A procedure for turbulent structure convection velocity measurements using time-correlated images

Abstract

This paper describes the development of a technique used to determine the convection velocity of large-scale turbulent structures captured in time-correlated images. The crux of the procedure centers on a cross-correlation routine that is used to determine the convection distance of eddies during the time separation between the image acquisitions. The convection velocity is then estimated as the convection distance divided by the time separation of the image-pair. This cross-correlation routine is capable of analyzing very large data sets in a completely automated manner, thereby improving the accuracy and objectivity of the results over manual or partially automated procedures. Guidelines for optimizing the experimental and computational components of this technique are also presented.