Prediction of microbial desulfurization of coal using artificial neural networks

Artificial neural networks procedures were used to predict the organic and inorganic sulfur reduction from coal using mixed culture consisted ferrooxidans species extracted from coal washery tailings, for pyritic sulfur, and Rhodococcus species, extracted from oily soils, for the organic sulfur removal. The particle size, pulp density, initial pH, shaking rate, leaching time and temperature, in pyritic sulfur removal prediction, and pulp density, shaking rate, leaching time and temperature, in organic sulfur removal prediction, were used as inputs to the network. Feed-forward artificial neural networks with 4-8-4-1 and 3-5-6-1 arrangements, were capable to estimate organic and inorganic sulfur removal, respectively. The outputs of the models were percentage of organic and inorganic sulfur reduction. It was achieved quite satisfactory correlations of R2 = 1.00 and 0.98 in training and testing stages for pyritic sulfur removal prediction and R2 = 1.00 and 0.97 in training and testing stages, respectively, for organic sulfur removal prediction. The proposed neural network models accurately estimate the effects of operational variables in organic and inorganic desulphurization plants and can be used in order to optimize the process parameters without having to conduct the new experiments in laboratory.