Study On The Vehicle Emission Regularity Of The Bridge Intersection Considering The Traffic Operation Factors

In the process of economic and social development,the call for"green transportation" has become increasingly popular,and the research on traffic emissions has become increasingly hot,and the research on the road network based on environmental factors has seen the results,but the research on intersection needs to be improved.As the throat of the urban road network,various operating conditions are changed repeatedly in the intersection,and its exhaust emission is significantly greater than that of the road.With the bridge becoming an important component of the urban road network,the research on the bridge intersection should be put on the agenda.Therefore,it is an inevitable trend that the research on traffic operation and emission of bridge intersection.First of all,the current research on exhaust emissions is elaborated from the aspects of model establishment and simulation,theoretical analysis and field measurement,the research status of traffic particulate matter(PM)regularity is clarified from its own character,road traffic influence factors,meteorological environment influence factors and spati-otemporal distribution characteristic,then the research content of bridge intersection is clarified.Secondly,the intersection of Juyuanzhou bridge(Jianping road)and Qishan avenue is selected as the test site;PMi,PM2.5,PM4,PM10 and TSP is selected as the subjects;traffic volume,queue length,headway and delay are the traffic operation factors;luminosity,temperature,noise,humidity and wind speed are the meteorological environmental factors;the survey points has been rationally designed in order to collect data;and the relationship between the entrance and exit routes/the location of the stop line and the concentration of PM concentration has been investigated.The research results show that the relationship between PM and influenc-ing factors can be considered mainly in the location of canalizetion island/stop line in the inlet channelThirdly,based on preliminary fitting analysis-simple correlation analysis-partial correlation analysis and cluster analysis,the relationships have been studied from the 3 aspects:the PMs,the influence factors and betweenthe the influence factors and PMs.The results show that accord-ing to the fitting rule analysis between the single factor and PM,the clustering analysis tree diagram about PMs,the five selected PMs have been classified as PM1,PM2.5 and PM4,and PM10 and TSP;there is a certain degree of correlation between the selected influencing factors,and the correlation between traffic operation factors/meteorologycal environ-mental factors is more obvious;in the relationship between the influenc-ing factors and the PMs,with the single factor fitting analysis,the positive effects include flow,headway,queue length,luminosity and noise,the negative correlation include delay and wind speed,but temperature and humidity have different effects on different PMs so that the relationships should be classified analysis,what's more,the smaller PMs' size is more obviously affected by traffic factors,such as traffic volume,delay,queue length and noise.Finally,under the foregoing analysis,the important factors have been selected by cluster analysis to eliminate the collinear,by using the multi-variate linear stepwise regression analysis,the variables what the model needs have been finalized,then the BP neural network models could be established.The results show that the advantage of cluster analysis is from data,and the consideration of the practical significance does not affect the classification of the influence factors of the bridge intersection;compared with the multiple linear regression model,BP neural network model has been improved.