Академический Документы
Профессиональный Документы
Культура Документы
Aeroacoustics
Overview of Lecture
Noise Sources and Generation Mechanisms
Conclusions
Approach
Takeoff
Rotor-Stator Interaction
Fluid-Structure Interaction as a
Noise Generation Mechanism
Propagation in the Duct: Sound Must Propagate in a Duct: therefore only high
frequency acoustic modes will propagate.
Rotor-Stator Interaction
Wakes and Tip Vortices
Scaling Analysis
B U s
R
BR
=
(
)(
) = O (1 )
=
= BM >> 1
Ux
R
Ux
co
co
Fast Variables:
Slow Variables:
r
r
x
~
x=
r
r
x
x* =
R
Blade passing
frequency BPF=B
+ p '
+ '
r
u
<<
r
'
+ 0 u = 0
t
r
1
u
+
p '= 0
0
t
1
2
) p '= 0
2
2
c0 t
Spherical symmetry:
=
f (t m
R
R
)
c , R = xr - yr
Greens Function
r
r
G ( x , t | y, ) =
( t
R
)
c
Compact/Noncompact
Sources
Plane Waves
p ' = pe
r
u =
E =
r r
i ( k x t )
r
k
0 c0 k
p' '
20
I =
0u
2
=
r
r
I = p'u =
1
r r
i ( k x t )
p'
r
k
0 c0 k
2 0 c0
p'
=
2
0 c0
r
p' k
0c k
Point Source
Dipole: Force
Quadrupole: Stress
Acoustic Pressure
Acoustic Intensity
..
..
m (t R / c )
4 R
r.
r
F ( t R / c ). e r
4 Rc
..
T ij Vx
4 R 3 c 2
R
)
c
0 R 2c
(t
16
xx
i
dFi dFj
16 c R
..
dt
Directivity
Sperical symmetry
cos
dt
..
T ij T kl V 2 xi x j xk xl
16 R c
2
6 4
cos sin
Monopole Directivity
Dipole Directivity
Quadrupole Directivity
Scaling
Dipole
Quadrupole
Ratio
1
Thus at low Mach number dipoles or forces are more efficient sources of noise than
quadrupoles or turbulence. However, this result is valid only for uniform flows and low
frequency.
As the vortex travels near the trailing edge it is no longer convected by the
mean flow. Its trajectory crosses the undisturbed mean flow. This increases
the amount of fluid energy converted into acoustic energy. The acoustic
power scales with M3, much higher than that predicted by a dipole (M6).
mx2
cos
Amne
i ( kmnx1 t )
+ Bmne
i ( kmnx1 +t )
Dispersion Relation
k mn =
mx3
2
2
2 (m 2 + n 2 )
2
c
a
c
>
or < 2a
a
The velocity and pressure of
evanescent or cut-off modes are
out of phase and there is no net
transport of energy (I=0).
Phase Velocity
c ph =
c
1 2 (m 2 + n 2 )
a 2 2
d
c2
2
2
2
cg =
= c 1 (m + n ) 2 2
dk
a
Incident Gust
r
r
i ( k x + m ' t )
u g = a (mg ) ( r ) e g
, kg =
U
Cut-on Modes
= m ' = m '
2
V
r
r
u a = a m , n ( r ) e i ( k mn x + m t )
m ,n
m=m-qV
The circumferential modal number m for propagating modes for a rotor with B blades and a
stator with V blades is given by
m=pB-kV
Example: B=18, V=40. m=-22, -4, 10, 14, 22
D 02 p
2
+
p = 0
2
D t
p
= 0
r
2
)
~ 2 (1 M x2 )( m 2 + mn
(1 M x2 )
Where
~ = mMs
m = n = 0
K n ( mn r ) e i (k mn x + m t )
Conclusions
Fan noise sources: Flow nonuniformities and
irregular flow pattern.
Mechanism: Fluid/structure interaction and link to
unsteady aerodynamics.
Classical acoustics concepts are essential to
understanding and modeling of noise.
The coupling to the duct determines the modal
content of the scattered sound and affects sound
propagation.
T ij
2
2
2
c0 =
2
t
xix
v j
vi
2
)
+
xi
x j
3
ij
vk
xk
T ij
D0
2
2
c0 =
t 2
xix
r
r
u = u
D0
Dt
1
( 2
c0
r
u
+
= 0
v
2
D0
2
) = 0
2
Dt
D0
s'= 0
Dt
r
D0
+ U .
Dt
t
D0
p'= 0
Dt
Potential Mode:
r
r
r r
u = u ( x U i)
v
r r
r
r i(k x t )
u = ae
g
2
D
1
0 2 ) = 0
(
c 2 Dt 2
0
r r
r r i(k x t)
u = ae
g
Transverse Gust: (0,a2,0), (k1,0,0)
Oblique Gust:(0,a2,0), (k1,0,k3)
Vector Diagram Showing the Real and Imaginary Parts of the Response
Function S(k1,0,M) versus k1 for a Transverse Gust at Various M
Vector Diagram Showing the Real and Imaginary Parts of the Response
Function S(k1,k3,0.8) versus k1 for a Transverse Gust at Various M
Conclusions
At low Mach number, low frequency, dipole effects
(unsteady airfoil pressure) dominate the scattered sound.
At moderate and high Mach number and/or high
frequency, the scattered sound strongly depends on both
dipole and quadrupole effects and sound directivity is
characterized by lobe formation.
