论文部分内容阅读
使用车载排放测试系统(PEMS)采集轻型电喷汽油车道路实际污染物排放率数据,并利用GPS系统获得测试车辆测试过程的实际行驶工况.定义一段较短时间内的车速变化历程为短时实际行驶工况,以短时实际行驶工况表示车辆运行状态并将其各时刻的速度作为排放模型的参数,用BP神经网络的方法建立了机动车微观排放模型.模型运行结果表明,二氧化碳(CO2)、氮氧化合物(NOx)、一氧化碳(CO)、碳氢化合物(HC)等污染物的排放率预测总体误差分别在4％、2％、5％、5％以下,检验了通过短时实际行驶工况各时刻速度计算机动车污染物排放率的方法的可行性.“,”Using a portable emission measurement system (PEMS),an emission test of a light-duty electronic injection gasoline vehicle was carried out to collect the on-road emission data.During the test,the real driving cycle of the test vehicle was obtained using a GPS system.The concept of a real short-term driving cycle was proposed to represent the vehicle operating conditions,and a microscopic emissions model was established based on a BP neural network.In the network,the input parameters were the instantaneous vehicle speeds of the real short-term driving cycle.The results show that the total errors in the carbon dioxide (CO2),nitrogen oxide (NOx),carbon monoxide (CO),and nitrogen oxide (HC) were less than 4％,2％,5％,and 5％,respectively.Accordingly,the feasibility of calculating the vehicle emission rates through the instantaneous vehicle speeds of the real short-term driving cycle was verified.