0% found this document useful (0 votes)
137 views16 pages

Rans Derivation

1) The governing equations for turbulent flow include the time-averaged continuity, momentum, and energy equations. Additional terms accounting for velocity fluctuations must be included. 2) The k-ε model is commonly used to determine turbulent eddy viscosity, and involves solving transport equations for turbulent kinetic energy (k) and dissipation rate (ε). 3) Reynolds number indicates transition from laminar to turbulent flow, and turbulent flow equations are more complex due to random fluctuations that must be time-averaged.

Uploaded by

uday gouthama
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
137 views16 pages

Rans Derivation

1) The governing equations for turbulent flow include the time-averaged continuity, momentum, and energy equations. Additional terms accounting for velocity fluctuations must be included. 2) The k-ε model is commonly used to determine turbulent eddy viscosity, and involves solving transport equations for turbulent kinetic energy (k) and dissipation rate (ε). 3) Reynolds number indicates transition from laminar to turbulent flow, and turbulent flow equations are more complex due to random fluctuations that must be time-averaged.

Uploaded by

uday gouthama
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 16

Governing Equations for

Turbulent Flow
(i) Boundary Layer on a Flat Plate
ρU ∞ x
Re x = = Reynolds Number
µ
u / U ∞ = 0.99
Rex=0 Rex=5(10)5 Rex=106

層流區域 過渡區域 紊流區域

Thickness of
boundary layer
The Origin of Turbulence

• Turbulence is believed to be induced by small disturbance


in flow. In laminar flow region, the disturbance is
restrained by the flow, thus it does not affect the flow
motion. In turbulent flow region, the disturbance is
amplified by the flow motion to form “eddies”.
• For laminar flow, as the derived continuity, momentum
and energy equations are used for predicting flow velocity
and temperature distributions. The result appears to
match well with experimental data. Nevertheless, for
turbulent flow, this match usually appears to be poor.
• Reynolds number can be used for indicating the
occurrence of turbulent flow.
Fluctuation of velocity and temperature
in turbulent flow region
u

u at a certain location
t1
period
t (time)

Define:
u = u + u ' (t); v = v + v' (t); w = w + w' (t); P = P + P ' (t)
T = T + T ' (t)

1 t1 1 t1 1 t1
u ≡
t1 ∫0
u dt , v ≡
t1 ∫0
v dt , w ≡
t1 ∫0
w dt etc.
(ii) Time-Averaged Continuity Equation
∂u ∂v ∂w
Continuity: + + =0 (incompressible flow)
∂x ∂y ∂z

u = u + u ' (t); v = v + v ' (t); w = w + w' (t);

substituting into the above equation,


∂u ∂u ' ∂v ∂v ' ∂w ∂w'
+ + + + + =0
∂x ∂x ∂y ∂y ∂z ∂z

Integrating over period 0 → t1


∂u ∂u ' ∂v ∂v ' ∂w ∂w'
+ + + + + = 0
∂x ∂x ∂y ∂y ∂z ∂z
=0 =0 =0
∂u ∂v ∂w
+ + =0
∂x ∂y ∂z
(iii) Time-Averaged x-direction Momentum Equation

x-direction momentum equation (neglecting body force):


∂u ∂u ∂u ∂u 1 ∂Pe ∂ 2u ∂ 2u ∂ 2u
+u +v +w = − +ν ( + + )
∂t ∂x ∂y ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2

Multiplying continuity equation by u:


∂u ∂v ∂w
u( + + ) = 0 adding to the above equation
∂x ∂y ∂z

∂u ∂u 2 ∂ (uv) ∂ (uw) 1 ∂Pe ∂ 2u ∂ 2u ∂ 2u


+ + + = − +ν ( + + )
∂t ∂x ∂y ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2
averaging over period 0 → t1

∂u ∂u 2 ∂ (uv) ∂ (uw) 1 ∂Pe ∂ 2u ∂ 2u ∂ 2u


+ + + = − +ν ( + + )
∂t ∂x ∂y ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2

