A novel double-image-sequence correlation method for time-resolved particle image velocimetry

Abstract

This work reports a novel method with a goal of improving the accuracy and robustness of time-resolved particle image velocimetry (TR-PIV) measurements. This method, named double-image-sequence correlation (DISC), accomplishes this goal by utilizing the temporal information stored in TR-PIV measurements. The key to such utilization is that the DISC method constructs two separate image sequences with well-selected sequence length and separation time. As a result, multiple cross-correlation maps from consecutive image pairs can be averaged to reduce the random measurement noises (e.g., due to laser shot-to-shot variation and out-of-plane flow motion), and also each individual cross-correlation map can be obtained using the image pair separated with optimal time to further reduce the uncertainties due to other factors such as non-optimal in-plane velocity gradients and flow accelerations. The DISC method was validated both numerically and experimentally. The results illustrated that this method was able to enhance the accuracy of velocity measurements by 20.4–47.2% in terms of velocity error and by 7.8–22.9% in terms of non-physical Eulerian acceleration, compared to past multi-frame PIV methods.