Modeling and optimization of biogas production from cow manure and maize straw using an adaptive neuro-fuzzy inference system

Abstract

This study was focused on the prediction and optimization of biogas production from cow manure with maize straw under various total solid content (TS), Carbon to Nitrogen (C/N) ratio and stirring intensity. This research used full-scale (1200 L) batch reactor under mesophilic condition. An adaptive neuro-fuzzy interference system (ANFIS) was utilized to predict and optimize biogas production from anaerobic digestion. C/N ratio, TS and stirring intensity of substrates, each of them in three levels, were considered as input variables and biogas production was regarded as the output variable of the model. The coefficient of determination (R2) between observed and predicted biogas production values was 0.99 which showed good match and accuracy of the model. Highest biogas production was achieved from C/N ratio 26.76, TS 9% and moderate stirring. Biogas production increased about 8% with optimal conditions suggested by the ANFIS model.