Development and Application of Combined CFD and Data-driven Models for Hydrocyclones

Hydrocyclones, widely employed for particle separation, exhibit performance strongly influenced by geometries, operating conditions, and material properties. This thesis develops data-driven models, trained on mechanistic and experimental data, to enhance the prediction, optimization, intelligent control, and industrial application of hydrocyclone performance.