Application of supervised learning models for enhanced lead (II) removal from wastewater via modified cellulose nanocrystals (CNCs)

Abstract

Heavy metal ions are acknowledged to impact the environment and human health adversely. CNCs are effective materials for removing heavy metal ions in industrial applications and process innovations since they can be used in static and dynamic adsorption processes. Cost-effective, uncomplicated water treatment technologies must be developed using biodegradable polymers, namely, modified cellulose nanocrystals. Adaptive neuro-fuzzy inference systems (ANFISs) and artificial neural networks (ANNs) were used to evaluate and examine the efficacy of modified cellulose nanocrystals in removing lead(II) from wastewater. The research indicated that the maximum adsorption capacity attained was 260 mg/g at a pH of 6, an initial concentration of 200 mg/L, a contact duration of 300 min, and a 5 g/200 mL dose. Influence of four input variables on the Pb(II) adsorption capacity: The experimental data were juxtaposed with the outcomes from ANN and ANFIS to ascertain the pH, contact time, starting concentration, and dose. The correlations of 0.9916 for the created artificial neural network (ANN) and 0.9953 for the adaptive neuro-fuzzy inference system ANFIS indicate that the study data may be predicted with precision. ANFIS had a Pearson's chi-square value of 0.638, surpassing the ANN's score of 0.979.