Artificial intelligence modelling approach for the prediction of CO-rich hydrogen production rate from methane dry reforming

Bamidele Victor Ayodele, Siti Indati Mustapa, May Ali Alsaffar, Chin Kui Cheng

Research output: Contribution to journalArticle

Abstract

This study investigates the applicability of the Leven–Marquardt algorithm, Bayesian regularization, and a scaled conjugate gradient algorithm as training algorithms for an artificial neural network (ANN) predictively modeling the rate of CO and H2 production by methane dry reforming over a Co/Pr2O3 catalyst. The dataset employed for the ANN modeling was obtained using a central composite experimental design. The input parameters consisted of CH4 partial pressure, CO2 partial pressure, and reaction temperature, while the target parameters included the rate of CO and H2 production. A neural network architecture of 3 13 2, 3 15 2, and 3 15 2 representing the input layer, hidden neuron layer, and target (output) layer were employed for the Leven–Marquardt, Bayesian regularization, and scaled conjugate gradient training algorithms, respectively. The ANN training with each of the algorithms resulted in an accurate prediction of the rate of CO and H2 production. The best prediction was, however, obtained using the Bayesian regularization algorithm with the lowest standard error of estimates (SEE). The high values of coefficient of determination (R2 > 0.9) obtained from the parity plots are an indication that the predicted rates of CO and H2 production were strongly correlated with the observed values.

Original languageEnglish
Article number738
JournalCatalysts
Volume9
Issue number9
DOIs
Publication statusPublished - Sep 2019

Fingerprint

artificial intelligence
Methane
hydrogen production
Carbon Monoxide
Reforming reactions
Hydrogen production
Artificial intelligence
methane
predictions
Neural networks
education
Partial pressure
partial pressure
gradients
Network architecture
neurons
Design of experiments
Neurons
parity
indication

All Science Journal Classification (ASJC) codes

  • Catalysis
  • Physical and Theoretical Chemistry

Cite this

@article{85921b3d6d044415836718195cf55415,
title = "Artificial intelligence modelling approach for the prediction of CO-rich hydrogen production rate from methane dry reforming",
abstract = "This study investigates the applicability of the Leven–Marquardt algorithm, Bayesian regularization, and a scaled conjugate gradient algorithm as training algorithms for an artificial neural network (ANN) predictively modeling the rate of CO and H2 production by methane dry reforming over a Co/Pr2O3 catalyst. The dataset employed for the ANN modeling was obtained using a central composite experimental design. The input parameters consisted of CH4 partial pressure, CO2 partial pressure, and reaction temperature, while the target parameters included the rate of CO and H2 production. A neural network architecture of 3 13 2, 3 15 2, and 3 15 2 representing the input layer, hidden neuron layer, and target (output) layer were employed for the Leven–Marquardt, Bayesian regularization, and scaled conjugate gradient training algorithms, respectively. The ANN training with each of the algorithms resulted in an accurate prediction of the rate of CO and H2 production. The best prediction was, however, obtained using the Bayesian regularization algorithm with the lowest standard error of estimates (SEE). The high values of coefficient of determination (R2 > 0.9) obtained from the parity plots are an indication that the predicted rates of CO and H2 production were strongly correlated with the observed values.",
author = "Ayodele, {Bamidele Victor} and Mustapa, {Siti Indati} and Alsaffar, {May Ali} and Cheng, {Chin Kui}",
year = "2019",
month = "9",
doi = "10.3390/catal9090738",
language = "English",
volume = "9",
journal = "Catalysts",
issn = "2073-4344",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "9",

}

Artificial intelligence modelling approach for the prediction of CO-rich hydrogen production rate from methane dry reforming. / Ayodele, Bamidele Victor; Mustapa, Siti Indati; Alsaffar, May Ali; Cheng, Chin Kui.

In: Catalysts, Vol. 9, No. 9, 738, 09.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Artificial intelligence modelling approach for the prediction of CO-rich hydrogen production rate from methane dry reforming

AU - Ayodele, Bamidele Victor

AU - Mustapa, Siti Indati

AU - Alsaffar, May Ali

AU - Cheng, Chin Kui

PY - 2019/9

Y1 - 2019/9

N2 - This study investigates the applicability of the Leven–Marquardt algorithm, Bayesian regularization, and a scaled conjugate gradient algorithm as training algorithms for an artificial neural network (ANN) predictively modeling the rate of CO and H2 production by methane dry reforming over a Co/Pr2O3 catalyst. The dataset employed for the ANN modeling was obtained using a central composite experimental design. The input parameters consisted of CH4 partial pressure, CO2 partial pressure, and reaction temperature, while the target parameters included the rate of CO and H2 production. A neural network architecture of 3 13 2, 3 15 2, and 3 15 2 representing the input layer, hidden neuron layer, and target (output) layer were employed for the Leven–Marquardt, Bayesian regularization, and scaled conjugate gradient training algorithms, respectively. The ANN training with each of the algorithms resulted in an accurate prediction of the rate of CO and H2 production. The best prediction was, however, obtained using the Bayesian regularization algorithm with the lowest standard error of estimates (SEE). The high values of coefficient of determination (R2 > 0.9) obtained from the parity plots are an indication that the predicted rates of CO and H2 production were strongly correlated with the observed values.

AB - This study investigates the applicability of the Leven–Marquardt algorithm, Bayesian regularization, and a scaled conjugate gradient algorithm as training algorithms for an artificial neural network (ANN) predictively modeling the rate of CO and H2 production by methane dry reforming over a Co/Pr2O3 catalyst. The dataset employed for the ANN modeling was obtained using a central composite experimental design. The input parameters consisted of CH4 partial pressure, CO2 partial pressure, and reaction temperature, while the target parameters included the rate of CO and H2 production. A neural network architecture of 3 13 2, 3 15 2, and 3 15 2 representing the input layer, hidden neuron layer, and target (output) layer were employed for the Leven–Marquardt, Bayesian regularization, and scaled conjugate gradient training algorithms, respectively. The ANN training with each of the algorithms resulted in an accurate prediction of the rate of CO and H2 production. The best prediction was, however, obtained using the Bayesian regularization algorithm with the lowest standard error of estimates (SEE). The high values of coefficient of determination (R2 > 0.9) obtained from the parity plots are an indication that the predicted rates of CO and H2 production were strongly correlated with the observed values.

UR - http://www.scopus.com/inward/record.url?scp=85073318624&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073318624&partnerID=8YFLogxK

U2 - 10.3390/catal9090738

DO - 10.3390/catal9090738

M3 - Article

AN - SCOPUS:85073318624

VL - 9

JO - Catalysts

JF - Catalysts

SN - 2073-4344

IS - 9

M1 - 738

ER -