Breakthrough Curves Prediction of Selenite Adsorption on Chemically Modified Zeolite Using Boosted Decision Tree Algorithms for Water Treatment Applications

This work describes an experimental and machine learning approach for the prediction of selenite removal on chemically modified zeolite water treatment. Breakthrough curves were constructed using iron-coated adsorbent adsorption behavior was evaluated as a function initial contaminant concentration well ionic strength. An elevated selenium in threatens human health aquatic life. The migration this metalloid from contaminated sites problems associated with its high releases into has become major environmental concern. mobility emerging prompted development efficient, cost-effective removal. Selenite [Se(IV)] aqueous solutions studied laboratory-scale continuous packed-bed columns natural adsorbents. proposed combines iron oxide zeolite’s ability to bind contaminants. initially obtained under variable conditions, including change Se (IV), strength solutions. Investigating effect these parameters will enhance retardation water. Continuous experiment findings evaluate efficiency economical naturally-based fate Multilinear non-linear regressions approaches utilized, yet low coefficients determination values respectively obtained. Then, comparative analysis five boosted regression tree algorithms breakthrough curve performed. AdaBoost, Gradient boosting, XGBoost, LightGBM, CatBoost models analyzed data columns. performance different operation such strength, discussed. applicability metrics (i.e., Mean Absolute Error (MAE), Root Square (RMSE), Percentage (MAPE), coefficient (R2). model provided best fit R2 equal 99.57. k-fold cross-validation technique statistical verify model’s accurateness. A feature importance assessment indicated that (IV) most influential variable, while had least effect. finding consistent column transport results, which observed sorption dependency inlet concentration; simultaneously, negligible. proposes implementing learning-based predicting remediation-associated processes. significance provide alternative method investigating machine-based approach. also highlighted management practices processes involved remediation.