Energy management for plug-in hybrid electric vehicles via vehicle-to-grid

As a paradigm of the incoming smart grid, vehicle-to-grid (V2G) has been proposed as a solution to increase the adoption rate of plug-in hybrid electric vehicles (PHEVs). In this work, we investigate the energy management strategies for PHEVs via bidirectional V2G. We first follow a cost-conscious approach. To minimize the daily energy cost, we formulate the energy management problem through dynamic programming. However, the “well-known” complexity in solving dynamic programming poses a computational challenge even for a small number of iterations. Therefore, we prove that a state-independent four-threshold (s, S, s´, S´) feedback policy is optimal for PHEV battery charging/discharging based on stochastic inventory theory. A backward iteration algorithm is further developed to practically implement the above (s, S, s´, S´) policy. Second, aiming to minimize the peak load and flatten the overall load profile, we propose an optimal PHEV charging scheme and derive a reminiscent “water-filling” solution for this scenario. Realistic PHEV battery models, time-of-use (TOU) electricity pricing rate and real data of household demand are integrated into our formulated PHEV model. The theoretical analysis and proofs are instrumental to the future large-scale PHEV adoption in smart grid.