Fig. 1 PIV experimental equipment for oil liquid
The initial velocity of the oil inlet
was set to V0=0.0362 m/s, the corresponding Reynolds
number was 138.3, and the oil in the pipeline was in a laminar flow
state. Since the temperature did not change much during an experiment,
the influence of temperature was not considered. After the oil samples
to be tested were thoroughly mixed, they were fed into the test pipeline
through a peristaltic pump. After flowing through the pipeline
development section, a uniform liquid-solid two-phase suspension flow
was formed in the test section and then entered the oil tank through the
tail section to form a circulation. The system was run for 30 min before
the PIV test to ensure that the oil and particles were evenly mixed
during the PIV measurement period.
The PIV equipment used in the experiment was a 2DPIV system from TSI
Corporation of the United States. This system consists of a double
yttrium aluminum garnet laser (200 mJ, pulse frequency 15 Hz, laser
wavelength 532 nm), light guide arm, sheet light source lens (focal
length from 30-3000 mm, two sets of cylindrical lenses 15 mm F.L. and
-25 mm F.L., and divergence angles 25° and 14° , respectively), a Nikon
CCD camera (15 frames per sec, 50 mm / F1.8 lens, 2K×2K pixel
resolution) and TSI Insight4G image acquisition, analysis and display
software platform. The coordinate frame of the camera is
three-dimensional and is automatically controlled by a microcomputer
with a movement accuracy of 0.01 mm. The laser light emitted from the
laser passes through the light guide arm and the light source lens to
form a 1 mm thick sheet light, which is irradiated from the normal
direction (z-direction) of the pipe test section. Then, oil image
information is obtained by the camera positioned in the direction of the
pipe (y-direction). In addition, the synchronization controller (time
resolution 0.25 ns) is set to trigger the laser and the camera
synchronously.
As the particle size in oil is less than 100 μm, the particles have a
good following behavior and uniform distribution of light scattering, so
no tracer particles need to be added during the test. The acquired image
adopts the algorithms of multigrid iterative mutual correlation and
multiadaptive deformation window, and the image query area adopts
pipeline horizontal and vertical initial query windows of 72×64 pixels,
a final query window of 64×48 pixels and 50% overlap to obtain the
original velocity vector and eliminate the impact of the query window
boundary. The method of processing the error vector is to use a local
average verification method, and this vector is replaced by the mean in
the adjacent 5×5 pixel range. The measurement position is 300 mm from
the inlet of the pipeline; the measurement area is (flow
direction×normal direction) 85.86 mm×22.20 mm, while the center of the
area is (flow direction×normal direction) 300 mm×30 mm. In addition, 200
image frames are captured in a test period.