A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm

Abstract In this study, GMDH neural network based on genetic algorithm was used to predict the physical and mechanical properties of laboratory made particleboard. To predict the mechanical and physical properties of particleboard we used input parameters such as neural network including press clos...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Zahra Jahanilomer, Saeed Reza farrokhpayam, Mohammad Shamsian
Formato: article
Lenguaje:FA
Publicado: Regional Information Center for Science and Technology (RICeST) 2014
Materias:
Acceso en línea:https://doaj.org/article/eeb2d32a64cc4842afc09778291d3542
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:eeb2d32a64cc4842afc09778291d3542
record_format dspace
spelling oai:doaj.org-article:eeb2d32a64cc4842afc09778291d35422021-12-02T03:52:28ZA mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm1735-09132383-112X10.22092/ijwpr.2014.6130https://doaj.org/article/eeb2d32a64cc4842afc09778291d35422014-09-01T00:00:00Zhttp://ijwpr.areeo.ac.ir/article_6130_301e53ab2aa4650e82fcb5c248734664.pdfhttps://doaj.org/toc/1735-0913https://doaj.org/toc/2383-112XAbstract In this study, GMDH neural network based on genetic algorithm was used to predict the physical and mechanical properties of laboratory made particleboard. To predict the mechanical and physical properties of particleboard we used input parameters such as neural network including press closing time (10,20 and 30 seconds), moisture content of the mat (8,10,12 and 14%) and press temperature (150,160,170 and 180°C) as the input data and the output data was the physical and mechanical properties. The efficiency of these techniques was evaluated with statistical criteria of mean square error (MSE), root mean square error, (RMSE), mean absolute deviation (MAD) and the correlation coefficient (R2). Results showed that the value of MSE, RMSE and MAD for MOR, IB, TS24h, TS2h, WA2h and WA24h is low. Errors obtained for the MOE model were very high. According to the results obtained, this model is not the appropriate for prediction of MOE. R2 values from the test and training set properties for MOR, IB, MOE, TS24h, TS2h, WA2h and WA24hwas more than 0.91%, which reflects that the performance of these models is better.Zahra JahanilomerSaeed Reza farrokhpayamMohammad ShamsianRegional Information Center for Science and Technology (RICeST)articleKeywords: ParticleboardmodelingGMDH-Type Neural NetwokPhysical Mechanical PropertiesForestrySD1-669.5FAتحقیقات علوم چوب و کاغذ ایران, Vol 29, Iss 3, Pp 376-389 (2014)
institution DOAJ
collection DOAJ
language FA
topic Keywords: Particleboard
modeling
GMDH-Type Neural Netwok
Physical Mechanical Properties
Forestry
SD1-669.5
spellingShingle Keywords: Particleboard
modeling
GMDH-Type Neural Netwok
Physical Mechanical Properties
Forestry
SD1-669.5
Zahra Jahanilomer
Saeed Reza farrokhpayam
Mohammad Shamsian
A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
description Abstract In this study, GMDH neural network based on genetic algorithm was used to predict the physical and mechanical properties of laboratory made particleboard. To predict the mechanical and physical properties of particleboard we used input parameters such as neural network including press closing time (10,20 and 30 seconds), moisture content of the mat (8,10,12 and 14%) and press temperature (150,160,170 and 180°C) as the input data and the output data was the physical and mechanical properties. The efficiency of these techniques was evaluated with statistical criteria of mean square error (MSE), root mean square error, (RMSE), mean absolute deviation (MAD) and the correlation coefficient (R2). Results showed that the value of MSE, RMSE and MAD for MOR, IB, TS24h, TS2h, WA2h and WA24h is low. Errors obtained for the MOE model were very high. According to the results obtained, this model is not the appropriate for prediction of MOE. R2 values from the test and training set properties for MOR, IB, MOE, TS24h, TS2h, WA2h and WA24hwas more than 0.91%, which reflects that the performance of these models is better.
format article
author Zahra Jahanilomer
Saeed Reza farrokhpayam
Mohammad Shamsian
author_facet Zahra Jahanilomer
Saeed Reza farrokhpayam
Mohammad Shamsian
author_sort Zahra Jahanilomer
title A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
title_short A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
title_full A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
title_fullStr A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
title_full_unstemmed A mathematical model to predict particleboard properties using the GMDH-type neural network and genetic algorithm
title_sort mathematical model to predict particleboard properties using the gmdh-type neural network and genetic algorithm
publisher Regional Information Center for Science and Technology (RICeST)
publishDate 2014
url https://doaj.org/article/eeb2d32a64cc4842afc09778291d3542
work_keys_str_mv AT zahrajahanilomer amathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
AT saeedrezafarrokhpayam amathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
AT mohammadshamsian amathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
AT zahrajahanilomer mathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
AT saeedrezafarrokhpayam mathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
AT mohammadshamsian mathematicalmodeltopredictparticleboardpropertiesusingthegmdhtypeneuralnetworkandgeneticalgorithm
_version_ 1718401620257013760