=0

∂u 2 ∂ (uv) ∂ (uw) 1 ∂ Pe ∂ 2u ∂ 2u ∂ 2u
+ + = − +ν ( + + )
∂x ∂y ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2

u = u + u ' (t); v = v + v' (t); w = w + w' (t);

substituting these three expressions into the above equation,

∂ (u ) 2 ∂ (u ' ) 2 ∂ (u v) ∂ (u 'v ' ) ∂ (u w) ∂ (u ' w' ) 1 ∂ Pe ∂ 2u ∂ 2u ∂ 2u


[ + ]+[ + ]+[ + ]= − +ν ( + + )
∂x ∂x ∂y ∂y ∂z ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2
where
∂ (u ) 2 ∂u ∂ (u v) ∂v ∂u ∂ (u w) ∂w ∂u
= 2u ; =u + v ; =u + w
∂x ∂x ∂y ∂y ∂y ∂z ∂z ∂z
Substituting back,

∂u ∂ (u ' ) 2 ∂v ∂u ∂ (u 'v ' ) ∂w ∂u ∂ (u ' w' )


[2u + ] + [u + v + ]+[ u + w + ]
∂x ∂x ∂y ∂y ∂y ∂z ∂z ∂z
1 ∂ Pe ∂ 2u ∂ 2u ∂ 2u
= − +ν ( + + ) (1)
ρ ∂x ∂x 2 ∂y 2 ∂z 2

Multiplying the time-averaged continuity equation by u ,


∂u ∂v ∂w
u ( + + ) = 0
∂x ∂y ∂z
Substracting equation (1) with this equation, it yields
∂u ∂ (u ' ) 2 ∂u ∂ (u 'v ' ) ∂u ∂ (u ' w' ) 1 ∂ Pe ∂ 2u ∂ 2u ∂ 2u
[u + ] + [v + ] + [w + ]=− +ν ( + + )
∂x ∂x ∂y ∂y ∂z ∂z ρ ∂x ∂x 2 ∂y 2 ∂z 2
After arrangement,

∂u ∂u ∂u 1 ∂ P ∂ (u ' ) 2 ∂ (u 'v ' ) ∂ (u ' w' )


u + v + w + = ν (∇ u ) −
2 − −
∂x ∂y ∂z ρ ∂x ∂x ∂y ∂z

Additional terms
∂ 2u ∂ 2u ∂ 2u
where ∇ 2u ≡ + +
∂x 2 ∂y 2 ∂z 2

Similarly, y and z-directional momentum equations can be


derived as well!

In tensor notation, i-direction time-averaged momentum equation is


∂ui 1 ∂P ∂ 2ui ∂ (ui' v 'j )
uj + = ν( ) −
∂x j ρ ∂x ∂x j ∂x j ∂x j
For two-dimensional turbulent boundary layer, the x-momentum equation is,
∂u ∂u 1 ∂P ∂ 2u ∂ (u 'v' )
u + v + ≈ ν −
∂x ∂y ρ ∂x ∂y 2 ∂y
∂ 2u ∂ 2u ∂(u ' )2 ∂(u 'v' )
where  and 
∂x 2 ∂y 2 ∂x ∂y
(velocity fluctuation in the y direction is larger than that in the x direction)
Negative value

∂u ∂u 1 ∂P 1 ∂ ∂u
u + v + = (µ − ρ u 'v' )
∂x ∂y ρ ∂x ρ ∂y ∂y
where
∂u ∂u
µ = molecular shear stress ; − ρ u 'v' = eddy shear stress ≡ ρε M ( )
∂y ∂y
ε M ≡ eddy momentum diffusivity

[Note: velocity fluctuations (u ' , v' ) are assumed to be induced by ∂u / ∂y.]


Hence,

∂u ∂u 1 ∂P 1 ∂ ∂u ∂u
u + v + = (µ + ρε M )
∂x ∂y ρ ∂x ρ ∂y ∂y ∂y
1 ∂ ∂u
= [ ρ (ν + ε M ) ]
ρ ∂y ∂y
∂ ∂u
= [(ν + ε M ) ]
∂y ∂y

∂u
where ρ (ν + ε M ) = apparent shear stress
∂y
(iv) k−ε Model for Turbulent Flow

„ εM is not a constant. For solving the momentum equation,


its value must be determined.

