TITLE:
Design of a Portable Digester and Prediction of Biogas Yield Using Artificial Neural Network
AUTHORS:
Temilola Taiwo Olugasa, Solomon Oladapo Olaniyan, Moses Omolayo Petinrin, Olanrewaju Miracle Oyewola
KEYWORDS:
Biodigester, Optimisation, Model, Training Algorithm, Yield Prediction
JOURNAL NAME:
Open Access Library Journal,
Vol.13 No.2,
February
12,
2026
ABSTRACT: Biogas generation through the anaerobic digestion of organic matter is one of the crucial technology interventions that brings about the transformation of the fossil fuel dependent energy system to a renewable energy based one. Biogas production needs further development and optimisation for the technical, economic, and environmental aspects to be fully marketable and economical. Thus, a broad knowledge of the reaction kinetics involved in the breaking down of organic matter by microbes into biogas and the effect of the fluid dynamics in the process of digestion pertinent to model, predict control biogas production accurately and effectively. A four-wheeled portable digester was developed from a 63 Litre drum and biogas was generated from cow dung at a retention time of 21 days. The digestion process was monitored by means of data loggers and sensors. Data of pressure, temperature, PH, volume of gas generated using a data logger, and biogas yield was modelled and predicted using Artificial Neural Network. The performance of the model was explored using Levenberg-Marquardt, Bayesian Regularisation and Scaled Conjugate Gradient training algorithms, with 10, 15 and 20 hidden layers. The Artificial Neural Network predicted biogas yield to high degree of accuracy. The Levenberg-Marquardt algorithm had the highest R value of 0.9999.