Loading strongly affects the scattered acoustic energy.
Exact nonreflecting boundary conditions are essential for
obtaining accurate results particularly at high Mach
number and reduced frequency.
=k1
M=0.3, k2
Magnitude (a) and Phase (b) of the first Harmonic Unsteady Pressure
Difference Distribution for the subsonic NACA 0006 Cascade Underhoing
an In-Phase Torsional Oscillation of Amplitude a=2o at k1=0.5 about
Midchord; M=0.7;
Magnitude (a) and Phase (b) of the first Harmonic Unsteady Pressure
Difference Distribution for the subsonic NACA 0006 Cascade Underhoing
an In-Phase Torsional Oscillation of Amplitude a=2o at k1=0.5 about
Midchord; M=0.7.
Conclusions
Turbulence Representation
r r
u ( x, t ) =
Fourier representation:
rr
r
r r
i ( k . x t )
a (k , )e
dk d
,k
r
r
r r r
ai (k )a j (k ' ) = ij (k ) (k k ' )
Velocity covariance
r r
R ij ( x , x ' , t ) =
r
r r
ui (x, t ) = ij ( x, k)a jeik1Ut
r ik Ut r
r r
r r
(0)
( x , k ) jl ( x ' , k ) k l ( k ) e 1 d k
r
k
*
ik
Source term
on blades
Blade unsteady
loading & radiated
sound field
Non-reflecting
boundary
conditions
Mathematical Formulation
Linearized Euler equations
Axisymmetric swirling mean flow
r r
r
r
U ( x ) = U x ( x , r )ex + U s( x , r )e
H = T S + U
f ( r )e
i ( t + m + k mn x )
Mode Spectrum
Spectral and Shooting Methods
Mx=0.55, M=0.24, M=0.21, =16, and m=-1
Normal Modes
Pressure-Dominated
Acoustic Modes
Propagating
Decaying
Nonreflecting
Boundary Conditions
Vorticity-Dominated
Nearly-Convected Modes
Singular Behavior
Computational
Domain
Outflow Conditions
Inflow Conditions
Formulation
Presssure at the boundaries is expanded in terms of the
acoustic eigenmodes.
r
p(x, t) =
c mn p mn ( , r ) e
i ( t + m + k mn x )
n=0
Causality
r
p(x,t) =
M 2
=M 2 n=0
c mn p mn ( r ) e
i ( t + m + k mn x )
L-1
[ p ] = [][c]
Computational
Domain
[ p ]L = []L [c]
Rotor/Stator Interaction
The rotor/stator system is
decoupled
The upstream disturbance can be
written in the form,
r
u I (r, ) =
p I (r, ) =
r
a m ( r ) e i ( m t )
m =
A m ( r ) e i ( m t )
m =
Quasi-periodic conditions
Wake discontinuity
Nonreflecting conditions
Numerical Formulation
The rotor/stator system is
decoupled
The upstream disturbance can be
written in the form,
r
u I (r, ) =
p I (r, ) =
r
a m ( r ) e i ( m t )
Two Vanes
m =
A m ( r ) e i ( m t )
m =
Tip
Quasi-periodic conditions
Wake discontinuity
Nonreflecting conditions
Hub
Computational Domain
Domain Decomposition
Inner and Outer Regions
O(R)
O(c)
Scattering Results
Narrow Annulus
rtip/rhub 1.0/0.98
0.5,1.0
1.5,2.0
2.5,3.0
3.5,4.0
Full Annulus
Data
rtip/rhub 1.0/0.7 M (mach number)
5
3.0
(disturbance)
B (rotor blades)
V (stator blades)
C (chord)
Stagger
L (length)
0.5
0.1
16
24
2/V
45o
3c
Computational Domain
B (rotor blades) 16
V (stator blades) 24
C (chord)
2/
V
3c
L (length)
nB rm
O (nB )
co
Acoustic Coefficients
m=-8, n=0,1
Acoustic Coefficients
m=-8, n=0,1
Mode m=-8
Downstream
n=0
Downstream
n=1
Upstream
n=0
Upstream
n=1
Mean
Flow 1
Mean
Flow 2
Mean Mean
Flow 3 flow 4
0.2015
0.1610
0.1375 0.1349
Cut off
0.0787
0.2140 0.1698
0.1363
0.0143
0.0313 0.0195
Cut off
0.0370
0.0763 0.0586
Acoustic Coefficients
Effect of Frequency
Mean Flow 4
Acoustic Coefficients
m=-8, n=0,1
Mode m=-8
Downstream
n=0
Downstream
n=1
Upstream
n=0
Upstream
n=1
2-D
Disturbance
3-D
Disturbance
0.1471
0.1094
0.0908
0.0617
0.0185
0.0143
0.0498
0.0418
Acoustic Coefficients
Vortical
Disturbance
Acoustic
Disturbance
0.1471
0.8608
0.0908
0.1458
0.0185
0.0718
0.0498
0.0775
Acoustic Coefficients
Three-Dimensional Effects
Comparison With Strip Theory
Effect of Swirl
Conclusions
Swirl affects the impedance of the duct and the number of
cut-on acoustic modes. The spinning modes are not
symmetric.
Strip theory gives good approximation as long as there is
no acoustic propagation.
Resonant conditions in strip theory are much more
pronounced than for 3D, i.e., lift variation in 3D is much
smoother along span.
Discrepancies between strip theory and 3D calculations
increase with the reduced frequency: it is a high frequency
phenomenon.