Wall-PIV

At this point in time, an efficient measurement system does not exist for the acquisition of near wall velocity distributions in flows (which, for example, show up in medically relevant problems) and the resulting derived values, such as sheer rate or sheer stress. These flows, among others, are investigated in biofluid mechanics (blood pumps, models of blood vessels) in order to reduce medical complications. Knowledge of the wall sheer stress fields in blood pumps and artificial heart valves is of great practical value, since a direct connection exists between wall sheer stress and the formation of thrombi. The near wall flow influences the progress of the deformations in aneurysms. This is why the measurement of the flow conditions in the previously named flow models is necessary for achieving medical progress.
Frequently, this concerns flows along vaulted and flexible walls where conventional systems do not work. Further examples where knowledge of the near wall velocity fields is important are the removal of biofilms in the medical field, the cleaning of containers in the food industry, the segmental streaming of various substances during chemical analysis and the cultivation of shear stress sensitive cells in the field of biotechnology.

For the previously non-researchable areas in the vicinity of vaulted walls, the Biofluid Mechanics Lab is developing the Wall-PIV-Method. This method concentrates on the observation and analysis of flows close to the wall and is a further development of Particle Image Velocimetry (PIV) for the above mentioned special case.
During Wall-PIV, light reflecting particles (tracers) which follow the flow are added to the fluid and the model is illuminated from the front with diffuse light. By adding a coloring agent to the fluid model, the penetration depth of the light can be controlled and therefore be restricted to the depth of the near wall region of interest.
The velocity distribution of the near wall area of interest can be determined from the recorded particle movements.

Figure 1: The functional principle of Wall-PIV: (Left) sketch of the illuminated flow field and setup of the camera and the light source outside of the transparent model. (Middle) an enlargement of the illuminated flow field of the sketch on the left, where the distance to the wall of the particles A, B and C is marked. (Right) stylized measurement result: b/w-image section with the gray values of the particles, which result from the distance to the wall.

The two velocity components normal to the optical axis can be obtained by the position of the particle within the image, the third component results from the gray values of the particle. In accordance with Beer-Lambert’s Law, there is an exponential relationship between the gray value of a particle and the path length of the light through the absorbing fluid. The closer a particle is to the wall, the brighter it appears to the observer.

Figure 2: Maximum-minimum images for each of the three orientations of two cerebral aneurysm models. A maximum image is comprised of the brightest gray value for each pixel (x,y) of all of the images in the measurement series at this position. Accordingly, a minimum image is comprised of the darkest gray value. For a maximum-minimum image, the minimum image is subtracted from the maximum image of the measurement series. For stationary areas, the maximum image matches the minimum image, resulting in the masking out of these areas. The flow paths of the particles remain, which depict a complex flow in the shown aneurysm.

Figure 3: Front view of the cerebral aneurysm shown in the bottom row in the previous figure. Next to the previously shown maximum-minimum image is an image from the sequence of the near wall flow field calculated with the optical flow (OF) method. Based on the particle path lines, it is clearly visible that the angle error of the estimated flow field is marginal. An exception is the inflow of the model, which was not within the area of depth of focus of the camera.

In cooperation with the Heidelberg Collaboratory for Image Processing (HCI) of the Interdisciplinary Center for Scientific Computing (IWR) of the University of Heidelberg, an evaluation process is being developed based on the optical flow method, which allows the determination of all three velocity components for particles close to the wall and the recording of three-dimensional particle path lines.
With the knowledge of the velocity field and the respective distance to the wall, it is possible by using Newton’s theory of viscosity to determine the wall shear stress.
The method was optimized and validated for complex flows with flat walls. Measurements were carried out using many complex, stationary and pulsating flows along curved and vaulted surfaces. These were then compared with CFD simulations. The first measurements using blood pumps and models of aneurysms are being carried out at this time. Preparatory work is being done on pulsating membranes (one to two Hertz).

Contact persons

Dr. Eng. Ulrich Kertzscher
Dipl.-Ing. André Berthe

Cooperation partner

Heidelberg Collaboratory for Image Processing (HCI) of the Interdisciplinary Center for Scientific Computing (IWR) of the University of Heidelberg

Links and Publications

• First entry in the GEPRIS-Databank of the DFG for funded projects
  
• Second entry in the GEPRIS-Databank of the DFG for funded projects
  

funded by