Particle image velocimetry (PIV) is an experimental approach widely used to investigate the flow inside stirred tanks. PIV noise sources deteriorate the velocity fields and compromise the measurements, mainly in the estimation of the turbulent parameters. PIV correction approaches based on the primary peak ratio (PPR), namely the ratio between the primary correlation peak to the second highest peak, are used when identifying outliers in the velocity field becomes difficult. However, in some cases, PPR is not sensitive to noise sources and may fail to identify outliers. Therefore, this work proposes a new method of field correction based on mutual information, which is estimated by dividing the height of the cross-correlation peak by the magnitude of the autocorrelation. Analysis were performed using PIV data measured in a 50 L stirred tank. Results show that the proposed methodology can be a valuable tool for the correction of noisy fields.
Read full abstract