A comparative study of response surface methodology and artificial neural network based algorithm genetic for modeling and optimization of EP/US/GAC oxidation process in dexamethasone degradation: Application for real wastewater, electrical energy consumption.

Dexamethasone (DXM) is a broadly used drug, which is frequently identified in the water environments due to its improper disposal and incomplete removal in wastewater treatment plant. The inability of conventional treatment processes of wastewater causes that researchers pay a great attention to study and develop effective wastewater treatment systems. This work deals with the study of integrated electro-peroxone/granular activated carbon (EP/US/GAC) process in the degradation of dexamethasone (DXM) from a water environment and the remediation of real pharmaceutical wastewater. Two approaches of response surface methodology based on central composite design (RSM-CCD) and artificial neural network based on algorithm genetic (ANN-GA) were employed for modeling and optimization of the process. Both the models presented significant adequacy for modeling and prediction of the process according to statistical linear and nonlinear metrics (R2  = 0.9998 and 0.9996 and RMSE = 0.2128 and 0.1784 for ANN-GA and RSM-CCD, respectively). The optimization study provided the same outcomes for both ANN-GA and RSM-CCD approaches, where approximately complete DEX oxidation was achieved at pH = 9.3, operating time = 10 min, US power = 300 W/L, applied current = 470 mA, and electrolyte concentration = 0.05 M. A synergistic study signified that the EP/US/GAC process made an 82% synergy index as compared to the individual US and EP processes. The calculated energy consumption for the integrated process was achieved to be 2.79 kW h/gCOD. Quenching test by tert-butanol and p-benzoquinone revealed that HO• radical possessed the largest contribution in DEX degradation. The efficiency of EP/US/GAC process in the remediation of real pharmaceutical wastewater showed a significant decline in COD content (92% removal after 180 min), and the ratio of initial BOD/COD ratio of 0.27 was elevated up to 0.7 after 100 min treatment time. The performance stability of EP/US/GAC system showed no remarkable drop in removal efficiency, and leakage of lead ions from the anode surface was negligible and below WHO guideline for drinking water. Generally, this research work manifested that the integrated EP/US/GAC system elevated the degradation efficiency and can be proposed as a pretreatment step before biological treatment processes for the remediation of recalcitrant wastewaters.