Optimizing Organic Dye Degradation via Electro-Peroxone Process: An Experimental and Machine Learning Approach

The electroperoxone (EPO) process, integrating ozonation and electrochemical hydrogen peroxide generation, has gained attention as an efficient advanced oxidation technology for treating recalcitrant pollutants. This study investigates the application of EPO for the removal of organic dye from synthetic wastewater using a two-stage analytical framework. In the first stage, a series of systematic batch experiments were conducted to explore the effects of key operational parameters, including initial pH, applied current, ozone dosage, and reaction time, on decolorization efficiency. In the second stage, predictive models were developed using machine learning algorithms—Support Vector Regression (SVR) and Random Forest (RF)—to capture the complex nonlinear behavior of the process. The Random Forest model outperformed others, achieving an R² value above ۰.۸۲۳ and demonstrating superior accuracy in predicting removal efficiency. Sensitivity analysis revealed ozone dosage and applied current as the most influential factors. These results highlight the potential of combining experimental optimization with robust data-driven modeling to enhance the design and scalability of advanced oxidation processes in wastewater treatment.

کلیدواژه ها:

Electroperoxone ، degradation ، Machine learning ، Experimental optimization ، Advanced oxidation processes