Artificial neural network-based modelling to predict biogas and methane production rates in a pilot-scale mesophilic up-flow anaerobic sludge blanket (UASB) reactor treating molasses wastewater

Two three-layer artificial neural network (ANN) models (8:9:1 and 8:12:1) were respectively developed to predict biogas and methane production rates in a pilot-scale mesophilic up-flow anaerobic sludge blanket (UASB) reactor treating molasses wastewater.Eight input variables such as volumetric organic loading rate (OLR), operating temperature, influent and effluent alkalinity, influent and effluent pH, effluent COD and volatile fatty acids (VFA) concentrations were modelled by the use of an artificial intelligence-based approach.In the study, proposed ANN-based models were created by using MATLAB® V7.0 software program. After backpropagation (BP) training combined with principal component analysis (PCA), the ANN models predicted biogas and methane production rates based on the experimental data, and all the predictions were proved to be satisfactory with correlation coefficients of about 0.967 and 0.961 for biogas and methane, respectively.In the ANN study, the scaled conjugate gradient algorithm was found as the best of 11 BP algorithms. The numbers of neurons in the hidden layer were optimized as 9 and 12 for the ANN models in estimation of biogas and methane production rates, respectively.The ANN predicted results were also compared with the outputs of two exponential non-linear regression models derived in this study. Findings of this study clearly indicated that, compared to non-linear regression models, the proposed ANN-based models produced smaller deviations and exhibited a superior predictive performance on forecasting of both biogas and methane production rates. Both ANN outputs and lineer/non-lineer study results were compared and advantages and further developments were evaluated.