Prediction of the As(III) and As(V) Abatement Capacity of Zea mays Cob Powder: ANN Modelling

Arsenic contamination of ground water is unfolding as one of the worst natural geo environmental disaster to date. Due to increasing environmental awareness and legal constraints imposed on discharge of effluents, the need for cost effective alternative technologies is essential for removal of arsenic from water bodies. Zea mays cob powder (ZMCP) is an excellent biomaterial as it is largely available as a waste and having high sorption capacity, hence, used for the present study. Optimization of the process variables (biomaterial dosage, contact time, arsenic concentration, volume and pH) for the decontamination of As(III) and As(V) using artificial neural network modeling were studied. Back-Propagation and Levenberg–Marquardt techniques are used to train various neural network architectures and the accuracy of the obtained models have been examined by using testing data set. The minimum mean square error in the group of five variables was determined for training and cross validation are 1.48275E−05 and 0.000140872 respectively. The performance of the network for predicting the sorption efficiency of biosorbent is found to be very impressive.