Prediction of adsorptive capacity of various agricultural wastes in the removal of heavy metals, dyes, and antibiotic in wastewater using ANN

Abstract

Artificial Neural Network model was proposed for the prediction of the adsorptive capacity of various agricultural wastes in the removal of heavy metals, dyes, and antibiotic in water. A total amount of 103 data sets were obtained from different literature and was split into training (70%), validation (15%) and testing (15%) data. After considering different architectures, an input layer that uses eight independent variables (molecular weight of the adsorbate, adsorbent, adsorbent pre-treatment preparation, average initial concentration of adsorbate in solution, mass of adsorbent, adsorbent dosage, pH, and temperature), one hidden layer with 18 neurons and one neuron in the output layer was found to give the best result. The overall mean square error was 3487, while the correlation coefficient for the test dataset is 0.91898.