Modeling the experimental results of lead removal efficiency from polluted solutions using Adaptive Neuro Fuzzy Inference System

The aim of this study is the investigation of the lead reduction from aqueous solution using zero valent iron nanoparticles (nZVI). The synthesis of nZVI was based on the reduction of ferrous iron with borohydride. The structure of synthesized nZVI particles was characterized by SEM, XRD, and BET analysis. BET analysis shows that, nZVI particles have high specific surface area. To model Pb(II) ions removal efficiency, the Adaptive Neuro Fuzzy Inference System (ANFIS) method and a polynomial function were applied. The Quantum behaved Particle Swarm Optimization algorithm (QPSO) and the least square error method were employed to train the ANFIS and polynomial function constant parameters. Based on the sensitivity analysis, the predicted Re is more sensitive to initial pH followed by adsorbent dosage, temperature, initial concentration of lead, and contact time. This study showed that the ANFIS model and polynomial function can be very effective alternatives to the expensive and time-consuming laboratory measurements.

کلیدواژه ها:

Adaptive neuro fuzzy inference system ، Lead ، Polynomial function ، Zero-valent iron nanoparticles