车联网环境下电动汽车主动充电引导模型

Active Charging Guidance Model of Electric Vehicles Based on Internet of Vehicles

为了适应电动汽车数量和充电需求的急剧增长,从电动汽车用户视角出发,提出了一种在车联网环境下基于改进A*路径规划算法与排队论的电动汽车主动充电引导模型。首先,融入红绿灯等待时间和不走回头路条件,改进A*路径规划算法,利用实际路网状态信息更新路网时空状态矩阵,实时优化电动汽车行驶路径,获取电动汽车充电行驶时间。其次,利用深度置信网络预测充电站电动汽车短时到达量,基于排队论M/G/k模型预测电动汽车充电等待时间。最后,以最小化电动汽车充电行驶时间和充电等待时间为目标,搭建电动汽车主动充电引导模型。以中国南京市中心区域为算例,验证了所提主动充电引导模型的有效性,所提算法能够提高充电桩的利用率并减少电动汽车用户综合充电时间。

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.