Prediction and optimization of heavy metal ions removal efficiency from the active sludge using intelligent systems
Morteza Esfandyari, Mesyam Khodadadi, Rezvan Nekoo Ghadirli, Dariush Jafari
Abstract
ABSTRACT In the current study, the capability of artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) techniques were investigated in prediction of heavy metal ions (lead and nickel) removal efficiency from the active sludge of an industrial zone’s wastewater treatment plant. The effect of experimental parameters such as solution pH, contact time, initial ions concentration, and temperature were studied to find the optimal values of these parameters. During the modeling multilayer perceptron network and a Sugeno Fuzzy model were applied in the proposed ANN and ANFIS models, respectively. The results of comparison between the experimental and the predicted data were satisfactory and the achieved correlation coefficients (>98%) approved the high accuracy of these two models, although ANFIS performed slightly better than ANN. Moreover, the optimal operating conditions were achieved by minimizing the objective function using genetic algorithm. Additionally, a real sludge sample was then treated under the achieved optimal conditions which yielded acceptable adsorption efficiency values. Based on these observations, the proposed intelligent models were acceptable tools for the prediction of these pollutants’ adsorption efficiencies.