Neuro-fuzzy logic system for prediction of dye removal from aqueous solutions by a polymeric nanocomposite

Document Type : Original Article


1 Department of Physics, Qom Branch, Islamic Azad University, Qom, Iran

2 Distributed Intelligent Optimization Research Laboratory, Department of Electrical Engineering, Amirkabir University of Technology, Tehran, Iran


Here we explain the successful synthesis of a new polymeric nanocomposite and study the removal of a cationic dye, brilliant green, from aqueous solutions. The nanocomposite is characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD). In addition, adsorption kinetics and effects of various variables such as solution pH, adsorbent mass, and initial dye concentration are investigated. Moreover, the fuzzy model is used to investigate the dye removal efficiency. The investigation of the results obtained using the fuzzy model shows that it is suitable for the prediction of dye adsorption onto nanocomposite. The coefficient of determination (R2) and mean squared error (MSE) for the optimal model were obtained to be 0.9981 and 0.1538, respectively. The equilibrium experimental data are represented by Langmuir, Freundlich, and Sips isotherms. Results briefly indicate that adsorption of dye by nanocomposite obeys Langmuir isotherm and pseudo-second-order kinetic model.