Temporally resolved reconstruction of sweeping jet flow field based on sub-Nyquist-rate PIV data

Abstract

Sweeping jets can produce kHz frequency actuation for effective flow control. However, high-frequency flow dynamics poses a challenge for the temporal resolution of particle image velocimetry (PIV) measurements. In this article, two methods for reconstructing the temporally resolved flow fields of sweeping jets from sub-Nyquist-rate PIV data, namely least squares (LS) regression and compressive sensing (CS) reconstructions, are proposed and compared. In both methods, proper orthogonal decomposition (POD) is first conducted to reduce the dimensionality of the data by acquiring the spatial modes and the corresponding sub-Nyquist-rate coefficients. For LS regression reconstruction, high-frequency data obtained from pointwise measurement (hotwires) are used to estimate the temporally resolved POD coefficients. For the CS method, the temporally resolved coefficients are calculated by solving a basic pursuit problem based on the sub-Nyquist coefficients. Together with spatial modes, the temporal sweeping jet flow fields can be reconstructed. The performance of LS regression in reproducing the ground truth is highly sensitive to the number of pointwise probes and has a high error rate. In comparison, the CS method can provide more accurate reconstruction with an error rate below 3% without the need for pointwise measurements.