Turbulence kinetic energy
 | This article needs additional citations for verification. (March 2009) |
| Turbulence kinetic energy | |
|---|---|
Common symbols | TKE, k |
| In SI base units | J/kg = m2⋅s−2 |
Derivations from other quantities | <math> k = \frac{1}{2} \left(\, \overline{(u')^2} + \overline{(v')^2} + \overline{(w')^2} \,\right) </math> |
In fluid dynamics, turbulence kinetic energy (TKE) is the mean kinetic energy per unit mass associated with eddies in turbulent flow. Physically, the turbulence kinetic energy is characterized by measured root-mean-square (RMS) velocity fluctuations. In the Reynolds-averaged Navier Stokes equations, the turbulence kinetic energy can be calculated based on the closure method, i.e. a turbulence model.
The TKE can be defined to be half the sum of the variances σ² (square of standard deviations σ) of the fluctuating velocity components: <math display="block"> k = \frac12 (\sigma_u^2 + \sigma_v^2 + \sigma_w^2 ) = \frac12 \left(\, \overline{(u')^2} + \overline{(v')^2} + \overline{(w')^2} \,\right), </math> where each turbulent velocity component is the difference between the instantaneous and the average velocity: <math> u' = u - \overline{u}</math> (Reynolds decomposition). The mean and variance are <math display="block"> \begin{align} \overline{u'} &= \frac{1}{T} \int_0^T (u(t) - \overline{u}) \, dt = 0, \\[4pt] \overline{(u')^2} & = \frac{1}{T}\int_0^T (u(t) - \overline{u})^2 \, dt \geq 0 = \sigma_u^2, \end{align}</math> respectively.
TKE can be produced by fluid shear, friction or buoyancy, or through external forcing at low-frequency eddy scales (integral scale). Turbulence kinetic energy is then transferred down the turbulence energy cascade, and is dissipated by viscous forces at the Kolmogorov scale. This process of production, transport and dissipation can be expressed as: <math display="block"> \frac{Dk}{Dt} + \nabla \cdot T' = P - \varepsilon, </math> where:<ref>Pope, S. B. (2000). Turbulent Flows. Cambridge: Cambridge University Press. pp. 122–134. ISBN 978-0521598866.</ref>
- <math>\tfrac{Dk}{Dt} </math> is the mean-flow material derivative of TKE;
- ∇ · T′ is the turbulence transport of TKE;
- P is the production of TKE, and
- ε is the TKE dissipation.
Assuming that molecular viscosity is constant, and making the Boussinesq approximation, the TKE equation is: <math display="block"> \underbrace{ \frac{\partial k}{\partial t}}_{\text{Local} \atop \text{derivative}} \!\!\! + \ \underbrace{\overline{u}_j \frac{\partial k}{\partial x_j}}_{\text{Advection} \atop {}}
= - \underbrace{ \frac{1}{\rho_o} \frac{\partial \overline{u'_i p'}}{\partial x_i} } _{\text{Pressure} \atop \text{diffusion}}
- \underbrace{ \frac{1}{2} \frac{\partial \overline{u_j' u_j' u_i'}}{\partial x_i} } _{{\text{Turbulent} \atop \text{transport}} \atop \mathcal{T}} + \underbrace{ \nu\frac{\partial^2 k}{\partial x^2_j} }_{ {\text{Molecular} \atop \text{viscous}} \atop \text{transport}} - \underbrace{\overline{u'_i u'_j}\frac{\partial \overline{u_i}}{\partial x_j} } _{\text{Production} \atop \mathcal{P}} - \underbrace{ \nu \overline{\frac{\partial u'_i}{\partial x_j}\frac{\partial u'_i}{\partial x_j}} } _{\text{Dissipation} \atop \varepsilon_k} - \underbrace{ \frac{g}{\rho_o} \overline{\rho' u'_i}\delta_{i3} } _{\text{Buoyancy flux} \atop b} </math>
By examining these phenomena, the turbulence kinetic energy budget for a particular flow can be found.<ref>Baldocchi, D. (2005), Lecture 16, Wind and Turbulence, Part 1, Surface Boundary Layer: Theory and Principles , Ecosystem Science Division, Department of Environmental Science, Policy and Management, University of California, Berkeley, CA: USA.</ref>
Computational fluid dynamics
In computational fluid dynamics (CFD), it is impossible to numerically simulate turbulence without discretizing the flow-field as far as the Kolmogorov microscales, which is called direct numerical simulation (DNS). Because DNS simulations are exorbitantly expensive due to memory, computational and storage overheads, turbulence models are used to simulate the effects of turbulence. A variety of models are used, but generally TKE is a fundamental flow property which must be calculated in order for fluid turbulence to be modelled.
