Boosting ensembles for estimation of discharge coefficient and through flow discharge in broad-crested gabion weirs

Abstract Gabion weirs are environment-friendly structures widely used for irrigation and drainage network purposes. These structures' hydraulic performance is fundamentally different from solid weirs' due to their porosity the existence of a through-flow discharge. This paper investigates reliability suitability number Machine learning models estimation gabion weirs. Generally, three Boosting ensemble models, including Gradient Boosting, XGBoost, CatBoost, compared well-known Random Forest Stacked Regression model, with respect accuracy in prediction discharge coefficient ratio free flow conditions. The Bayesian optimization approach fine-tune model hyper-parameters automatically. Recursive feature elimination analysis also performed find optimum combination features each model. Results indicate that CatBoost has outperformed other terms estimating through ( Q /Q t ) R 2 = 0.982, while both XGBoost have shown close C d equal 0.994 0.992. Weakest results were produced by Decision tree regressor 0.821 0.865 /Qt values.