Support Vector Regression prediction of two dyes removal using Electrocoagulation

ABSTRACT The aim of the present study is comparison of Electrocoagulation (EC) removal of Acid Red ۱۸ (AR ۱۸) and Acid Orange ۷ (AO ۷) dyes and finding the Support Vector Regression (SVR) models for prediction of the process. For these purposes three parameters of applied current (۱-۳ A), initial dye concentration (۱۰۰-۴۰۰ mg/l) and experiment time (۰-۴۰ min) were examined. Average removal efficiencies of ۳۴.۴۳% for AR ۱۸ and ۷۲.۲۱% for AO ۷ indicated higher degrees of treatment for AO ۷ withdrawn samples from the system. According to the results of the present study, AO ۷ and AR ۱۸ removal efficiencies were obtained between ۷.۹۲-۹۴.۸۲% and ۱.۳۴-۸۵.۰۹%, respectively. High R۲ values (≥۹۷%) for AR ۱۸ and AO ۷ SVR models showed that the models were reliable. As a result of this investigation, robust SVR is an effective tool for prediction of the EC process. Furthermore, final pH of the treated solutions was reported ۸.۵-۹.۵.

کلیدواژه ها:

_Electrocoagulation ، Acid Red ۱۸ ، Acid Orange ۷ ، Support Vector Regression