COMPUTATIONAL FLUID DYNAMICS (CFD)

Module – 18
Reynolds Averaged Navier-Stokes Equation
Eddy Viscosity Models
Module – 18

Reynolds Averaged Navier-Stokes Equation & Eddy Viscosity Model

• Reynolds Averaged Navier-Stokes Equation

• The Closure problem
• Boussinesq approach
• Eddy Viscosity Models
• K-Epsilon Turbulence Model
• K-Omega Turbulence Model
• Derivation of Turbulent Kinetic Energy Equation
Reynolds-Averaged Navier Stokes Equation
Reynolds-Averaged Navier Stokes Equation
The Reynolds-Averaged Navier–Stokes equations (incompressible, steady form) expanded in x, y, z components are:
Reynolds-Averaged Navier Stokes Equation
The Reynolds-Averaged Navier–Stokes equations (incompressible, steady form) expanded in x, y, z components are:
Reynolds-Averaged Navier Stokes Equation
The Reynolds-Averaged Navier–Stokes equations (incompressible, steady form) expanded in x, y, z components are:

These are additional unknowns in the RANS equations and

must be modeled, this is the closure problem in turbulence.
The Closure Problem

• We have 3 Momentum and 1 Pressure Equation. When the flow is Laminar we have 4 variables x, y, z velocity, and
pressure & 4 equations. Hence, the problem is closed.

• In the Turbulence flow problem, We have additional unknown stress terms in our momentum equation. But, the number of
equations is still 4. Hence no. of equations are not equal to the number of variables hence Problem remains unclosed.

• Now to close the problem either one has to get an additional equation or replace the unknown variables with suitable
known variables.

• We can derive the additional equations to model our Reynolds Stresses but it will end up adding more unknowns.

• The alternative way is to replace the unknown terms with known variables. This process is known as Turbulence Modelling.

• One of the famous Modelling methods is Eddy Viscosity Method. It is based on the Boussinesq hypothesis where Reynolds
Stress is related to mean velocity gradient.
Boussinesq hypothesis

• In 1887, Boussinesq proposed that the Reynolds Stresses (τij) can be related to the mean velocity gradients using turbulent
(eddy) viscosity, µt.

• Turbulence is characterized by isotropic eddy viscosity μt which enhance mixing between various layers of the flow.
𝝁𝒕 = 𝒄𝝁𝝆𝑽𝒕𝒍𝒕
• Based on the dimensional argument, Eddy Viscosity is proportional to the product of density, eddy velocity scale (Vt), and
eddy length scale (lt). 𝒄𝝁 is proportionality constant.

• Between Eddy length, time, and velocity scale, If any two quantities are known then the third quantity could be computed.

• Eddy or Turbulent viscosity μt is a flow property, not a fluid property.

Boussinesq hypothesis

𝝁𝒕 = 𝝆𝒄𝝁𝑽𝒕𝒍𝒕

• Momentum exchange through turbulence eddies. The largest integral eddies are responsible for mixing. They have a time
scale T (turn-over time) and a length scale L (size).

• Now we need two additional equations to describe L and T so that we can compute eddy viscosity.

• Once the eddy viscosity is computed, we can put it back into the Boussinesq formula and calculate the unknown Reynolds
Stress.

• In this way, we can close the RANS equation.

Laminar Mixing Layer: Molecular Viscosity
y High speed flow Downstream of plate
Inviscid Flow Mixing Layer U
• Negligible effect of viscosity
• No random motion of fluid No Mixing Layer
• No momentum transfer and mixing Flat Plate
• Sharp jump in U(y). x
• Discontinuous velocity profile u U>u
=0
Low speed flow

y
Downstream of plate
Laminar Flow Mixing Layer U
• Molecular viscosity
• Brownian motion (random mixing of fluid Laminar
molecules) Mixing Layer
• Momentum exchange/transfer through viscous x
Flat Plate u
effects
U>u ≠0
• Low speed and high speed flow molecules are
getting mixed.

Molecular viscosity x strain rate

Turbulent Mixing Layer: Eddy Viscosity
Turbulent Flow Mixing Layer
• Instabilities lead to turbulent flow and eddies
interactions. Turbulent large scale motion.

• Large eddies interact and generate smaller

eddies (Energy cascade)

• Large scale eddies leads to enhanced mixing

across layers and enhanced momentum
transport across fluid layers.

• Turbulent mixing is much stronger than

Laminar/Molecular mixing. y

U
• Analogy between molecular and turbulent
mixing.
x
Eddy Viscosity Formulation Flat Plate u
U>u
Turbulent Mixing Layer: Eddy Viscosity
Eddy Viscosity Formulation

Turbulent Flow Mixing Layer

• Fron dimensional analysis, viscosity has a length scale
square over time scale.

• Ratio of size of eddies to their turn over time.

• Large scale eddies drives the mixing so we need

information on large turbulent scales.

• Not interested in small scale eddies. y

• Size of eddies can be different at different location. U

They can have different turn over time as well.

• Required additional transport equation to compute x

Turbulent length and time scale on top of continuity, Flat Plate u
momentum & energy equation. U>u
• Foundation of One/Two Equation Eddy Viscosity
Turbulence Models.
One Equation Turbulence Model: Spalart-Almaras Model
• The Spalart–Allmaras model is a one-equation model that solves a modelled transport equation for the kinematic
eddy turbulent viscosity.

