Improvements on Particle Tracking Velocimetry: model-free calibration and noiseless measurement of second order statistics of the velocity field

TL;DR

This paper introduces a model-free stereoscopic calibration method and a noise reduction technique for particle tracking velocimetry, enhancing the accuracy of flow velocity measurements and second order statistics in turbulent flows.

Contribution

It presents a novel optical calibration approach that outperforms traditional models and a noise removal method that improves the accuracy of velocity statistics without filtering.

Findings

Calibration method yields better results than Tsai model.

Noise reduction technique accurately recovers second order velocity structure functions.

Enhanced measurement accuracy aligns with validated turbulence metrics.

Abstract

This article describes two independent developments aimed at improving the Particle Tracking Method for measurements of flow or particle velocities. First, a stereoscopic multicamera calibration method that does not require any optical model is described and evaluated. We show that this new calibration method gives better results than the most commonly-used technique, based on the Tsai camera/optics model. Additionally, the methods uses a simple interpolant to compute the transformation matrix and it is trivial to apply for any experimental fluid dynamics visualization set up. The second contribution proposes a solution to remove noise from Eulerian measurements of velocity statistics obtained from Particle Tracking velocimetry, without the need of filtering and/or windowing. The novel method presented here is based on recomputing particle displacement measurements from two consecutive frames for multiple different time-step values between frames. We show the successful application of this new technique to recover the second order velocity structure function of the flow. Increased accuracy is demonstrated by comparing the dissipation rate of turbulent kinetic energy measured from the second order structure function against previously validated measurements. These two techniques for improvement of experimental fluid/particle velocity measurements can be combined to provide high accuracy 3D particle and/or flow velocity statistics and derived variables needed to characterize a turbulent flow.