Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal International Journal of Multidisciplinary Approach Research and Science
Usman, Abdulgafar
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Chemical Engineering, Chemistry & Bioengineering Economics, Econometrics & Finance Immunology & microbiology Social Sciences

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Forecasting of Biogas and Biomethane Outputs from Anaerobic Co-digestion Using Multilayer Perceptron Artificial Neural Networks (MLP-ANN) Samuel, D.A.; Eziefula, B.I.; Orkuma, Gbushum; Usman, Abdulgafar
International Journal of Multidisciplinary Approach Research and Science Том 3 № 02 (2025): International Journal of Multidisciplinary Approach Research and Science
Publisher : PT. Riset Press International

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59653/ijmars.v3i02.1529

Abstract

The intricate, nonlinear interactions between several process parameters make it difficult to accurately forecast biogas and methane (CH₄) yields in anaerobic digestion systems. Conventional kinetic models frequently do not capture these dynamics well, particularly in co-digestion systems with heterogeneous substrates. This was addressed by developing and testing a Multilayer Perceptron Artificial Neural Network (MLP-ANN) for predicting cumulative biogas and CH₄ production from the anaerobic co-digestion of soymilk dregs and cow manure. Different pH values, substrate mixing ratios, and hydraulic retention durations were used to operate batch-mode digesters. Three input parameters, one hidden layer with three neurons, and two output nodes that represented the yields of methane (CH4) and biogas made up the feed-forward ANN architecture. The result showed negligible prediction errors and very high coefficients of determination (R2) of 0.999 for both outputs, and the model exhibited exceptionally accurate predictive power. The average relative errors for the training and testing stages were 0.002 and 0.004, respectively, confirming the model's excellent generalization capabilities. These findings support the optimization of anaerobic digestion systems by validating the MLP-ANN as a reliable and efficient forecasting method. This MLP-ANN can be used for forecasting biogas and CH₄ outputs based on pH values, substrate mixing ratios, and hydraulic retention times.
Co-Authors Eziefula, B.I. Orkuma, Gbushum Samuel, D.A.