-
Numerical study of the effect of velocity perturbations on the
mechanics of vortex shedding in synchronized bluff-body wakes
S. BALABANI1 E. KONSTANTINIDIS1,2 C. LIANG3 G. PAPADAKIS1
1Department of Mechanical Engineering, Kings College London,
WC2R 2LS, UK
2Department of Engineering and Management of Energy Resources,
University of Western Macedonia, Kozani 50100, Greece
3Department of Mathematics, University of Glasgow, Glasgow G12
8QW, UK
-
Outline
Background and motivation Computational details Comparison with
experiments Observations from a typical simulation Effect of
excitation frequency Summary and remarks
-
Wake and vortex shedding control by periodic excitations
Rectilinear cylinder oscillations Rotational cylinder
oscillations Flow pulsation Acoustic forcing Vibration of control
rod in wake Surface suction/blowing Surface
modulation/deformation
Question: are there any universal features among the different
methods?
-
Flow configuration
0.21 /o mf U d Inflow with superimposed periodic velocity
oscillations
Circular cylinder perpendicular to the oncoming flow
( ) sin(2 )m eU t U U f t= +
Main parameters:
2 e
m
A UD f d
UU
=
*
0
e
e
UUf d
ff
=UDRe =
-
Forced oscillation studies (inline)
0.5 1.0 1.5 2.0 2.5 3.00.0
0.1
0.2
0.3
0.4
=
Flow Oscillations Cylinder OscillationsArmstong et al. (1986),
Re = 21
500 Griffin & Ramberg (1976), Re = 190Barbi et al. (1986),
Re = 3
000 Tanida et al. (1973), Re = 80Barbi et al. (1986), Re =
40
000 Tanida et al. (1973), Re = 4
000Konstantinidis et al (2003), Re = 2 150 Tatsuno (1972), Re =
100Konstantinidis et al (2005), Re = 2 150 Nishihara et al (2004),
Re = 17000
A UD 2 feD
fundamentalsynchronization
region
fe / fo
-
Vortex-induced inline vibration
Okajima et al (2004) Eur J Mech B/Fluids 23: 115-125
-
Forced vs. free streamwise vibration
In the middle of the lock-on region associated with no
excitation, i.e. zero energy transfer from the fluid to the
structure (Konstantinidis et al. 2005, JFM)
1.5 2.0 2.5 3.0 3.5 4.0
4 3.5 3 2.5 2 1.50.00
0.05
0.10
0.15
0.20
2nd responsebranch
1st responsebranch
A D
lock-onregion
fe / fo
U *= U/feD
fefnfefw
-
Objectives
Compare LES to PIV data at moderate Reynolds numbers
Determine forces (magnitude and phase) exerted on the cylinder
within lock-on range
Determine sign of energy transfer Improve understanding of flow
physics for
this class of problems by generalization of the results
-
Present simulations & previous experiments
1.5 2.0 2.5 3.0 3.5
4 3.5 3 2.5 2 1.50.00
0.05
0.10
0.15
0.20 PIV (Konstantinidis et al 2005) LES (present)
2nd responsebranch
1st responsebranch
A D
lock-onregion
fe / fo
U *= U/feD
-
Computational details
Unstructured collocated grid (746,688 cells)
3D LES (standard Smagorinsky model)
33 planes along span (L=D) Liang & Papadakis (2007)
Comp. Fluids
-
Flow statistics
Mean flow streamlines Mean vorticity
Streamwise mean velocity Reynolds stress
LES
PIV LES
Re 2150 2580
fe/fo 1.87 1.88
u/U 0.045 0.050PIV
-
Instantaneous flow
SimulationExperiment
-
3-D flow structureSpanwise vorticity
z /UD = 3
Streamwise vorticity (Mode B instability)
x /UD = 1
Spanwise correlation of velocity fluctuations in separating
shear layerAspect ratio effectsexperiment: 10D (end
walls)simulation: D (periodic b. condition)
-
Typical simulation
0
1
2
3
-1
0
1
0 10 20 30 400
45
90
135
180
0 10 20 30 400
45
90
135
180
CD CL
drag
t /Te
lift
t /Te
fe/fo =
1.88
-
Lissajous figures
1.0 1.5 2.0-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
1.0 1.5 2.0 1.0 1.5 2.0 1.0 1.5 2.0
2.092.001.88fe / fo = 1.77
CD
CL
-
Timing of vortex shedding
U(t)
fe/fo = 1.77
1.88
2.00
2.09
-
Force phase and energy transfer
1.7 1.8 1.9 2.0 2.1 2.2 2.30
60
120
180
E < 0
E > 0
fe/fo
drag (LES) lift (LES) Vmax(PIV)
U()
V()
-
Mean drag vs. excitation frequency
1.0 1.5 2.0 2.5 3.01.0
1.2
1.4
1.6
present LES, Re = 2580 Nishihara et al, Re = 17000
M
e
a
n
d
r
a
g
fe/fo
-
Drag amplification vs amplitude
Flow forcing
Present study
Jarza & Podolski (2004)
Konstantinidis et al (2005)
Streamwise oscillations
Tanida et al (1973)
Nishihara et al (2005)
Transverse oscillations
Tanida et al (1973)
Sarpkaya (1978)
Gopalkrishnan (1993)
Dong & Karniadakis (2004)0.0 0.2 0.4 0.6 0.8
1.0
1.5
2.0
2.5
3.0
3.5
4.0
0.00 0.05 0.10 0.15
Streamwise, Ax /D Transverse, Ay /D1+
8.2(
A x /D
)
C
D
m
a
x
/
C
D
o
1+2.1
(A y /D)
A/D
A/D
-
Effect of velocity perturbation
0.0 0.1 0.2 0.31.0
1.5
2.0
2.5
C
D
m
a
x
/
C
D
o
U/Uo
-
Summary
Comparison between computations and experiments shows agreement
on flow statistics instantaneous flow structure spanwise
correlation
Simulations revealed 3-D flow structure (Mode B type) primary
vortices fully-correlated along span (D) vortex-induced drag
in-phase with relative displacement (zero
energy transfer) Equivalence between inline cylinder
oscillations and inflow
fluctuations lock-on limits phase between induced forces and
relative flow/cylinder oscillation drag amplification
-
Closing remarks
Within lock-on region vortex shedding frequency is controlled by
the excitation frequency
Energy transfer between the fluid and the structure is
controlled by the timing of vortex shedding which is imposed by the
velocity perturbation
Zero-phase difference between force and relative displacement
when fe = 2fo for streamwise and fe = fo for transverse
oscillations
Drag amplification appears more pronounced in streamwisethan in
transverse oscillations
-
Acknowledgement
Financial support from the ECLAT group at Kings College London
for attending this conference
Numerical study of the effect of velocity perturbations on the
mechanics of vortex shedding in synchronized bluff-body
wakesOutlineWake and vortex shedding control by periodic
excitationsFlow configurationForced oscillation studies
(inline)Vortex-induced inline vibrationForced vs. free streamwise
vibrationObjectivesPresent simulations & previous
experimentsComputational detailsFlow statisticsInstantaneous
flow3-D flow structureTypical simulationLissajous figuresTiming of
vortex sheddingForce phase and energy transferMean drag vs.
excitation frequencyDrag amplification vs amplitudeEffect of
velocity perturbationSummaryClosing remarksAcknowledgement