A grid based surface tracking algorithm for efficiently computing Lagrangian coherent structures in 3D unsteady flows

Abstract

A grid based surface tracking algorithm is developed for the efficient computation and extraction of Lagrangian coherent structures (LCS) in 3D unsteady flows. To avoid the costs of mesh generation, a grid based “marching cubes” type algorithm is used. Large savings in computational time are achieved by performing computations only near the ridges in the finite time Lyapunov exponent (FTLE) field, which identify LCS. This algorithm improves the computational order of the LCS algorithm from O(1/dx 3 ) to O(1/dx 2.2 ) and typical applications may see more than an order of magnitude of speedup.