Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study

The global solar radiation prediction is the most necessary part of the project and performance of solar energy applications. The objective of the present work is to predict global solar radiation (GSR) received on the horizontal surface using an artificial neural network (ANN). For the city (Reliza...

Full description

Saved in:
Bibliographic Details
Main Authors: Dahmani Abdennasser, Yamina Ammi, Salah Hanini
Format: Article
Language:English
Published: University of El Oued 2022-12-01
Series:International Journal of Energetica
Subjects:
Online Access:https://www.ijeca.info/index.php/IJECA/article/view/198
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849410069070872576
author Dahmani Abdennasser
Yamina Ammi
Salah Hanini
author_facet Dahmani Abdennasser
Yamina Ammi
Salah Hanini
author_sort Dahmani Abdennasser
collection DOAJ
description The global solar radiation prediction is the most necessary part of the project and performance of solar energy applications. The objective of the present work is to predict global solar radiation (GSR) received on the horizontal surface using an artificial neural network (ANN). For the city (Relizane) in the west region of Algeria. The inputs used in the neural network are: time (h), day, month, year, temperature (k), relative humidity (%), pressure (mbar), wind speed (m/s), wind direction (°), and rainfall (kg/m2). The neural network-optimal model was trained and tested using 80 %, and 20 % of whole data, respectively. The best results were obtained with the structure 10-25-1 (10 inputs, 25 hidden, and 1 output neurons) presented an excellent agreement between the calculated and the experimental data during the test stage with a correlation coefficient of R = 0.9879, root means squared error of RMSE = 47.7192 (Wh/m2), mean absolute error MAE = 27.7397 (Wh/m2), and mean squared error MSE = 2.2771e+03(Wh/m2), considering a three-layer Feedforward neural network with Regularization Bayesienne (trainbr)  training algorithm, a hyperbolic tangent sigmoid and linear transfer function at the hidden and the output layer, respectively. The results demonstrate proper ANN’s predictions with a root mean square error (RMSE) of less than 0.50 (Wh/m2) and coefficient of correlation (R) higher than 0.98, which can be considered very acceptable. This model can be used for designing solar energy systems in the hottest regions.
format Article
id doaj-art-7647002a6c394ec8981b9099e537e821
institution Kabale University
issn 2543-3717
language English
publishDate 2022-12-01
publisher University of El Oued
record_format Article
series International Journal of Energetica
spelling doaj-art-7647002a6c394ec8981b9099e537e8212025-08-20T03:35:16ZengUniversity of El OuedInternational Journal of Energetica2543-37172022-12-0172818120Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis studyDahmani Abdennasser0Yamina Ammi1Salah Hanini2Department of Mechanical Engineering, University Ahmad Zabana of RelizaneLaboratory of Biomaterials and Transport Phenomena (LBMPT), University of MedeaLaboratory of Biomaterials and Transport Phenomena (LBMPT), University of MedeaThe global solar radiation prediction is the most necessary part of the project and performance of solar energy applications. The objective of the present work is to predict global solar radiation (GSR) received on the horizontal surface using an artificial neural network (ANN). For the city (Relizane) in the west region of Algeria. The inputs used in the neural network are: time (h), day, month, year, temperature (k), relative humidity (%), pressure (mbar), wind speed (m/s), wind direction (°), and rainfall (kg/m2). The neural network-optimal model was trained and tested using 80 %, and 20 % of whole data, respectively. The best results were obtained with the structure 10-25-1 (10 inputs, 25 hidden, and 1 output neurons) presented an excellent agreement between the calculated and the experimental data during the test stage with a correlation coefficient of R = 0.9879, root means squared error of RMSE = 47.7192 (Wh/m2), mean absolute error MAE = 27.7397 (Wh/m2), and mean squared error MSE = 2.2771e+03(Wh/m2), considering a three-layer Feedforward neural network with Regularization Bayesienne (trainbr)  training algorithm, a hyperbolic tangent sigmoid and linear transfer function at the hidden and the output layer, respectively. The results demonstrate proper ANN’s predictions with a root mean square error (RMSE) of less than 0.50 (Wh/m2) and coefficient of correlation (R) higher than 0.98, which can be considered very acceptable. This model can be used for designing solar energy systems in the hottest regions.https://www.ijeca.info/index.php/IJECA/article/view/198prediction, global solar radiation, artificial neural networks, relizane
spellingShingle Dahmani Abdennasser
Yamina Ammi
Salah Hanini
Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
International Journal of Energetica
prediction, global solar radiation, artificial neural networks, relizane
title Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
title_full Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
title_fullStr Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
title_full_unstemmed Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
title_short Neural network for prediction solar radiation in Relizane region (Algeria) - Analysis study
title_sort neural network for prediction solar radiation in relizane region algeria analysis study
topic prediction, global solar radiation, artificial neural networks, relizane
url https://www.ijeca.info/index.php/IJECA/article/view/198
work_keys_str_mv AT dahmaniabdennasser neuralnetworkforpredictionsolarradiationinrelizaneregionalgeriaanalysisstudy
AT yaminaammi neuralnetworkforpredictionsolarradiationinrelizaneregionalgeriaanalysisstudy
AT salahhanini neuralnetworkforpredictionsolarradiationinrelizaneregionalgeriaanalysisstudy