Research On Air Pollutant Emission Of Passenger Cars Based On Multi-source Data

Passenger cars are irreplaceable in the urban transportation system due to the characteristics of fast,comfortable,personal,and door-to-door travel.The growing demand for passenger cars is impacted by car-purchases limitation,car-usage restrictions,and health considerations in special periods(smog,epidemic,etc.to avoid long-term outdoor exposure).Thus,the total number of passenger cars will show a steady upward trend,and the traffic-related pollution will not be ignored.Non-operating passenger cars have become the main type for traffic emissions due to the large number and unavoidable commuting demand.While operating passenger cars have higher per capita air pollution emissions because of the higher average daily mileage,more frequent acceleration,deceleration and non-operational idling.Therefore,the control on passenger cars-related emissions will become an important way to reduce traffic emissions,and it is also an effective method to alleviate air pollution and improve environmental quality.The primary issue in controlling passenger cars-related air pollution emissions is to measure their emissions levels,and then explore the deep causes of emissions.In previous studies,passenger car emissions used small sample for laboratory or road tests to directly monitor vehicle operation and pollutant emissions.With the updating of monitoring instruments and measurement methods,as well as the rise of Big Data processing technology,passenger cars' travel research based on trajectory data has developed rapidly.Moreover,combining with image visualization,basic data and visual platform for passenger cars emissions have been created.All above make measurement more accurate and can represent the overall emission level,without errors due to sample size and sampling problems.As for the deep causes of emissions,the existing researches mainly focus on micro-factors such as operating modes,however,have little to do with the macro-factors--traffic states.Therefore,this paper focuses on air pollutant emissions of passenger cars.The operating characteristics and pollutant emissions of operating and non-operating passenger cars are compared and calculated by using multi-source data including software simulation,road test and laboratory test.Then,the deep reason of emission and its change is explored,that is,the relationship with traffic state.The main content is as follows:Firstly,travel demand of passenger cars and the current status of air pollutant emissions are studied.Then,the impact of population,economy,energy and transportation on air pollution is modeled.The relationship between passenger cars and air pollution is clarified by the contribution factor of car number to pollutants.The results show that the number of passenger cars is the second influential factor,followed by economy.They also reveal the severity of passenger car emissions and the significance of emission calculation.Secondly,this study calculates emissions separately,according to the differences in operation and emission between operating and non-operating passenger cars.For operating passenger cars,Global Positioning System(GPS)trajectory data is used to characterize the operating state.Average speed is used as a link,and MOVES(Motor Vehicle Emission Simulator)software is used to simulate emissions.Thus meso-mean speed emission method and micro-mean speed emission method are established based on GPS raw data and MOVES localization correction.The results confirm that there are significant differences in the operating characteristics and emission levels between occupied state and vacant state,the working day and non-working day patterns are also observed.Then,the speed reconstruction of the GPS data is performed to obtain second by second data,and the MOVES micro-condition emission method and VSP(Vehicle Specific Power)distribution emission method are constructed.The studies show that there are peak hours for vehicle emission,and emission level during evening peak is higher than that of during morning peak hour.Moreover,emission factors of occupied/vacant states on different date types(working,weekend and holiday)are different.The results of the above four methods are different for operating passenger cars.The calculation of meso-mean speed method is the highest,besides,the micro-mean speed method is higher than the micro-condition emission method.For non-operating passenger cars,an urban tunnel test is conducted to infer the emission levels of total vehicles by measuring the pollutant concentrations at the entrance and exit of the tunnel,and then estimating the emission factors for different types of vehicles.At the same time,the Vehicle Mass Analysis System(VMAS)based on the chassis dynamometer is used to monitor the emissions of the test vehicles.Vehicles are firstly classified as low-,medium-,and high-emission through principal component analysis and K-means clustering.Then the operating conditions and emission characteristics are studied for each classification separately.Furthermore,the relative error of VMAS based on tunnel test results is analyzed.It is found that the emission factors of operating passenger cars are significantly higher than those of nonoperating passenger cars.The vehicle emission increases sharply and fluctuates strongly in the acceleration stage,which is closely related to the operating condition.The middle emission vehicles in the cluster are closer to the local running vehicles.The combination of the above two methods can reveal the meso and micro characteristics of emissions in a comprehensive way.Finally,above calculation method are compared,and a total emission measurement method for passenger cars based on multi-source data(GPS reconstruction speed,tunnel test,traffic roadside station,as well as road information)is proposed.The case study of Xi'an South Ring Road verifies the effectiveness of the method.In practical applications,various methods can be comprehensively used according to the research object,traffic data,etc.Based on multi-source data types,coupled with multi-information,localized traffic emissions measurement will be performed more accurately.At the same time,taking the average speed as a link and considering traffic states,the influence of traffic states on emissions is revealed through modeling of the relationship between emission factor and average speed,measuring emission benefit of traffic states improvement,and analyzing the trend between Traffic Performance Index(TPI)and emission.It is found that with the increase of travel speed,the emission factors first decrease rapidly and then slowly decrease to stable state,finally gradually increase at high speed zone.In addition,the improvement of traffic conditions has significant positive benefits in emission reduction,and the emissions of passenger cars also increase with the increase of TPI.The above results all indicate that the alleviation of traffic congestion is conducive to the reduction of pollutant emissions of passenger cars,revealing the driving influence of emissions with their changes and the necessity of traffic control.