Open AccessDiffusion Test and Prediction of PM2.5 Concentration in Urban Street Traffic Microenvironment
Author(s) -
Yanshu Ni,
Lianglin Ma
Publication year - 2020
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.179
H-Index - 26
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/585/1/012084
Subject(s) - artificial neural network , matlab , environmental science , simulation , meteorology , reliability (semiconductor) , traffic flow (computer networking) , diffusion , test data , computer science , transport engineering , engineering , geography , artificial intelligence , power (physics) , physics , computer security , quantum mechanics , programming language , operating system , thermodynamics
A detailed test plan was developed by analyzing the factors affecting PM 2.5 diffusion in the urban street traffic microenvironment. After field investigation, several test streets within the third Ring Road of Harbin city were selected to investigate and test the hourly traffic flow of the streets, and the geometric structure of the block (road width, building height) was measured. The research carried out tests on the PM 2.5 concentration, wind speed, temperature and relative humidity of actual street test points. Based on the actual test data, BP artificial neural network and the improved LMBP neural network were used to carry out simulation research on MATLAB platform respectively. A PM 2.5 concentration prediction model was established to compare and analyze the reliability of the model. Scientific and reasonable prediction of PM 2.5 pollution in the traffic microenvironment in a specific area.