• The Spalart–Allmaras model was designed specifically for aerospace applications involving wall-bounded flows and
has been shown to give good results for boundary layers subjected to adverse pressure gradients.

• It is also gaining popularity in turbomachinery applications.

One Equation Turbulence Model: Spalart-Almaras Model
Two Equation Turbulence Model

𝝁𝒕 = 𝝆𝒄𝝁𝑽𝒕𝒍𝒕

• The most natural framework to compute the two independent scales required for the eddy-viscosity μt ~ L2∕T ~ k2∕ε ~ k∕ω.
• One turbulence scale is computed from the turbulent kinetic energy, k, which is provided from the solution of its transport
equation.

• The second scale is typically estimated from the dissipation rate of turbulence, ε, or its turn-over frequency, ω.

• From turbulent kinetic energy (k), we can get velocity scale.

• Dissipation rate (ε) will give a length scale or specific dissipation (ω) will give a time scale of integral eddy.

• Now using k and ε or ω, We can calculate the eddy viscosity.

The k- ε Model
The standard k–ε model (Launder and Spalding, 1974) has two model equations, one for k and one for ε, based on best
understanding of the relevant processes causing changes to these variables.

We use k and ε to define velocity scale and length scale representative of the large-scale turbulence as follows:

𝝁𝒕 = 𝝆𝒄𝝁𝑽𝒕𝒍𝒕
The k- ε Model
The k- ω Model

The model attempts to predict turbulence by two partial differential equations for two variables, k and ω, with the first
variable being the turbulence kinetic energy (k) while the second (ω) is the specific rate of dissipation/turbulence
frequency (of the turbulence kinetic energy k into internal thermal energy).

Proposed by: David C. Wilcox (1988, refined in 1993 and 2006)

The k- ω Model

The model attempts to predict turbulence by two partial differential equations for two variables, k and ω, with the first
variable being the turbulence kinetic energy (k) while the second (ω) is the specific rate of dissipation/turbulence
frequency (of the turbulence kinetic energy k into internal thermal energy).

Proposed by: David C. Wilcox (1988, refined in 1993 and 2006)

The closure coefficients and auxiliary relations are derived through a

combination of dimensional analysis, asymptotic behavior in
boundary layers, and empirical calibration using experimental and DNS
data to ensure model accuracy across a range of turbulent flows.
The k- ω Model
Derivation of Turbulent Kinetic Energy Equation
Derivation of Turbulent Kinetic Energy Equation
• Start with obtaining the equation of Reynolds Stresses.

• Take the difference between the Instantaneous Navier-Stokes equation and Reynolds Averaged Navier-Stokes Equation.

Instantaneous Navier-Stokes Reynolds Average Navier-Stokes

• By taking the difference, We can obtain the Equation of Reynold Stresses as shown below.
Derivation of Turbulent Kinetic Energy Equation

X - Component

Y - Component

• Multiply the above two-equation and apply time averaging then what is obtained is equation for Reynolds stress
as shown below.
Derivation of Turbulent Kinetic Energy Equation

• While analyzing the equation we can see that we have products of fluctuating velocity quantities i.e. Triple product of
fluctuating velocities, a third order correlation which is an unknown.

If we substitute value of i,j and k, We will end up having additional 27 unknowns.

• Now, Finding those 27 unknowns is almost impossible so we can not close the problem.

• We have to plan some alternative way, This is where Turbulent Kinetic Energy comes into the picture.
Derivation of Turbulent Kinetic Energy Equation

• In order to reduce the unknowns, We will follow one step of tensor algebra which is known as contraction.

• Contraction is taking the trace or the isotropic part of the Reynolds Stress Tensor Matrix.

• The isotropic Part is the diagonal element of the Reynolds Stress Tensor Matrix.

• If we make i = j, We can obtain the Isotropic Part.

• Substitute i = j = 1 and we obtain an equation for first normal stress, 𝒖 ′ 𝒖 ′
• Substitute i = j = 2 and we obtain an equation for second normal stress,v′ v ′
• Substitute i = j = 3 and we obtain an equation for third normal stress, w ′ w ′
• Summing up all 3 equations we will get an Equation for Turbulent Kinetic Energy, 𝒌.
Derivation of Turbulent Kinetic Energy Equation

• Turbulence Kinetic Energy (k) is the mean kinetic energy per unit mass associated with eddies in a turbulent flow.

• Physically, the turbulence kinetic energy (k) is characterized by root-mean-square (RMS) of velocity fluctuations.

• Turbulence Kinetic Energy (k) only contains the normal/isotropic part of the Reynolds Stress Tensor.

• The shear component or anisotropic part of the Reynolds Stress (Which is very important in strong rotating flows) is not
considered in the 2 Equation Model Turbulence Modelling Approach.

• After adding all the equations of normal stresses we will get an Equation for Turbulent Kinetic Energy, k as shown below.
Turbulent Kinetic Energy Equation

Change =Dissipation
in Time Pk=Production

turbulent Diffusion molecular Diffusion

Convective
Transport