Ozonation of offshore produced water: kinetic study and fuzzy inference system modeling

Ozonation has been recently proposed as a treatment option to remove toxic and recalcitrant organics from offshore produced water (OPW). In this study, experimental and modeling approaches were used to investigate the removal of polycyclic aromatic hydrocarbons (PAHs) from OPW by ozonation. It was found that ozonation can effectively remove PAHs by following the pseudo first-order kinetic model, which could cover both direct and indirect ozonation pathways. The decay rate constants of most PAHs were as low as 0.01-0.16 min(-1), possibly due to the interference caused by OPW matrix. The effects of ozone dose, bubble size, pH, and temperature on five representative PAHs were investigated along with their possible interactions. Based on the experimental results, a novel multi-output adaptive network-based fuzzy inference system (MO-ANFIS) was developed to model the removal of four individual PAHs and total PAHs. The overall RMSE and R between measured and modeled removal rates were 6.60% and 0.98, respectively, indicating a good model fit. This study demonstrated the effectiveness of ozonation in OPW treatment and the potential applicability of MO-ANFIS for process modeling and control.