Adaptive neuro‐fuzzy inference system modeling of 2, 4‐dichlorophenol adsorption on wood‐based activated carbon. Issue 5 (3rd March 2020)

• Record Type: Journal Article
• Authors: Alver, Alper; Baştürk, Emine; Tulun, Şevket; Şimşek, İsmail
• Is Part Of: Environmental progress & sustainable energy. Volume 39:Issue 5(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-03-03
• Subjects: activated carbon -- adaptive neuro‐fuzzy inference system -- adsorption -- isotherm -- kinetic behavior -- phenol; Environmental engineering -- Periodicals; Sustainable engineering -- Periodicals; Environmental chemistry -- Periodicals; 333.7
• DOI: 10.1002/ep.13413
• Languages: English
• ISSNs: 1944-7442

Abstract

Abstract: Phenolic compounds cause significant problems both in drinking water and wastewater due to their toxicity, high oxygen requirements, and low biodegradability. They are listed as primary pollutants by the United States Environmental Protection Agency and the European Union. In this study, the adsorption efficiency of 2, 4‐dichlorophenol (2, 4‐DCP) on activated carbon, which is commonly used in treatment plants, was investigated under different experimental conditions including adsorbent dose, initial phenol concentration, initial pH, and contact time. As a result of experimental studies, it was determined that the adsorption isotherm and kinetics could be perfectly fitted to Langmuir and the assumption of pseudo‐second order model, respectively. Then, the adaptive neuro‐fuzzy inference system (ANFIS) model was developed, which was the primary purpose of this study. The correlation between training and testing data and the ANFIS output was over 0.999. The generalization ability of the model was found to be 0.999. The input variables such as adsorbent dosage (14.2%), initial concentration (14.6%), initial pH (13.9%), and the contact time (57.2%) showed a higher effect on 2, 4‐DCP removal efficiency in the sensitivity analysis. To summarize, modeling studies that are frequently preferred in treatment plants for the removal of different pollutants will reduce the number of experiments harmful to human health and save time, labor, and economy.