Improving the accuracy of the calculated environmental monitoring of atmospheric air quality in cities based on a corrective neural network

To obtain the values of the concentrations of atmospheric impurities with a spatial-temporal resolution throughout the city, which is not achieved during systematic experimental observations, it is recommended to use calculation methods. A significant advantage of the calculated concentrations, in c...

Full description

Saved in:
Bibliographic Details
Published in:IOP conference series. Earth and environmental science 2021-09, Vol.864 (1), p.12029
Main Authors: Tunakova, Y A, Novikova, S V, Shagidullin, A R, Valiev, V S, Novikova, K N
Format: Article
Language:English
Subjects:
Air monitoring
Air pollution
Air quality
Environmental monitoring
Impurities
Mathematical analysis
Natural resources
Neural networks
Outdoor air quality
Pollutants
Pollution sources
Temporal resolution
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To obtain the values of the concentrations of atmospheric impurities with a spatial-temporal resolution throughout the city, which is not achieved during systematic experimental observations, it is recommended to use calculation methods. A significant advantage of the calculated concentrations, in contrast to the experimental ones, is the ability to determine the concentration at any point in the study area at the present time and in the future when the characteristics of stationary pollution sources change. To calculate the dispersion of emissions of pollutants in the air and determine the surface concentrations of impurities, the order of the Ministry of Natural Resources and Environment of the Russian Federation regulates the use of the universal program for calculating atmospheric pollution (UPCAP) “Ecolog”. To increase the convergence of the measured concentrations of impurities and calculated concentrations obtained using UPCAP “Ecolog-Gorod” we propose the use of neural network technologies.
ISSN:1755-1307
1755-1315
DOI:10.1088/1755-1315/864/1/012029