Study On The Emission And Diffusion Of Fine Particles Of Motor Vehicles At Urban Intersection

In recent years,with the continuous development of urban transportation in China,convenient transportation brings convenience to people,but also brings many problems,such as traffic congestion and traffic environment pollution,among which environmental issues such as urban haze are particularly serious.Urban intersections have become the hardest-hit areas for urban traffic environment pollution.Pedestrians face the risk of high concentration exposure of fine particles when they frequently cross or stay at intersections.However,there are few reports on the spatial-temporal distribution characteristics of fine particle concentrations at intersections.Therefore,this article takes the Xi'an Xiao Zhai intersection as the research object,through the fixed-point monitoring experiment collection data,to explore the internal relations between the traffic,meteorology and other factors and the space-time distribution of fine particles,and provide scientific basis for the control strategy of vehicle exhaust emission,urban traffic emission reduction and air pollution reduction.The prime objective of this paper is to study the concentration distribution of fine particle in urban intersections and reduce the exposure of the fine particles for pedestrians in urban public areas.To achieve it,the study investigated the concentration distribution of the fine partticles with an abundant of field data at Xiaozhai signal intersection in Xi'an.A reliable fine particle concentration prediction model for the signal intersection,which is based on motor vehicle exhaust model MOVES and diffusion model CAL3 QHC,was proposed,calibrated and validated with real-world data,which has the potential to serve as a practical tool for simulating the concentration distribution of the fine particles at signal intersection.In addition,an artificial neural network model was proposed to predict the concentration of fine particles with the experimental data in the intersection.The main research work of this paper is as follows:Firstly,in order to study the exhaust emission and diffusion of motor vehicles more subtly,this paper selects MOVES model based on driving conditions as the basic calculation principle and the vehicle exhaust diffusion model CAL3 QHC suitable for signalized intersections.By analyzing the structure and principle of the emission model MOVES and the diffusion model CAL3 QHC,the two models are matched and integrated,and a prediction model of fine particle concentration is established.The basic principle of this model is to use the MOVES model to output the vehicle exhaust emission factors in the actual road network,and to use the input data of the CAL3 QHC model to calculate the fine particle diffusion concentration.Secondly,the Xiaozhai intersection in Xi'an was selected as the research object,and the proposed fine-particle concentration prediction model was used to simulate the fine-particle concentration distribution at the intersection.Through the localized modified emission model,the emissions and emission factors at the intersection were calculated,the highest concentration of particulate matter at the south entrance of the intersection was obtained,followed by the north,west,and east inlets;the diffusion model CAL3 QHC was used to simulate the diffusion characteristics of the motor vehicle exhaust around the intersection.The results of the study showed that the concentration of fine particles diffused at the southeast and northeast corners of the intersection was relatively high,which was consistent with the results of the measured values.Through comparison with measured values,the results show that the CAL3 QHC model has a prediction error of 5.1% when the background concentration of fine particles is low.Finally,through the analysis of fixed-point monitoring particle concentration data,the results show that the southeast and northeast corners of the intersection are the locations with the highest concentration of fine particles,while the concentration of fine particulate matter is accumulated during the period from the early rush hour to the early morning peak period,and the average concentration of PM2.5 in the week is from 73.22 ?g/m3 to 80.31 ?g/m3.The concentration of particulate matter at the weekend decreased significantly compared with the weekday,and the concentration of PM2.5 decreased by 40%.In addition,in order to overcome the deficiency of the current pollutant diffusion concentration prediction model CAL3 QHC,and based on the measured data,the corresponding BP neural network and wavelet neural network model were established to predict the concentration of fine particles in the intersection,and the accuracy of the two neural network models is verified by the measured values.The results show that the wavelet neural network is used,compared with the traditional BP neural network model,the prediction accuracy is better,in which the prediction error of the wavelet neural network model is 4.61%,while the BP neural network model is 5.55%.The prediction results of the CAL3 QHC model are compared and analyzed at the same time,and considering the shortcoming of BP neural network model such as local optimum and over fitting,it is concluded that the wavelet neural network is more generalized in application by compared and analyzed with each other.