Study On The Relationship Between Mileage And Emission Deterioration Of Light-duty Gasoline Vehicles

In recent years,China’s private car ownership has grown rapidly,and the pollutant emissions caused by it cannot be ignored.Among it,the accumulated mileage is closely related to vehicle emissions,and exploring the relationship between the accumulated mileage of vehicles and the emission of pollutants is of great significance for the management of high-emission vehicles.This study used the Acceleration Simulation Mode(ASM)Test Procedure data of light-duty gasoline vehicles in Nanjing in 2018,and studied the technical distribution characteristics of light-duty gasoline vehicles through mathematical statistical methods,and revised the emission limits of ASM5025 Test Procedure.With the help of commonly used mathematical models,the emission deterioration law of vehicle pollutants with accumulated mileage was analyzed,and different models were constructed to predict the accumulated mileage of vehicles when they reach the emission limit.Finally,the model results were compared and evaluated.The main research contents of this article are as follows:(1)In terms of vehicle technical distribution characteristics,a classification system based on emission standards,vehicle series and quality intervals was determined.Based on this,the distribution characteristics of accumulated mileage and pollutants of the vehicle were analyzed,and the vehicle statistical results under the classification system were given.And the results are as follows:1)The box chart was combined with the accumulated mileage distribution to determine the effective accumulated mileage range;2)The distribution of the three pollutants under the ASM5025 and ASM2540 Test Procedure basically exhibits a left-skewed distribution characteristic,and the emission control effect brought by the national standard IV stage and national standard V stage is more excellent;3)The emission level of domestic cars and German cars is higher than that of European/American cars and Japanese/Korean cars.(2)In terms of emission deterioration law of vehicle pollutants.Logarithmic regression and exponential regression models were selected to establish the deterioration model of vehicle pollutants and accumulated mileage in each classification interval.The modeling results are generally good.And the deterioration of vehicle pollutants in Nanjing City,Jilin Province and Foshan City were compared and analyzed.The following conclusions are obtained:1)Pollutant emissions from light-duty gasoline vehicles increase with increasing mileage;2)The emission levels of domestic-made cars at different national stages are the worst of the four cars,the medium to high quality cars in German and European/American have lower emission levels than the low-quality cars,but the Japanese/Korean cars have the opposite law;3)In the national standard I/II stage,the emission deterioration of the four vehicle series at the stage of 40,000-80,000 km is relatively smooth,and deteriorate rapidly at the stage of 80,000-200,000 km,then tends to oscillate gently;4)In the national standardⅢstage,the degree of vehicle deterioration is reduced,manifested as slow deterioration and then rapid deterioration,but there is no obvious characteristic that the pollutants oscillate gently with the increse accumulated mileage;5)In the national standard IV stage,in addition to domestically produced vehicles,the degree of deterioration of vehicles has been further reduced.Among them,the degree of deterioration of pollutants in European/American and Japanese/Korean vehicles is very small;6)The vehicle pollutants in Jilin Province has the highest deterioration degree,followed by Nanjing,and Foshan is the lowest.Among them,Nanjing should focus on NO control.(3)In terms of accumulated mileage prediction modeling.First,according to the plotted relative concentration of pollutants and cumulative frequency curve,the ASM5025 Test Procedure emission limit was revised.Then,using three categorical variables:emission standards,car series and quality interval,one continuous variables:divided pollutants value as the input variables,multiple linear regression,random forest regression and neural network models were established to predict the accumulated mileage when the vehicle reached the emission limit.The modeling results show that the neural network prediction model is optimal,and the R~2 of the NO,CO and HC models are 0.83,0.74 and 0.78,respectively.Based on this,a table of accumulated mileage when the emission of light-duty gasoline vehicles reach limit is given.