„ k−ε Model Involves


(i) Turbulence kinetic energy (k) equation
(ii) Dissipation energy (ε) equation

„ The above two equations plus the momentum equation


can be used to solve the εM value and velocity distribution.
Analog to molecular kinetic theory for gas:

ε M = Cu k 1/ 2 L

where k = turbulence energy ; L = length scale ; Cu = empirical coefficient

k 3/ 2
Dissipation rate = ε = C D ( ) where CD = drag coefficient = 1
L
k2
Sustituting back to eliminate L, it yields εM = ( )Cu
ε

Dk ∂ ε M ∂k ∂u
k − equation: = ( ) + ε M ( )2 − ε
Dt ∂y σ k ∂y ∂y
Dε ∂ ε M ∂ε ∂u 2 ε ε2
ε − equation: = ( ) + c1ε M ( ) − c 2 ( )
Dt ∂y σ ε ∂y ∂y k k

The following values are suggested for :


Cu = 0.09, c1 = 1.44, c 2 = 1.92, σ k = 1, σ ε = 1.3
(v) Time-Averaged Energy Equation
Energy equation (neglect dissipation term):
∂T ∂T ∂T ∂T ∂ 2T ∂ 2T ∂ 2T
ρcp ( +u +v +w ) = k( + + )
∂t ∂x ∂y ∂z ∂x 2 ∂y 2 ∂z 2

where u = u + u ' (t) ; v = v + v' (t) ; w = w + w' (t)


P = P + P ' (t) ; T = T + T ' (t)

Multiplying continuity equation by T:


∂u ∂v ∂w
T( + + ) = 0 adding to the above equation
∂x ∂y ∂z

∂T ∂ (uT ) ∂ (vT ) ∂ ( wT ) ∂ 2T ∂ 2T ∂ 2T
+ + + = α( + + )
∂t ∂x ∂y ∂z ∂x 2 ∂y 2 ∂z 2
Time averaged from 0 → t1 ,

∂T ∂ (uT ) ∂ (vT ) ∂ ( wT ) ∂ 2T ∂ 2T ∂ 2T
+ + + = α( 2 + 2 + 2 )
∂t ∂x ∂y ∂z ∂x ∂y ∂z

∂ (uT ) ∂ (vT ) ∂ ( wT ) ∂ 2T ∂ 2T ∂ 2T
After arrangement, + + = α( 2 + 2 + 2 )
∂x ∂y ∂z ∂x ∂y ∂z

∂T ∂u ∂T ∂v ∂T ∂w
[u + T ] + [v + T ] + [w + T ]
∂x ∂x ∂y ∂y ∂z ∂z
∂ (u 'T ' ) ∂ (v 'T ' ) ∂ ( wT
' '
) ∂ 2T ∂ 2T ∂ 2T
+ + + = α( 2 + + )
∂x ∂y ∂z ∂x ∂y 2
∂z 2

For two-dimensional flow (also using continuity equation), it yields :


∂T ∂T ∂ (v 'T ' ) ∂ 2T
u + v + = α 2
∂x ∂y ∂y ∂y
∂ 2T ∂ 2T ∂ (u 'T ' ) ∂ (v 'T ' )
[Note: In the boundary layer,  ;  ]
∂x 2
∂y 2
∂x ∂y
For two-dimensional flow in the boundary layer:
∂T ∂T ∂ 2T ∂ (v 'T ' )
u + v = α 2 −
∂x ∂y ∂y ∂y
1 ∂ ∂T ∂ ∂T
= [k − ρ c p v 'T ' ] = [(α + ε H ) ]
ρ c p ∂y ∂y ∂y ∂y
where
∂T
k = molecular heat flux (without considering direction)
∂y
∂T
− ρ c p v 'T ' = eddy heat flux ≡ ρ c pε H
∂y
ε H ≡ eddy thermal diffusivity ; α = molecular thermal diffusivity

∂T ∂T ∂ ε ∂T
u + v = [(α + M ) ]
∂x ∂y ∂y Prt ∂y

where Prt ≡ turbulent Prandtl number


eddy momentum diffusivity ε M
≡ = (empirally determined)
eddy thermal diffusivity εH

You might also like