Prediction and optimization of biogas production from POME co-digestion in solar bioreactor using artificial neural network coupled with particle swarm optimization (ANN-PSO)

Abstract

Biogas production from anaerobic co-digestion (ACoD) of palm oil mill effluent (POME) and cattle manure (CM) is getting a lot of attention due to its wide availability and relatively simple energy conversion technology. The ACoD process is extremely complex to model with conventional mathematical modeling methods and requires the use of advanced computational tools due to the mixing of different substrates. Artificial neural network (ANN) is a very recent alternative to modeling tools used to predict complex ACoD problems. To get the best performance from ANN, the parameters of ANN need to be optimized. Here, particle swarm optimization (PSO) algorithms can be a great option. The present study investigates the possibility of using the combined ANN-PSO framework to simulate the process and to predict biogas production from the ACoD of POME and CM. The mixture ratio of POME and CM, oxidation by hydrogen peroxide, and ammonium bicarbonate effects were analyzed separately to increase biogas production using solar-assisted bioreactors. From the experiment, five data volumes of the amounts of POME, CM, hydrogen peroxide, ammonium bicarbonate, and biogas yield were recorded. This dataset has been used to design the proposed model. The results of the proposed ANN-PSO framework with an understanding of mean square error (MSE) and correlation coefficient (R) are 0.0143 and 0.9923, respectively. This result indicates that the proposed method is found to be effective and flexible in predicting biogas production from the ACOD of POME and CM.