehicle Battery Switch Service

“…The CS-selection scheme in [6] adopts a pricing strategy to minimize congestion and maximize profit, by adapting the price depending on the number of EVs been parked. Game theory strategy [8] is also applicable by balancing the charging plans among EV drivers.…”

Coalitional Control for Self-Organizing Agents

“…4). Power setpoints are computed with the RTO (23). minimization of the energy transferred between connected areas can be implicitly addressed by penalizing large values of ∆θ i ; additionally, measures available from cooperating nodes can be exploited by penalizing the angle difference between the members of a given coalition.…”