Optimal recharging policies for electric vehicles

Recharging decisions for electric vehicles require many special considerations because of battery dynamics. Battery longevity is prolonged by recharging less frequently and at slower rates, and also by not charging the battery too close to its maximum capacity. In this paper, we address the problem of finding an optimal recharging policy for an electric vehicle along a given path. The path consists of a sequence of nodes, each representing a charging station, and the driver must decide where to stop and how much to recharge at each stop. We present efficient algorithms for finding an optimal policy in general instances with deterministic travel costs and homogeneous charging stations, and also for two specialized cases - one where the vehicle can stop anywhere along the path to recharge and another with equidistant charging stations along the path. In addition, we develop two heuristic procedures that we characterize analytically and explore empirically. We further analyze and test our solution methods on model variations that include stochastic travel costs and nonhomogeneous charging stations.