Study On The Influence Of Multidimensional Traffic Flow Based On Urban Road On Air Quality

At present,the pollution problem of mobile sources is increasingly prominent in China.Jinan,as the capital city of shandong province,is an important transportation hub of shandong peninsula.By the end of 2018,the number of motor vehicles in jinan has exceeded 2.3 million,with the highest level of new registrations and annual increment.Air pollution types in jinan city are gradually transforming from the traditional"coal smoke"pollution to the composite pollution of"coal smoke+motor vehicle emission".Motor vehicle exhaust pollution is becoming one of the main"culprits"affecting air quality in jinan.This paper analyzes the temporal and spatial changes of the main road traffic flow and air pollution indexes in jinan city,and carries out grey correlation analysis on the sample data to explore the impact of vehicle exhaust on urban air quality.In order to increase the interpretability of the prediction model,meteorological factors were added into the characteristic factors,real-time traffic data and meteorological data were used,and integrated learning algorithm was used to predict road air pollution more accurately by combining the monitoring data of air quality ground monitoring stations.Finally,the classification prediction model was established by using the integrated learning algorithm,and the performance of the traditional machine learning algorithm and the integrated algorithm were compared and analyzed to provide the corresponding data support for the road traffic air pollution control and control in jinan city,so as to realize the goal of improving the urban air quality from the mobile pollution source.At the same time,because of considering,in addition to the polluted by motor vehicles,road air quality is affected by industrial pollution is also more obvious,therefore this article through the crawl city point of interest data judgment road near the factory point intensity,selecting the distribution in the plant roads and the distribution in the plant road two set of sample data,and carries on the grey relation analysis and prediction of the results by comparing the two groups,with analysis of industrial pollution on the degree of interference in this paper,the research content.It is found that traffic factors and meteorological factors have a high correlation degree with urban road air quality,and the influence degree of vehicle flow factors on urban road air quality is greater than that of vehicle speed.Among the meteorological factors,the wind speed has the largest influence,and the temperature has a significant influence on O₃₁.The variation of O₃₁ concentration has a significant seasonal trend.Based on the comprehensive score of correlation degree,it is found that O₃₁ is the pollutant most affected by traffic factors.In the prediction analysis,it is found that the model performance of integrated algorithm is generally better,among which XGBoost algorithm is the best.In practice,it is suggested that the environmental protection department should access the traffic flow and meteorological data to predict the O₃₁pollution status at each time of the next day.In the future,the prevention and control of road air pollution in jinan should focus on O₃₁,especially in the summer season with high occurrence of O₃₁ pollution,and the road traffic flow should be controlled by different models in different periods.Relevant departments can also make timely prevention strategies based on the prediction of road O₃₁ pollution by the model,and suggest that people reduce their travel from afternoon to evening and reasonably arrange their travel time.