Bio-Electro-Fenton process for efficient degradation of metronidazole: RSM and ANN-GA optimization

Bio-electro-Fenton (BEF) processes utilize biological electrons generated by Bio-Electrochemical Systems (BES) for wastewater treatment. In this report, this process was utilized to remove metronidazole (MTZ) from pharmaceutical wastewater. The performance of the BEF system was investigated and optimized based on key parameters including MTZ concentration, reaction time, pH, anodic chamber substrate concentration (chemical oxygen demand (COD)) and iron concentration using response surface methodology with Box-Behnken Design (RSM-BBD) and artificial neural network combined with genetic algorithm (ANN-GA) approaches. The results indicated that the ANN model outperformed the RSM model, as indicated by its lower Sum of Squared Errors (SSE), Average Relative Errors (ARE), Mean Squared Error (MSE), Absolute Average Deviation (AAD), Sum of Absolute Errors (SAE), and Root Mean Squared Error (RMSE) values, as well as its higher R 2 value. However, both models showed excellent performance in predicting the elimination of MTZ. Under optimal operating conditions (pH = 3, reaction time = 24 h, MTZ concentration = 5 mg/L, COD concentration = 6000 mg/L and iron concentration = 6 mg/L), the removal efficiency exceeded 90%. Additionally, decreasing MTZ concentration, lowering pH, and increasing COD concentration resulted in higher power generation, with maximum power densities of 67.68, 88.45, and 92.38 mW/m 2 , respectively. In conclusion, both models have significant potential for effectively simulating wastewater treatment processes.