How PIV Works
How PIV Works
PIV Basics
Particle Image Velocimetry (PIV) is an optical experimental technique engineers use to both visualize and measure fluid flow fields (i.e., liquid or gas). A common laboratory set up for PIV is depicted in Figure 1.
Figure 1. Example laboratory PIV set-up, where a camera images a laser light sheet that is illuminating a flow field seeded with particles.
As shown in Figure 1, laser sheet optics (i.e., a cylindrical lens) focus the collimated laser beam into a thin laser sheet. The laser sheet illuminates neutrally buoyant particles (i.e., seeds) that have been introduced into the flow. These small (1-100 $\mu m$) seeds both follow the flow field and effectively scatter the laser light so that they can be seen and imaged. A digital camera (directed orthogonal to the laser sheet plane) images the particles in the plane as shown in Figure 2. The velocity can be calculated by taking the displacement of the particles divided by a known time step. In general, the particles shouldn’t move too much, the flow field being images should be the same between time steps and the time should be small enough that the particles have only moved a few particle diameters.
Figure 2. Image of a flow field that was illuminated with a laser light-sheet and imaged by a camera and the mi-PIV application. The neutrally buoyant particles in the flow field appear white as they reflect the laser light.
By taking two images of the same space of a flow field a small, known time apart, the displacement of the particle position over time (i.e., velocity) can be computed.
PIV Algorithms
Instead of tracking the movement of each individual particle (known as PTV), PIV algorithms segment each of the two images into smaller images referred to as interrogation regions (Figure 3).
Figure 3. Main image 1 (from Figure 2) partitioned into interrogation regions.
Basic PIV algorithms employ statistical correlation methods to identify the most likely location of a set of particles as they move within an interrogation region in the first image to an interrogation window in the second image. The “search area” is the interrogation region in the second image. The translation of a set of particles is identified by finding the change in x and y location from the first image to the second image. PIV algorithms identify the velocity of the fluid in each region by dividing this translation by the prescribed time between images.
More accurate PIV measurements result when particle movements are contained within overlapping interrogation regions, since a small search radius increases the likelihood of maintaining valid correlations (i.e., correlating the translation of the same particles). The likelihood of a poor or erroneous correlation increases as the “search area” increases. In other words, the farther you move from the particles’ initial position in the first interrogation region, the more likely you are to correlate to the wrong set of particles in the second interrogation region.
References:
[1] B. L. Smith and D. R. Neal, “Particle Image Velocimetry,” Part. Image Velocim., p. 27, 2016.
Author: Jack Elliott
Date Published: June, 2022