Investigation of artificial neural network performance in the aerosol properties retrieval

Aerosols are an integral part of Earth's climate system and their effect on climate makes this field a relevant research problem. The artificial neural network (ANN) technique is an upcoming technique in different research fields. In the current work, we have evaluated the performance of an ANN...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Nishi Srivastava, D. Vignesh, Nisheeth Saxena
Formato:	article
Lenguaje:	EN
Publicado:	IWA Publishing 2021
Materias:	aeronet aerosols aod artificial neural network Environmental technology. Sanitary engineering TD1-1066 Environmental sciences GE1-350
Acceso en línea:	https://doaj.org/article/f1833868710d4a6fbd5c6c1abbcf01ed
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:f1833868710d4a6fbd5c6c1abbcf01ed
record_format	dspace
spelling	oai:doaj.org-article:f1833868710d4a6fbd5c6c1abbcf01ed2021-11-05T19:08:08ZInvestigation of artificial neural network performance in the aerosol properties retrieval2040-22442408-935410.2166/wcc.2021.336https://doaj.org/article/f1833868710d4a6fbd5c6c1abbcf01ed2021-09-01T00:00:00Zhttp://jwcc.iwaponline.com/content/12/6/2814https://doaj.org/toc/2040-2244https://doaj.org/toc/2408-9354Aerosols are an integral part of Earth's climate system and their effect on climate makes this field a relevant research problem. The artificial neural network (ANN) technique is an upcoming technique in different research fields. In the current work, we have evaluated the performance of an ANN with its parameters in simulating the aerosol's properties. ANN evaluation is performed over three sites (Kanpur, Jaipur, and Gandhi College) in the Indian region. We evaluated the performance of ANN for model's hyperparameter (number of hidden layers) and optimizer's hyperparameters (learning rate and number of iterations). The optical properties of aerosols from AERONET (AErosol RObotic NETwork) are used as input to ANN to estimate the aerosol optical depth (AOD) and Angstrom exponent. Results emphasized the need for optimal learning rate values and the number of iterations to get accurate results with low computational cost and to avoid overfitting. We observed a 23–25% increase in computational time with an increase in iteration. Thus, a meticulous selection of these parameters should be made for accurate estimations. The result indicates that the developed ANN can be utilized to derive AOD, which is not assessed at AERONET stations. HIGHLIGHTS In designing an ANN, we must choose the optimal number of iterations based on computational cost and quality of results.; Our finding indicates that ANN with more hidden layers can perform reasonably well at a low number of iterations.; The specific site may need a different set of hyperparameters for the best performance of the ANN.; The developed ANN can be utilized to derive AOD, which is not assessed at AERONET stations;Nishi SrivastavaD. VigneshNisheeth SaxenaIWA Publishingarticleaeronetaerosolsaodartificial neural networkEnvironmental technology. Sanitary engineeringTD1-1066Environmental sciencesGE1-350ENJournal of Water and Climate Change, Vol 12, Iss 6, Pp 2814-2834 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	aeronet aerosols aod artificial neural network Environmental technology. Sanitary engineering TD1-1066 Environmental sciences GE1-350
spellingShingle	aeronet aerosols aod artificial neural network Environmental technology. Sanitary engineering TD1-1066 Environmental sciences GE1-350 Nishi Srivastava D. Vignesh Nisheeth Saxena Investigation of artificial neural network performance in the aerosol properties retrieval
description	Aerosols are an integral part of Earth's climate system and their effect on climate makes this field a relevant research problem. The artificial neural network (ANN) technique is an upcoming technique in different research fields. In the current work, we have evaluated the performance of an ANN with its parameters in simulating the aerosol's properties. ANN evaluation is performed over three sites (Kanpur, Jaipur, and Gandhi College) in the Indian region. We evaluated the performance of ANN for model's hyperparameter (number of hidden layers) and optimizer's hyperparameters (learning rate and number of iterations). The optical properties of aerosols from AERONET (AErosol RObotic NETwork) are used as input to ANN to estimate the aerosol optical depth (AOD) and Angstrom exponent. Results emphasized the need for optimal learning rate values and the number of iterations to get accurate results with low computational cost and to avoid overfitting. We observed a 23–25% increase in computational time with an increase in iteration. Thus, a meticulous selection of these parameters should be made for accurate estimations. The result indicates that the developed ANN can be utilized to derive AOD, which is not assessed at AERONET stations. HIGHLIGHTS In designing an ANN, we must choose the optimal number of iterations based on computational cost and quality of results.; Our finding indicates that ANN with more hidden layers can perform reasonably well at a low number of iterations.; The specific site may need a different set of hyperparameters for the best performance of the ANN.; The developed ANN can be utilized to derive AOD, which is not assessed at AERONET stations;
format	article
author	Nishi Srivastava D. Vignesh Nisheeth Saxena
author_facet	Nishi Srivastava D. Vignesh Nisheeth Saxena
author_sort	Nishi Srivastava
title	Investigation of artificial neural network performance in the aerosol properties retrieval
title_short	Investigation of artificial neural network performance in the aerosol properties retrieval
title_full	Investigation of artificial neural network performance in the aerosol properties retrieval
title_fullStr	Investigation of artificial neural network performance in the aerosol properties retrieval
title_full_unstemmed	Investigation of artificial neural network performance in the aerosol properties retrieval
title_sort	investigation of artificial neural network performance in the aerosol properties retrieval
publisher	IWA Publishing
publishDate	2021
url	https://doaj.org/article/f1833868710d4a6fbd5c6c1abbcf01ed
work_keys_str_mv	AT nishisrivastava investigationofartificialneuralnetworkperformanceintheaerosolpropertiesretrieval AT dvignesh investigationofartificialneuralnetworkperformanceintheaerosolpropertiesretrieval AT nisheethsaxena investigationofartificialneuralnetworkperformanceintheaerosolpropertiesretrieval
_version_	1718444076557139968

Investigation of artificial neural network performance in the aerosol properties retrieval

Ejemplares similares