Abstract
This study investigates urea removal from wastewater using zinc-based electrocoagulation method, supported by Gradient Boosting Regressor modeling. The highest urea removal of 42% was obtained for 1.2 g/L initial urea concentration, 22 mA/cm2 current density, and a pH solution of 10, while natural pH (≈7.50) gave 30%. By applying the optimum conditions 27% of urea was removed from real hospital effluent. Application of GBR model leveraging Artificial Intelligence (AI) demonstrated a high predictive accuracy (R2 = 0.9825, RMSE = 0.01666) with experimental results. Treatment combination processes were investigated: Chemical Coagulation–Electrocoagulation achieved 35% efficiency, while two EC cycles yielded 45%. Electrocoagulated sludge characterization by scanning electron microscope/energy-dispersive X-ray spectroscopy, Fourier-transform infrared spectroscopy, X-ray diffraction analysis revealed surface irregularities as well as the presence of zinc, carbon, nitrogen, and sodium. These findings confirm the treatment’s effectiveness in removing urea and support the safe valorization and reuse of the sludge. EC proves effective and cost-efficient for industrial-scale implementation.