Research And Application Of PM Concentration Spatio-temporal Model In Micro Area

Particulate matter pollution is becoming more and more serious,and the existing PM concentration data are mostly obtained by macroscopic monitoring.that ignored the micro-regional concentration changes.However,the micro-regional environmental quality is closely related to people’s health.In this study,a set of remote environmental air monitoring system is designed for monitoring micro-regional air pollution,and it records the micro-regional PM concentration changes by real-time monitoring parameters.Moreover,a Kalman prediction model combined with cubic spline interpolation is proposed for concentration prediction and analysis.The main contributions of this thesis are shown as follows:1.The current status of ambient air information monitoring is analyzed.We decide to design an intelligent,real-time and integrated low-cost PM concentration monitoring system.It reflects the ambient area and the specific area of air information from the microscopic.In order to study micro-regional PM concentration distribution and its impact,an appropriate space-time model is established.2.A solar-powered environment monitoring box based on GPRS wireless network is developed.At the same time,the remote micro-area environment air real-time monitoring system platform is designed based on RTDBMS system architecture.The system includes the subsystem,the communication network,system back-end,the scheduling center and the application terminal.In our experiment,the campus area was taken as a micro region,and the 15 environmental monitoring boxes were arranged according to the topography of Chongqing University of Posts and Telecommunications(CQUPT),and the data of PM concentration was collected for one year.3.An improved Kalman prediction model based on cubic spline interpolation is proposed to analyze and simulate the changes of campus PM concentration.The theme is illustrated by the example of monitoring point hourly average and daily mean data.The mean absolute error of the daily average forecast results is 1.2μg/m3,the mean relative error is 0.16 and the correlation coefficient is 0.87.Through a number of experimental comparative analysis,the results show that the error and relative error of predicted are small,and the correlation coefficient is large.Moreover,the results prove the effectiveness of the model in the prediction of PM concentration,and superiority of simulation for the spatial distribution of PM and local pollution characteristicsThe thesis realizes the design of micro-regional ambient air monitoring system.An improved Kalman prediction model based on cubic spline interpolation is established.The concentration of PM in the four seasons of the campus was predicted and analyzed by the model,and the model graphically describes the temporal and spatial distribution of PM and its regularity in the CQUPT campus.