Neuro fuzzy approach for arsenic(III) and chromium(VI) removal from water

The present research work focuses on the predictive method for the removal of As(III) and Cr(VI) from water using a ZEDA hybrid material. A simple sol–gel technique has been used to synthesize the hybrid material followed by physicochemical characterization. Batch and column adsorption experiment has been conducted as a function of different variables. The maximum removal percentage at an optimal condition of pH 6.5 and dose = 1.2 mg/l for As(III) and optimal condition of pH 5.5 and dose = 1.4 mg/l for Cr(VI) has been observed to be 97% and 91%, respectively with contact time of 40 min, temperature 60 °C and initial concentration of 5 mg/l. The experimental adsorption data fitted best to pseudo second-order kinetic and Langmuir isotherm model. The maximum adsorption capacity of the material is 36 and 31 mg/g for As(III) and Cr(VI), respectively. A predictive model has been used to estimate the removal percentage without resorting to costly experimental efforts. The data series of adsorption experiments are used for prediction using adaptive neuro-fuzzy inference system (ANFIS). The accuracy of the model has been found to be satisfactory with the correlation coefficient (R2) and average absolute relative percentage error (AARE) ranging between 0.94–0.99 and 0.652–0.041, respectively.