Vortex detection has been a long-standing and challenging topic in fluid analysis. Recent state-of-the-art extraction and visualization of vortices in unsteady fluid flow employ objective vortex criteria, which makes feature extraction independent of reference frames or observers. However, even objectivity can only guarantee that different observers reach the same conclusions, but not necessarily guarantee that these conclusions are the only physically meaningful or relevant ones. Moreover, a significant challenge is that a single observer is often not sufficient to accurately observe multiple vortices that follow different motions. This thesis presents a novel mathematical framework that represents physically realizable observers as the Lie algebra of the Killing fields on the underlying manifold, together with a software system that enables the exploration and use of an interactively chosen set of observers, resulting in relative velocity fields and objective vortex structures in real-time. Based on our mathematical framework, our system facilitates the objective detection and visualization of vortices relative to well-adapted reference frame motions, while at the same time guaranteeing that these observers are physically realizable. We show how our framework speeds up the exploration of objective vortices in unsteady 2D flow, on planar as well as on spherical domains.
|Date made available
|KAUST Research Repository