1.Electric Power Research Institute of State Grid Jiangsu Electric Power Co., Ltd., Nanjing 211103, China;2.State Grid Jiangsu Electric Power Co., Ltd., Nanjing 210024, China;3.School of Electrical Engineering, Southeast University, Nanjing 210096, China

In order to adapt to the rapid growth of electric vehicles (EVs) and charging demand, this paper proposes an active charging guidance model of EVs based on the Internet of vehicles by using the improved A* path planning algorithm and the queuing theory from the perspective of EV users. Firstly, incorporating the traffic light waiting time and the no-backtracking condition, the A* path planning algorithm is improved to update the spatiotemporal state matrix of the road network using the actual road network state information, which can optimize the EV driving path in real time and obtain the EV traveling time for charging. Secondly, the deep belief network (DBN) is utilized to predict the short-time arrival numbers of EVs at the charging station, and the EV waiting time for charging is predicted based on the M/G/k model using the queuing theory. Finally, the active charging guidance model of EVs is constructed to minimize the traveling time and waiting time of EVs for charging. Taking the central area of Nanjing, China as an example, the effectiveness of the proposed active charging guidance model is verified. The proposed algorithm can improve the utilization rate of charging piles and reduce the comprehensive charging time of EV users.