Abstract:

To encourage the energy saving and emission reduction at the signalized intersections and improve the efficiency of road traffic, this paper establishes an ecological driving induction speed control strategy and a fuel consumption emission model based on the target vehicle speed correlation. The multi- objective genetic algorithm is used to optimize the target speed of ecological driving under the scenario of acceleration and deceleration crossings. The fuel consumption, emission, and travel time are defined as the objective functions, and the posted speed limit and non-stop passing speed are used as the constraints. Then, the joint simulation of different algorithm permeability and road saturation scenarios was carried out using MATLAB and VISSIM simulations. The joint simulation results were imported into the micro emission model MOVES to estimate the energy consumption and emission. The results show that the control strategy can increase the average speed by 13.8%, reduce the proportion of idling mode by 33%, and increase the proportion of medium speed cruise mode by 18%. The energy consumption and emissions of N2O, NOX, HC and CH4 were reduced by 6.6%, 12.2%, 4.0%, 6.3%, 2.9% respectively, and CO emissions increased by 1.5% under the scenario of high algorithm permeability and low road saturation. At last, the chassis dynamometer test was performed for the speed trajectories under different control strategies. The results show that compared with the scenario without speed control, the proposed strategy reduced the energy consumption by 53.1% and the emissions of CO, CO2 and PN respectively by 47.6%, 50.4%, 39.8% with NOX emission increase of 13.6%.

Key words: intelligent transportation, eco- driving, multi- objective genetic algorithm, signalized intersection, traffic simulation

摘要：

为实现车辆在信号交叉口区域的节能减排及提高道路通行效率，本文构建基于目标车速关联的油耗排放模型，建立生态驾驶诱导车速控制策略。在加减速通过场景下以油耗、排放和通行时间为优化目标，以道路限速和不停车通过车速为约束，利用多目标遗传算法优化生态驾驶目标车速；基于MATLAB与交通仿真软件VISSIM进行不同算法渗透率及道路饱和度场景下的联合仿真，将仿真结果导入微观排放模型MOVES测算能耗排放。仿真结果表明：控制策略与无控制时相比，在高算法渗透率、低道路饱和度场景下，车辆平均速度提高13.8%，怠速工况比例下降 33%，中速巡航工况比例上升18%，能耗及N2O、NOX、HC、CH4排放分别减少6.6%及12.2%、4.0%、 6.3%、2.9%，CO排放增加2.5%。最后，依据仿真得到不同控制策略下的速度轨迹在底盘测功机上完成实车实验，实验结果表明，基于交通流优化的控制策略与无控制场景相比，能耗及 CO、 CO2、PN排放分别减少53.1%及47.6%、50.4%、39.8%，NOX排放增加13.6%。

关键词: 智能交通, 生态驾驶, 多目标遗传算法, 信号交叉口, 交通仿真