Vortex detection and tracking in massively separated and turbulent flows

The vortex produced at the leading edge of the wing, known as the leading edge vortex (LEV), plays an important role in enhancing or destroying aerodynamic force, especially lift, upon its formation or shedding during the flapping flight of birds and insects. In this thesis, we integrate multiple new and traditional vortex identification approaches to visualize and track the LEV dynamics during its shedding process. The study is carried out using a 2D simulation of a flat plate undergoing a 45 degree pitch-up maneuver. The Eulerian 1 function and criterion are used along with the Lagrangian coherent structures (LCS) analyses including the finite-time Lyapunov exponent (FTLE), the geodesic LCS, and the Lagrangian-Averaged Vorticity Deviation (LAVD). Each of \h{these} Lagrangian methods \h{is} applied at the centers and boundaries of the vortices to detect the vortex dynamics. The techniques enable the tracking of identifiable features in the flow organization using the FTLE-saddles and -saddles. The FTLE-saddle traces have shown potential to identify the timing and location of vortex shedding, more precisely than by only studying the vortex cores as identified by Eulerian techniques. The traces and the shedding times of the FTLE-saddles on the LEV boundary matches well with the plate lift fluctuation, and indicates a consistent timing of LEV formation, growth, shedding. The formation number and vortex shedding mechanisms are compared in the thesis with the shedding time and location by the FTLE-saddle, which validates the result of the FTLE-saddles and provide explanations of vortex shedding in different aspects (vortex strength and flow dynamics). The techniques are applied to more cases involving vortex dominated flows to explore and expand their application in providing insight of flow physics. For a set of experimental two-component PIV data in the wake of a purely pitching trapezoidal panel, the Lagrangian analysis of FTLE-saddle tracking identifies and tracks the vortex breakdown location with relatively less user interaction and provide a more direct and consistent analysis. For a simulation of wall-bounded turbulence in a channel flow, tracking FTLE-saddles shows that the average structure convection speed exhibits a similar trend as a previously published result based on velocity and pressure correlations, giving validity to the method. When these Lagrangian techniques are applied in a study of the evolution of an isolated hairpin vortex, it shows the connection between primary and secondary hairpin heads of their circulation and position, and the contribution to the generation of the secondary hairpin by the flow characteristics at the channel wall. The current method of tracking vortices yields insight into the behavior of the vortices in all of the diverse flows presented, highlighting the breadth of its potential application