| In recent years,as the problem of air pollution has become more and more serious,particulate pollution has become an important part of it,the monitoring requirements of particulates have become more and more vigorous and the monitoring requirements have gradually increased.However,with the advancement and deepening of air pollution monitoring and control,it is found that it is difficult for traditional air quality automatic monitoring stations to obtain data with high temporal and spatial resolution to meet the needs of refined management and control.The development and advancement of sensor technology makes it possible to monitor air pollution with high resolution.As one of the more supplementary monitoring methods,the particle sensor based on light scattering method has shown explosive growth.However,there is relatively little work on the performance evaluation of light scattering particle sensors at present,and the research work on specific applications is relatively lacking.This thesis first evaluates the performance of three types of partical sensors based on light scattering method which has a large share of the domestic environmental monitoring market,and studies the factors that affect the quality of sensor data and proposes sensor improvements.At the same time,in the main urban area of Jinan City,Shandong Province,taxis equipped with sensors were used to monitor the mobile particulate matter concentration online.Based on the monitoring data of 9 automatic ambient air quality monitoring standard stations in the main urban area,the distribution characteristics and influencing factors of road environmental particulate matter are studied under the background of big data.The main conclusions of the two works are as follows:(1)Through the evaluation and study of 3 different types of particulate matter sensors(10 units of each type,denoted as A,B,C),it is found that the correlation within the sensor category is above 0.9,the Inter-sensor variability of A,B,C is 30.19%,15.43%,and 28.96%respectively.The time series analysis of the particle concentration data shows that the sensor and the standard instrument can make the same response to the change of PM2.5 concentration in the environment,and the correlation coefficient R2 of the two data is 0.59,0.80,0.61,respectively,and the sensor data has a positive error relative to the standard instrument data.Through multi-parameter correlation analysis,it is found that temperature(T),humidity(RH)and PM2.5/PM10 will affect the quality of sensor data.The average absolute error(MAE)of PM2.5 concentration data ofA,B,and C three types of sensors are 23.31 μg m-3,10.14 μg m-3,28.17 μm-3.The normalized root mean square error(RMSE)is 25.80 μg m-3,14.01 μg m-3,32.98 μg m’3,and the accuracy(A%)is 51.39%,72.97%,46.51%,respectively.(2)In June 2020,a one-month mobile monitoring of the environmental quality of roads was carried out by means of taxis equipped with sensors in the main urban area of Jinan City,Shandong Province.Through analysis of high temporal and spatial resolution particulate matter concentration data,it was found that:the correlation coefficients of the particle concentration data between the sensor and the standard station are:PM2.5:R=0.816;PM10:R=0.922.The concentration of particulate matter in the main urban area presents a distribution characteristic of "high in the north and low in the south" and has a diurnal change trend.The concentration of particulate matter presents different changing trends under different working conditions,and the change of particulate matter concentration is related to human activities and meteorological factors.At the same time,the diurnal trends of different road types(main roads,secondary roads,branch roads,other roads)indicate that the occurrence of peaks is closely related to traffic flow and human activities.Concentration differences in different administrative regions and four-level roads can explain the distribution characteristics of particulate matter concentration "high in the north and low in the south". |
PDF Full Text Request