基于远程在线监控车载终端的柴油车NOx排放等级诊断研究

远程在线监控车载终端集成了远程通讯模块、卫星定位模块、发动机OBD信息解析模块，能够实时读取车辆排放相关运行信息，但无法直接判断车辆NO_x排放情况.为了快速、准确地评估车辆排放情况，诊断和监测NO_x高排放车，同时为了克服有些重型柴油车监测数据中缺失进气流量、燃油流量、车速等重要的实时信息，无法计算出车辆NO_x排放因子的问题.本文提出了由NO_x浓度分布特征驱动的高排放重型柴油车识别算法，通过远程在线监控车载终端设备获取车辆的发动机信息和SCR系统运行信息，运用NO_x浓度分布计算车辆每天NO_x排放浓度占比，通过系统聚类法对车辆NO_x排放浓度占比进行聚类，结果聚为优、良、中、差4类.利用车辆NO_x排放浓度区间分布及其聚类结果分别作为训练集的输入和输出，选择BP神经网络作为训练算法，训练获得的模型分类准确率为90%，利用训练好的模型判断在用柴油车NO_x排放等级，从而识别及监测NO_x高排放车辆.研究结果可为柴油车NO_x高排放诊断及监测提供依据，有助于监管部门能够快速识别NO_x高排放车辆.

Diagnosis of NOx emission level of diesel vehicles based on remote online monitoring terminal

The remote online monitoring terminal integrates a remote communication module, a satellite positioning module and an OBD information analysis module, which can read the operational information related to emissions vehicles in real time, but cannot judge NO_x emissions vehicles directly. In order to assess vehicle emissions quickly and accurately, diagnose and monitor NO_x high-emission vehicles, and overcome the problem of incomplete monitoring data for some heavy-duty diesel vehicles, based on the monitoring results, a recognition algorithm for high-emission heavy-duty diesel vehicles classified by NO_x concentration distribution characteristics was proposed. In this article,vehicle engine information and SCR system operation information were obtained by the remote online monitoring terminal.The NO_x concentration distribution was used to calculate daily NO_x emission concentration ratio of vehicles. The systematic clustering method was used to cluster NO_x emission concentration ratio of vehicles, and the results were classified into four categories:excellent, good, medium and poor. The distribution of vehicle NO_x emission concentration and its clustered results were used as the input and output of the training set, and BP neural network was selected as the training algorithm. The classification accuracy of the trained model was 90%, and the trained model was used to judge the NO_x emission level of diesel vehicles in use, which was used to identify and monitor NO_x high-emission vehicles. The research results are intended to provide a basis for the diagnosis and monitoring of diesel vehicles with high NO_x emissions, and help regulatory agencies to identify vehicles with high NO_x emissions quickly.