dc.description.abstract |
A comprehensive understanding of the mechanism of particle entrainment is crucial to
obtain accurate prediction of sediment transport rates. However, due to complexity of the
parameters (e.g., particle shape, bed morphology, particle protrusion) involved in the particle
entrainment process, a generally accepted criterion for the particle entrainment condition
remains elusive. An inertial measurement unit (IMU) can directly measure particle inertial
dynamics during particle movement, thus providing novel insights into particle entrainment.
This thesis describes the use, including detailed calibration, of a Smart Sediment Particle
(SSP) with instrumented IMU sensors for measuring the particle’s acceleration during the
entrainment (particle dislodgement) process. Simultaneous two-dimensional particle tracking
technology (PTV) was used to obtain the flow field. A series of laboratory experiments were
conducted to study the particle entrainment on a regular roughened bed composed of
hexagonally packed spheres. Three different types of bed arrangement types (BA-1, BA-2, and
BA-3) are defined regarding the alignment angle between the main flow direction and the local
packing reference line (0⁰, 30⁰, and 60⁰ for BA-1, BA-2, and BA-3 respectively). The effect of
these bed arrangements and the relative particle protrusions P/D (P is the protrusion height and
D is the diameter of the target particle) on particle entrainment was examined.
Experimental results show that bed packing can significantly affect the particle’s
entrainment threshold and the subsequent motion. For the three bed arrangements (BA-1, BA-
2 and BA-3), the average ratios of the lateral force impulse to the resultant force impulse at
entrainment were measured to be 9.2%, 32.6% and 46.4%; regarding the longitudinal impulse,
the corresponding ratios were 54.0%, 25.6% and 10.7%. The data show that lateral movement
is predominant for some cases because the entrainment path is based on the least work done by the flow, even if the direction is not aligned with the flow. A concept of effective protrusion
height is proposed to quantify the effect of bed packing on entrainment.
The SSP explicitly captures the particle motion phases (i.e., no motion, vibration, and
entrainment). Acceleration results show larger maximum accelerations and longer duration of
motion for cases with greater obstruction by surrounding bed roughness, which entails larger
momentum transfer required to initiate the entrainment. Further analysis on instantaneous flow
structures indicates that the occurrence and persistence of the “sweep-outward interaction”
quadrant motion sequence (i.e. Q4-Q1 sequence) is the key to momentum transfer and
entrainment. During vibration events, the passage of anti-clockwise vortices may induce a
transient Q4-Q1 sequence, but the short duration does not lead to complete entrainment. The
Q4-Q1 sequence may be caused by the lower half portion of anti-clockwise vortices or it may
occur independently.
The protruding target particle substantially changed the spatial distributions of timeaveraged
velocities, turbulent kinetic energy, and Reynolds shear stresses, compared to those
on a flat rough bed (without protruding bed particles). Above the target particle, a zone of low
Reynolds shear stress was observed. A prevalence of sweep (Quadrant 4) and outwards
interaction (Quadrant 1) events near the protruding particle indicates that the particle
dislodgement is associated with a positive fluctuation of streamwise velocity. Furthermore, this
result implies that the drag force related to strong streamwise velocity may play a more
significant role in protruding particle entrainment than the shear stress.
The measurement of inertial particle forces was conducted for particle entrainment at
different relative protrusion heights. Experimental results show that the inertial drag and lift
forces required to initiate particle entrainment systematically decrease as relative particle
protrusion increases. The ratio of inertial lift force to drag force reveals that drag force slightly dominates the entrainment process at P/D > 0.7 while lift force slightly dominates at P/D <
0.62. Inertial drag and lift coefficients were computed from the force and local velocity data.
The inertial drag force coefficients were found to be independent of particle protrusion for low
P/D, but rapidly declined for P/D > 0.7. Differently, the inertial lift force coefficients remain a
relatively constant value of 0.043. Additionally, the critical Shields number is linearly related
to the normalised inertial forces for the present particle protrusion range. In summary, the
inertial forces demonstrate a significant impact of protrusion on the particle entrainment
process.
In general, the SSP shows the potential to analyse the particle entrainment process both
qualitatively and quantitatively. Also, this study is the first to apply a combined analysis of
inertial particle forces and instantaneous flow structures, thus enhancing the understanding of
the complex particle entrainment mechanism. |
|