基于网联车轨迹重构的交通油耗和排放估计方法

Fuel consumption and transportation emissions evaluation based on connected vehicles trajectory reconstruction

如何基于有限的轨迹数据对油耗和交通排放进行精准地估计一直是研究的难点和热点,为解决该问题,提出了基于轨迹重构的油耗和排放估计方法。首先,基于跟驰模型,利用网联车(Connected Vehicle,CV)轨迹重构道路上常规车(Regular Vehicle,RV)轨迹。然后,将轨迹重构结果与VT-Micro排放模型相结合,对道路上所有车辆的油耗和排放进行估算。最后,通过数值仿真试验验证模型在不同CV渗透率和交通流密度条件下的可行性和有效性。结果表明:该方法能够实现对油耗和排放的精准估算,且随着CV渗透率和交通流密度的增加,油耗和排放估计的准确性不断提高;当CV渗透率不低于30%,且交通流密度为60 veh/km时,交通油耗估计的平均绝对百分比误差小于2.17%,排放估计的平均绝对百分比误差小于8.66%;当CV渗透率为50%时,在不同交通流密度条件下车辆油耗估计的平均绝对百分比误差小于2.45%,排放估计的平均绝对百分比误差小于13.68%。

How to accurately estimate fuel consumption and transportation emissions based on limited trajectory data has always been a hot and difficult issue.To solve this problem,this paper proposes a fuel consumption and emissions evaluation method based on trajectory reconstruction.First,two optimization models based on the IDM car-following model are proposed to reconstruct the regular vehicle’s trajectory by using the connected vehicle’s trajectory.Secondly,the reconstructed trajectories and VT-Micro emission model are combined to evaluate the fuel consumption and transportation emissions of all vehicles on the road.Then,the evaluation indexes are developed to measure the accuracy of the trajectory reconstruction model and the estimated value of fuel consumption and transportation emissions.Finally,the feasibility and effectiveness of the method proposed in this paper under different penetration rates of connected vehicles(CV)and traffic densities are verified by numerical simulation experiments.Results show that(1)this method can accurately evaluate the fuel consumption and transportation emissions,and the accuracy of estimated fuel consumption and transportation emissions increases with the increase in the penetration rate of CV and traffic flow density;(2)When the penetration rate of CV more than 30%and the traffic flow density is 60 veh/km,The Mean Absolute Error(MAE)of the estimated fuel consumption is less than 1.25 L;the Mean Absolute Percentage Error(MAPE)is less than 2.17%,and the Root Mean Square Error(RMSE)is less than 1.42 L.The MAE of estimated transportation emissions is less than 1.34 g;the MAPE is less than 8.66%,and the RMSE is less than 1.43 g.(3)When the penetration rate of CV is 50%,the MAE,the MAPE and the RMSE of estimated fuel consumption under different traffic flow densities is less than 0.489 L,2.45%and 0.613 L,and the MAE,the MAPE,and the RMSE of estimated transportation emissions is less than 0.739 g,13.68%,and 0.875 g.