ANN-assisted forecasting of adsorption efficiency to remove heavy metals

In wastewater treatment, scientific and practical models utilizing numerical computational techniques such as artificial neural networks (ANNs) can significantly help to improve the process as a whole through adsorption systems. In the modeling of the adsorption efficiency for heavy metals from wastewater, some kinetic models have been used such as pseudo first-order and second-order. The present work develops an ANN model to forecast the adsorption efficiency of heavy metals such as zinc, nickel, and copper by extracting experimental data from three case studies. To do this, we apply trial-and-error to find the most ideal ANN settings, the efficiency of which is determined by mean square error (MSE) and coefficient of determination (R$^{2})$. According to the results, the model can forecast adsorption efficiency percent (AE%) with a tangent sigmoid transfer function (tansig) in the hidden layer with 10 neurons and a linear transfer function (purelin) in the output layer. Furthermore, the Levenberg--Marquardt algorithm is seen to be most ideal for training the algorithm for the case studies, with the lowest MSE and high R$^{2}$. In addition, the experimental results and the results predicted by the model with the ANN were found to be highly compatible with each other.