Aggregate size and density estimation from settling tests: An evaluation of available approaches using computational fluid dynamics

Estimating the size and density of fine particle aggregates in suspension is challenging, often involving analysing 2D images captured during settling tests. This study explores methods to better estimate drag and to derive size parameters, aiming to reduce calculation errors when characterizing complex aggregate structures. We compare 81 models for drag coefficient estimation, considering factors like aggregate shape and flow conditions. Using Particle-Resolved Direct Numerical Simulation, exact drag coefficient values are obtained under specific conditions for > 300 simulated structures. The accuracy and reliability of each model in drag and density calculations from 2D aggregate images is then assessed. Employing ellipsoid volume for estimating aggregate size yielded the smallest errors in drag coefficient calculations (12.2 %), in contrast to models assuming rigid spheres (71.3 %). Notably, the ellipsoid volume radius consistently provided the most precise aggregate drag estimates across all models, establishing its recommendation for dependable predictions. Models adaptable to a wider Reynolds number range displayed superior performance compared to those tailored for Stokes flow. Derived aggregate density was highly sensitive to the selected drag coefficient model. The guidelines provided for selecting the appropriate drag model and equivalent size when analysing 2D images enhances the scope for distinguishing changes in aggregate structural properties.