A Validation Framework for Lattice Boltzmann – LES Simulations of Single-Phase Mixing in Stirred Tanks
Please login to view abstract download link
During the past years the focus on developing digital twins has led to an increased interest in computational approaches to reduce the amount of experimental work to develop and establish manufacturing processes. Questions on the credibility of results provided by computational modelling have therefore gained a lot of attention. In the current work, we employ the ASME Validation & Verification (V&V) 40 standards [3], to create a validation framework for the Computational Fluid Dynamics (CFD) modelling of mixing of Newtonian miscible fluids in stirred tanks. The focus of this simulation validation framework is to develop a rationale to identify which phenomena are the most important for single phase mixing processes in stirred tanks and generate relevant validation datasets. Large Eddy Simulations (LES) using the Lattice Boltzmann method (LB) are the state of the art for modelling turbulent flows during mixing. We are using the LB-LES software SimVantage to model different stirred tank experiments from the literature, in order to assess the capability of the simulations to predict: - The hydrodynamics of single fluid mixing (mean, periodic and turbulent flow components) near the blades of a Rushton turbine impeller. [1] - The mixing time of a passive scalar in a liquid bulk. [2] - The mixing time of miscible liquids with significant differences in material properties such as density and viscosity. According to the FDA guidance document for assessing the credibility of simulations [4], this work is a practice of validation performed against retrospective datasets. The validations performed are not specific to the Context of Use (COU), but provide strong credibility evidence for the accuracy of the simulation results, in an effort to increase the applicability of these validation activities.
