Comparative study of response surface methodology and generalized regression artificial neural network in the optimization of Ofloxacin degradation

Ofloxacin (OFX) is considered an emerging pollutant that is commonly found in surface water and wastewater. Due to its potential environmental impact, there is a growing need to develop effective technologies for its removal. One promising approach is the use of advanced oxidation processes, which have shown success in eliminating antibiotics. This study focuses on investigating the application of UV/H2O2 and ozonation (O3) for the degradation of OFX. To obtain meaningful results, experimental designs were implemented, and the data was modeled using RSM and generalized regression neural networks (GRNN) [1]. These modeling techniques allowed for the development of models to accurately describe and predict the degradation of OFX. Additionally, the study aimed to identify the most significant variables that affect the degradation process and to determine the byproducts formed during the treatment [2, 3]. This analysis was performed using high-performance liquid chromatography-mass spectrometry (HPLC-MS). The concentration of the antibiotic and the pH level were identified as the most influential variables affecting the degradation process. In a short span of 30 minutes, a high removal percentage of 90- 97% was achieved, with pH being the most significant variable. Utilizing the GRNN analysis, a predictive model was developed, which showed a good fit between the model and experimental data. This indicates the reliability of the model in accurately predicting the degradation of OFX. Furthermore, the study identified the main byproducts formed during the degradation process, with many of them disappearing after 30 minutes of treatment.