Coupling a mathematical and a fuzzy logic-based model for prediction of zinc ions separation from wastewater using electrodialysis

This paper presents experimental data, a fuzzy logic (FL) model, a mathematical model (MM) and a coupled MM–FL model for a laboratory scale electrodialysis (ED) cell. The aim was to predict separation percent (SP) of zinc ions as a function of concentration, temperature, flow rate and voltage. At first, a Sugeno type FL inference system was applied to model zinc ions separation from wastewater using ED. FL modeling results showed that there is an excellent agreement between the experimental data and the predicted values, with mean squared relative error (MSRE) of less than 0.01. Then, the results of a previously developed MM were presented. The MM related SP to hydrodynamic dimension of the ED cell and operation conditions via two distinct parameters. This ability favored the MM for scale-up applications. However, based on MSRE of the MM (about 24), it could not obviously predict the experimental data as well as FL. Hence, as a final step, the MM was coupled with FL to achieve benefits of both. It was found out that the developed coupled model (MM–FL) is able to predict SP of zinc ions at all operating condition and almost every dimension to a high degree of accuracy (MSRE = 0.05).