Reynolds-averaged Navier–Stokes (RANS) simulations use the Boussinesq eddy viscosity hypothesis <ref>Boussinesq, J. V. (1877). "Théorie de l'Écoulement Tourbillant". Mem. Présentés Par Divers Savants Acad. Sci. Inst. Fr. 23: 46–50.</ref> to calculate the Reynolds stress that results from the averaging procedure: <math display="block"> \overline{u'_i u'_j} = \frac23 k \delta_{ij} - \nu_t \left( \frac{\partial \overline{u_i}}{\partial x_j} + \frac{\partial \overline{u_j}}{\partial x_i} \right), </math> where <math display="block"> \nu_t = c \cdot \sqrt{k} \cdot l_m. </math>
The exact method of resolving TKE depends upon the turbulence model used; k–ε (k–epsilon) models assume isotropy of turbulence whereby the normal stresses are equal: <math display="block"> \overline{(u')^2} = \overline{(v')^2} = \overline{(w')^2}. </math>
This assumption makes modelling of turbulence quantities (k and ε) simpler, but will not be accurate in scenarios where anisotropic behaviour of turbulence stresses dominates, and the implications of this in the production of turbulence also leads to over-prediction since the production depends on the mean rate of strain, and not the difference between the normal stresses (as they are, by assumption, equal).<ref>Laurence, D. (2002). "Applications of Reynolds Averaged Navier Stokes Equations to Industrial Flows". In van Beeck, J. P. A. J.; Benocci, C. (eds.). Introduction to Turbulence Modelling, Held March 18–22, 2002 at Von Karman Institute for Fluid Dynamics. Sint-Genesius-Rode: Von Karman Institute for Fluid Dynamics.</ref>
Reynolds-stress models (RSM) use a different method to close the Reynolds stresses, whereby the normal stresses are not assumed isotropic, so the issue with TKE production is avoided.
Initial conditions
Accurate prescription of TKE as initial conditions in CFD simulations are important to accurately predict flows, especially in high Reynolds-number simulations. A smooth duct example is given below. <math display="block"> k = \frac32 ( U I )^2, </math> where I is the initial turbulence intensity [%] given below, and U is the initial velocity magnitude. As an example for pipe flows, with the Reynolds number based on the pipe diameter: <math display="block"> I = 0.16 Re^{-\frac{1}{8}}. </math>
Here l is the turbulence or eddy length scale, given below, and cμ is a k–ε model parameter whose value is typically given as 0.09;
<math display="block"> \varepsilon = {c_\mu}^\frac34 k^\frac32 l^{-1}. </math>
The turbulent length scale can be estimated as <math display="block"> l = 0.07L, </math> with L a characteristic length. For internal flows this may take the value of the inlet duct (or pipe) width (or diameter) or the hydraulic diameter.<ref>Flórez Orrego; et al. (2012). "Experimental and CFD study of a single phase cone-shaped helical coiled heat exchanger: an empirical correlation". Proceedings of ECOS 2012 – The 25th International Conference on Efficiency, Cost, Optimization, Simulation and Environmental Impact of Energy Systems, June 26–29, 2012, Perugia, Italy. ISBN 978-88-6655-322-9.</ref>
References
Further reading
- Turbulence kinetic energy at CFD Online.
- Absi, R. (2008). "Analytical solutions for the modeled k-equation". Journal of Applied Mechanics. 75 (44501): 044501. Bibcode:2008JAM....75d4501A. doi:10.1115/1.2912722.
- Lacey, R. W. J.; Neary, V. S.; Liao, J. C.; Enders, E. C.; Tritico, H. M. (2012). "The IPOS framework: linking fish swimming performance in altered flows from laboratory experiments to rivers." River Res. Applic. 28 (4), pp. 429–443. doi:10.1002/rra.1584.
- Wilcox, D. C. (2006). "Turbulence modeling for CFD". Third edition. DCW Industries, La Canada, USA. ISBN 978-1-928729-08-2.
