5
$\begingroup$

The Brinkman equations for steady flow of an incompressible fluid through rigid porous solid are:

$-\dfrac{\mu_0}{k}\mathbf{v} - \mathrm{grad}p + \mu_0 \Delta \mathbf{v} =0$ and $\mathrm{div}(\mathbf{v}) = 0$ in $\Omega$

where $k > 0$ is the permeability of the solid, $\mu_0 >0$ is the viscosity of the fluid, and $p,\mathbf{v}$ are the pressure and velocity fields.

Suppose that the domain is divided into two disjoint regions and $k$ assumes the values $k_0$ and $k_1$ on them.

My question: Is there a jump condition for this problem due to the discontinuity of $k$ ?

Ans: $\mathrm{jump}(\mathbf{t_n})=0$ where $\mathbf{t_n} := -p\mathbf{n} + \mu_0\nabla \mathbf{v}\;\mathbf{n}$

|cite|improve this question
$\endgroup$
1
$\begingroup$

There is a jump condition, but it is automatically satisfied by the solution (i.e., you don't need to do anything special). This is no different than solving $-\nabla \cdot k \nabla u = f$: there is a jump condition discussed in Jump condition for elliptic equation in standard finite element method but it is automatically satisfied.

|cite|improve this answer
$\endgroup$
  • $\begingroup$ Could you please elaborate what will be the expression of the jump condition ? If there is a jump, I am unable to see how the term $\mu \Delta\mathbf{v}$ will make sense. I had asked the linked question too. $\endgroup$ – me10240 Aug 31 '14 at 22:26
  • $\begingroup$ The physically correct equation will have the term $\nabla \cdot \mu \nabla \mathbf v$ instead of the one you list. The jump condition will then be obvious. $\endgroup$ – Wolfgang Bangerth Aug 31 '14 at 23:59
  • $\begingroup$ I looked at the derivation of jump conditions from a continuum mechanics text, and realized that to get the correct form, I need to multiply throughout by $k$ first. Then, I obtained the jump condition as: $ \mathrm{jump}(k\boldsymbol{t_n}) = 0 $ where $\boldsymbol{t_n} = [- pI + \mu_0(\nabla \mathbf{v}]\mathbf{n} $, where $\mathbf{n}$ is the unit normal. Could you please confirm if this is correct ? If so I shall add it to the thread as an answer and mark the issue solved. $\endgroup$ – me10240 Sep 1 '14 at 1:38
  • $\begingroup$ Yes, that seems right. It's the flux/stress that's got the jump condition and that's exactly what you state here. $\endgroup$ – Wolfgang Bangerth Sep 1 '14 at 12:35
  • $\begingroup$ Based on your answer to another question of mine on jump conditions, I think my jump condition stated above is wrong. It should not have the $k$ factor. I am adding the correct answer to my original question. $\endgroup$ – me10240 Sep 2 '14 at 16:30

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.