Optimization of parameters that affect biogas production of anaerobic digestion using a particle swarm algorithm

Renewable energy sources are significant in countries' energy policies; because of the increasing demand for energy and environmental concerns regarding fossil fuels, the world is seeking renewable energy alternatives. Biofuels derived from biomass are a good alternative to fossil fuels. Anaerobic digestion is a biochemical strategy for biogas production. Biomass is one of the most important sources, and biogas is produced by the anaerobic biodegradation of biomass. Since the anaerobic digester processes internal and external factors and depends on many parameters such as temperature, time, and NaOH concentration that affect the amount and quality of biogas, predictive and computer-optimized algorithms are used to predict and optimize biogas production using anaerobic digestion. In this study, an artificial neural network system was used to predict biogas production and the PSO algorithm was used to find the best neural network performance in optimizing biogas production. In evaluating the performance of the proposed method, the error values of MAE, MSE, and RMSE were assessed and the values of these three parameters were 0.0304, 0.0045, and 0.0538 for the best neural network prediction, respectively. Also, with optimization by the PSO algorithm, the optimum values of methane production parameters, methane concentration, methane efficiency, and residual volatile solids were 195.6451, 77.032%, 214.4211, and 6.9841%, respectively.