From data to decisions: Bayesian modelling and global sensitivity analysis for flotation control
Abstract
A data-driven framework for interpretable modelling and decision support in flotation. A Gaussian process surrogate trained on laboratory data relates operating variables such as air velocity, froth height and bubble size to air recovery, and is combined with Sobol global sensitivity analysis and SHAP explanations to identify which variables, and which interactions, matter most. The work lays foundations for data-driven control and optimisation of flotation.
Type
Publication
arXiv preprint
Dr Paulina Quintanilla
Assistant Professor in Process Systems Engineering
Assistant Professor in Process Systems Engineering at UCL, leading the LOOPS research group. Research interests include machine learning, optimisation and control for processes and physical systems.