Assessment Of Black Carbon Actual Emissions From Light Gasoline Vehicles At Urban Road Intersections

As a key component of the urban road network,intersections are not only places where residents commute frequently,but also areas with high pollution in the city.The automobile exhaust emission is the main reason for serious pollution in the intersection area.Black carbon(BC)is a kind of ultra-fine particles aggregated in the form of carbon pellets,which is the main component of automobile exhaust particulates.BC has direct harm to public health and can cause diseases such as lung cancer,cardiovascular disease,central nervous system injury,etc.Therefore,it is necessary to study the BC emission of vehicles in the intersection area.In this study,the portable emission measurement system(PEMS)was used to measure the BC emissions of three light passenger vehicles at multiple signalized intersections and roundabouts on selected routes,and the BC emission factors in the intersection area and the contribution of vehicle operating conditions(acceleration,deceleration,idle speed and constant speed)to the total BC emissions of vehicles at intersections were evaluated,The multiple linear regression model is used to analyze the relevant influencing factors of BC emission of vehicles in the intersection area,and the BP neural network(BPNN)model is further used to predict the BC instantaneous emission factors of vehicles in the intersection area and compared with the generalized additive model(GAM),and then the vehicle BC emission prediction model with the best prediction effect is proposed.The main research conclusions of this paper are as follows:Firstly,a mobile measurement method of vehicle BC emission in the intersection area is established,and the BC emission factors of vehicles at signalized intersections and roundabouts are calculated,and the BC emission contribution of each driving condition(acceleration,deceleration,constant speed and idle speed)in the intersection area is calculated by combining the On-Board Diagnostics(OBD)data.The results show that the BC emission factor generated by vehicles in the intersection area is higher than that generated in the entrance and exit connection section.In the signalized intersection,the BC emission factor of vehicles is 64.79%higher than that of the entrance connection section and 70.35% higher than that of the exit connection section;At the roundabout,it is 26.13% higher than the entrance connection section and 49.05% higher than the exit connection section on average.Among the four driving conditions,the contribution of acceleration condition to emissions is the largest,followed by deceleration condition.The contribution of idle condition increases with the complexity of the change of working conditions.The trend of constant speed condition and idle condition is opposite.Secondly,the multiple linear regression method is used to analyze the effect of various influencing factors on vehicle BC emissions in the intersection area.The analysis results show that at the 95% confidence level,the BC emission factor of vehicles in the intersection area has a significant positive correlation with vehicle specific power(VSP)(P=0.000<0.05),delay time(P=0.001<0.05)and engine intake technology(P=0.016<0.05),but has no significant correlation with the emission standard(P=0.650>0.05).For the engine intake technology and emission standards,the significant results show that there are significant differences in BC emission factors between GDI vehicles and PFI vehicles in the intersection area,but there is no significant difference between the China-5 vehicles and the China-6 vehicles.The influence of each variable on vehicle BC emission factors in the intersection area is VSP,delay time and engine technology from the largest to the smallest.Finally,based on the measured data,this paper uses the BPNN algorithm to establish a prediction model of vehicle BC emissions in the intersection area and compares it with GAM.The results show that the predicted value of the BPNN model is close to the distribution of the true value in the validation set,indicating that the model has strong generalization ability.Compared with GAM,BPNN model has better prediction effect and can more accurately predict the BC emissions of vehicles in urban intersection areas.