A team of Nigerian researchers has developed an advanced computational framework that could improve renewable energy production from oil palm waste by combining anaerobic digestion experiments with artificial intelligence and optimisation technologies.
The study, published in Biomass Conversion and Biorefinery, focused on converting oil palm empty fruit bunch (OPEFB), one of the most abundant agricultural residues from the palm oil industry, into methane-rich biogas.
The research highlights how agricultural waste, often discarded or burned, can become a valuable resource for clean energy production, while supporting a circular bioeconomy.
Oil palm empty fruit bunches represent a major residue generated by palm oil processing. According to the study, large quantities of this lignocellulosic material are often burned or disposed of, creating environmental concerns and wasting potential bioenergy resources.
The researchers, led by a lecturer at the Federal College of Agriculture, Ibadan, Dr Idowu Olugbenga Adewumi, explored anaerobic digestion as a sustainable method for converting OPEFB into methane, a valuable renewable energy source.
However, the complex structure of lignocellulosic biomass makes efficient conversion challenging because lignin can restrict microbial access to cellulose and hemicellulose.
To overcome the challenges, the researchers developed an integrated framework combining experimental measurements, machine learning, explainable artificial intelligence and evolutionary optimisation.
The research analysed operational factors including temperature, pH, organic loading rate, hydraulic retention time and carbon-to-nitrogen ratio. The experiments produced 5,452 measured observations from laboratory-scale anaerobic digestion trials conducted between January and November 2025.
Methane yields recorded during the experiments ranged from 181.44 to 282.42 mL CH₄ g⁻¹ VS, with an average yield of 231.89 ± 15.22 mL CH₄ g⁻¹ VS.
The study found that temperature was the strongest predictor of methane production. Explainable artificial intelligence analysis showed temperature had the highest influence among the evaluated variables, followed by cellulose content.
The researchers explained that higher temperatures can enhance microbial activity, improve substrate breakdown and support methane-producing microorganisms.
Five predictive models were assessed in the study: Multiple Linear Regression (MLR), Artificial Neural Network (ANN), Random Forest (RF), Support Vector Regression (SVR) and Gradient Boosting Regression (GBR).
The results showed that Multiple Linear Regression achieved the strongest predictive performance with R² = 0.5513, RMSE of 10.14 mL CH₄ g⁻¹ VS, MAE of 8.17 mL CH₄ g⁻¹ VS and MAPE of 3.54%.
Using evolutionary optimisation algorithms, including Genetic Algorithm and Particle Swarm Optimisation, the researchers identified conditions capable of maximising methane production.
The optimum conditions were predicted at a temperature of 54.87°C, pH 7.18, organic loading rate of 3.62 g VS L⁻¹ day⁻¹, hydraulic retention time of 28.41 days and carbon-to-nitrogen ratio of 27.83.
Under these conditions, the model predicted maximum methane production of 281.84 mL CH₄ g⁻¹ VS.
ALSO READ: Tinubu asks US court to block release of alleged drug trafficking records — Legal team
The researchers stated that the integrated framework provides a transparent decision-support system for improving anaerobic digestion efficiency and advancing sustainable waste-to-energy technologies.
The findings are particularly relevant for Nigeria, where agricultural residues from industries such as palm oil production represent a significant opportunity for renewable energy generation and environmental protection.
By combining biotechnology with artificial intelligence, the study demonstrates how digital technologies can support more efficient conversion of agricultural waste into valuable energy resources.
