Modeling industrial hydrocyclone operational variables by SHAP-CatBoost - A “conscious lab” approach

Undoubtedly hydrocyclones play a critical role in powder technology, which can considerably affect the plants' process efficiency. However, were rarely modeled on an industrial scale, where model be used to train operators and minimize potential scale-up errors lab costs. The novel approach for filling such gap would using conscious “CL” as new concept that builds based dataset explainable artificial intelligence (XAI). As approach, this study developed CL explored interactions between hydrocyclone variables by most recent XAI method called “SHapley Additive exPlanations (SHAP)”, machine-learning model, “CatBoost”. output particle size of plant magnetic separator SHAP-CatBoost. SHAP could successfully all relationships, CatBoost predict O80 K80, outcomes had higher accuracy (R2 ∼ 0.90) than other conventional AIs.