Response surface methodology and artificial neural network modelling of palm oil decanter cake and alum sludge co-gasification for syngas (CO+H2) production.

Syngas (CO + H 2) production through biomass gasification offers a promising and sustainable alternative to conventional fuels. This study investigates the co-gasification of palm oil decanter cake (PODC) and Alum Sludge (AS), utilizing response surface methodology (RSM) and artificial neural network (ANN) techniques to optimize and predict syngas production. Conducted in a fixed bed horizontal reactor, the experiment investigates temperature, airflow rate, and particle size as input parameters. Results revealed that optimal condition of 900 °C temperature, 10 mL/min airflow rate, and 2 mm particle size yielded the highest syngas production at 39.48 vol%. The RSM showed an R2 value of 0.9896, whereas ANN network revealed an overall R2 value of 0.971. Both models demonstrated strong alignment with experimental data and the modelled equation. This research demonstrates the effective use of statistical modelling to enhance the efficiency and effectiveness of syngas production, thereby fostering advancements in sustainable energy production. [Display omitted] • PODC and AS co-gasification was carried out in fixed bed horizontal tube furnace reactor. • RSM and ANN were applied for optimization and prediction of syngas production. • At optimal condition of 900 °C, 10 mL/min, and 2 mm, 39.48 vol% syngas was produced. • RSM and ANN model prediction accuracy were 98.96% and 97.1%, respectively. • Both RSM and ANN satisfactory validate and predict the response. [ABSTRACT FROM AUTHOR]