0
$\begingroup$

I am looking for turbulent Navier Stokes equation for cylindrical coordinates.

I know that RANS (Reynolds Averaged Navier Stokes) eq. is the solution, I understood the point of it but only for Cartesian coordinates.

Without killer mathematical expressions, can I ask the formula ? You will see what I mean by saying "killer mathematical expressions" by looking at the picture I post here.

enter image description here

$\endgroup$
0
$\begingroup$

Perhaps you are looking for a coordinate-free RANS equation, which you can then write in any coordinate system of your choice. The actual incompressible NS equation, with usual choice for symbols, is: $$\nabla\cdot\tilde{\mathbf{u}}=0\\ \frac{\partial\tilde{\mathbf{u}}}{\partial t}+\tilde{\mathbf{u}}\cdot\nabla\tilde{\mathbf{u}}=-\frac{\nabla \tilde{p}}{\rho}+\nu\nabla^2\tilde{\mathbf{u}}$$ in which the quantities with the tilde are total quantities. We split the total quantity into the sum of mean and fluctuating parts: $$\left<\tilde{\mathbf{u}}\right>=\mathbf{U},\left<\tilde{p}\right>=P\\ \tilde{\mathbf{u}}=\mathbf{U}+\mathbf{u},\quad\tilde{p}=P+p\\ \left<\mathbf{u}\right>=\left<p\right>=0$$ in which $\left<\right>$ is the averaging operator, and being an ensemble average, commutes with the time and spatial derivative operators. Averaging the continuity equation gives: $$\left<\nabla\cdot\tilde{\mathbf{u}}\right>=0\\ \nabla\cdot\left<\tilde{\mathbf{u}}\right>=0\\ \nabla\cdot\mathbf{U}=0$$ Therefore $$\nabla\cdot\tilde{\mathbf{u}}=\nabla\cdot(\mathbf{U}+\mathbf{u})=0\\ \nabla\cdot\mathbf{u}=0$$

Similarly averaging the NS equation gives: $$\frac{\partial\mathbf{U}}{\partial t}+\left<\tilde{\mathbf{u}}\cdot\nabla\tilde{\mathbf{u}}\right>=-\frac{\nabla P}{\rho}+\nu\nabla^2\mathbf{U}$$ Now $$\left<\tilde{\mathbf{u}}\cdot\nabla\tilde{\mathbf{u}}\right>=\left<(\mathbf{U}+\mathbf{u})\cdot\nabla (\mathbf{U}+\mathbf{u})\right>\\ =\mathbf{U}\cdot\nabla\mathbf{U}+\left<\mathbf{u}\cdot\nabla\mathbf{u}\right>$$ Since $\nabla\cdot(\mathbf{u}\mathbf{u})=\mathbf{u}\cdot\nabla\mathbf{u}+\mathbf{u}\nabla\cdot\mathbf{u}=\mathbf{u}\cdot\nabla\mathbf{u}$, the RANS equation in coordinate-free notation is: $$\nabla\cdot\mathbf{U}=0\\ \rho\left( \frac{\partial\mathbf{U}}{\partial t}+\mathbf{U}\cdot\nabla\mathbf{U}\right)=-\nabla P+\mu\nabla^2\mathbf{U}+\nabla\cdot (-\rho\left<\mathbf{u}\mathbf{u}\right>)$$ in which $-\rho\left<\mathbf{u}\mathbf{u}\right>\equiv\sigma_R$ is the Reynolds stress tensor, which is obviously symmetric.

In the picture that you have posted $\sigma_R$ is written in Cartesian coordinates, whose orthonormal coordinate vectors are $\{\mathbf{e}_x,\mathbf{e}_y,\mathbf{e}_z\}$. The box you have drawn on the picture actually shows the matrix representation of $\sigma_R$ in that coordinate system. To write $\sigma_R$ in cylindrical coordinates you must adopt, at a given point in the flow field, orthonormal coordinate vectors $\{\mathbf{e}_r,\mathbf{e}_\theta,\mathbf{e}_z\}$ but the procedure of multiplying out to get the components of the tensor $\mathbf{u}\mathbf{u}$ is the same. While taking divergence however beware the fact that $\{\mathbf{e}_r,\mathbf{e}_\theta,\mathbf{e}_z\}$ are not constant vectors.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.