Artificial neural network for prediction of color adsorption from an industrial textile effluent using modified sugarcane bagasse: Characterization, kinetics and isotherm studies

Abstract

The present work aims to study the biosorption process of color removal from real textile effluent using chemically modified sugarcane bagasse (SBM) as a biosorbent material and the prediction of the process by modeling and simulation of an artificial neural network (ANN). The raw sugarcane bagasse and the biosorbent SBM were characterized by scanning electron microscopy analysis (SEM), Fourier-transform infrared spectroscopy (FTIR), and porous structure. Batch experiments were carried out on the effect of the effluent pH, contact time between adsorbent and adsorbate, adsorbent dosage, particle size, and effluent color concentration on the adsorption process. The best-operating conditions found were in an acid medium, using an SBM particle size of 0.7 mm, and a dosage of 0.6 g, which allowed a color removal of 100 % for an initial true color concentration of 149 PtCo.L−1. The multilayer feed-forward neural network, with five inputs and one output, was trained with eight neurons in the hidden layer. A comparison between the experimental data and the predicted by ANN model showed that color removal results fitted very well to the model with a coefficient of determination (R2) of 0.928 and a mean square error (MSE) of 0.013. Nonlinear adjustments were made to the kinetic and adsorption isotherm models. In general, the adsorption process with SBM proved to be a promising method for the treatment of textile effluent, and the developed ANN model can be successfully used to make predictions for the final color of the effluent.