Please use this identifier to cite or link to this item:
Title: Modeling of surface dust concentration in snow cover at industrial area using neural networks and kriging
Authors: Sergeev, A. P.
Tarasov, D. A.
Buevich, A. G.
Shichkin, A. V.
Tyagunov, A. G.
Medvedev, A. N.
Issue Date: 2017
Publisher: American Institute of Physics Inc.
Citation: Modeling of surface dust concentration in snow cover at industrial area using neural networks and kriging / A. P. Sergeev, D. A. Tarasov, A. G. Buevich, A. V. Shichkin, et al. . — DOI 10.1063/1.4981973 // AIP Conference Proceedings. — 2017. — Iss. 1836. — 20033.
Abstract: Modeling of spatial distribution of pollutants in the urbanized territories is difficult, especially if there are multiple emission sources. When monitoring such territories, it is often impossible to arrange the necessary detailed sampling. Because of this, the usual methods of analysis and forecasting based on geostatistics are often less effective. Approaches based on artificial neural networks (ANNs) demonstrate the best results under these circumstances. This study compares two models based on ANNs, which are multilayer perceptron (MLP) and generalized regression neural networks (GRNNs) with the base geostatistical method-kriging. Models of the spatial dust distribution in the snow cover around the existing copper quarry and in the area of emissions of a nickel factory were created. To assess the effectiveness of the models three indices were used: the mean absolute error (MAE), the root-mean-square error (RMSE), and the relative root-mean-square error (RRMSE). Taking into account all indices the model of GRNN proved to be the most accurate which included coordinates of the sampling points and the distance to the likely emission source as input parameters for the modeling. Maps of spatial dust distribution in the snow cover were created in the study area. It has been shown that the models based on ANNs were more accurate than the kriging, particularly in the context of a limited data set. © 2017 Author(s).
Access: info:eu-repo/semantics/openAccess
SCOPUS ID: 85021324104
WOS ID: 000409539000033
PURE ID: 1931259
ISSN: 0094-243X
ISBN: 9780735415065
DOI: 10.1063/1.4981973
Appears in Collections:Научные публикации ученых УрФУ, проиндексированные в SCOPUS и WoS CC

Files in This Item:
File Description SizeFormat 
10.1063-1.4981973.pdf759,94 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.