Enhancing sustainability in sewage treatment: A least squares support vector regression-based modeling approach for optimizing regeneration conditions of iFeCu

Nanoparticles in wastewater treatment offer sustainable, efficient removal of persistent pollutants, with minimal resource usage and reusability. This study examines the reusability of immobilized iron‑copper nanoparticles (iFeCu) for sewage treatment. The effect of the regeneration condition of iFeCu including the regeneration time and temperature on the performance of iFeCu for sewage treatment was investigated. Moreover, a soft sensor model, namely principal component analysis discrimination analysis ensemble least squares support vector regression (PCA-DA-E-LSSVR) model was developed to predict the performance of iFeCu, and an online performance of the model was studied to evaluate its accuracy. Investigating regeneration conditions like time and temperature, it was found that iFeCu regenerated in 1 mol (M) sodium carbonate (Na2CO3) at 40 °C for 5 h performed comparably to that regenerated at 60 °C for 15 h. Higher regeneration temperatures (>40 °C) decreased carbon dioxide and monoxide emissions but reduced ammonia (NH3) removal by ∼22 %. Shorter regeneration times (<5 h) decreased Na2CO3-iFeCu interaction, leading to an ∼11 % NH3 removal reduction. The correlation coefficient, R2 value of 0.8514 was obtained for the proposed PCA-DA-E-LSSVR model. Overall, the developed model well-fits the nonlinear data in sewage treatment and provides acceptable ranges of online performance results.