Efficient and fair scheduling of PHEV charging with neuro dynamic programming

Abstract

Plug-in hybrid electric vehicles (PHEVs) are a midterm solution to reduce the nation's dependency on oil. The batteries of these PHEVs can be recharged in the charging stations. However, with the extensive adoption of PHEVs, the charging load of PHEVs will pose considerable challenges to the grid because these PHEVs consume a large amount of electrical power. Therefore, it is important to implement coordinated charging. In this paper, an optimal charging scheduling of the PHEVs is proposed by minimizing the power losses and guaranteeing the fairness of PHEV charging. We employ a neuro-dynamic programming(NDP), which has in-built capability of studying from outside environment and adapting system parameters based on the performance requirements. Simulation results indicate that NDP outperforms two scheduling algorithms and simultaneously considers: (i) charging efficiency, (ii) charging fairness, and (iii) less power losses.