An automated predictor-corrector method for vortex detection

Detection of vortices from PIV data has been a challenging task driving efforts in experimentation with various algorithms and physical criteria. In the present study, three algorithms are employed and are arranged in a predictor-corrector fashion to reduce computation time while maintaining accuracy and robustness. Two predictor methods, geometric gridding and streamlines method have been employed to ensure that there is redundancy in finding the regions of interest at the predictor stage. A redundancy in the predictor method also ensures generality and robustness in detection. The corrector method is a pattern matching algorithm using the Hamel-Oseen model for velocity which acts on the reduced domain obtained through the predictor algorithms. The performance of the proposed scheme is demonstrated by applying them to detect vortices from two component PIV data obtained from swirl flow measurements. Copyright © 2008 by Arun Manohar, A. V. Varun and R. I. Sujith.