This is a read only copy of the old FEniCS QA forum. Please visit the new QA forum to ask questions

Is there a way to solve "moving boundary" problem??

+1 vote

I am trying to do the FEM simulation of temperature distribution in tissues while freezing. It is to be solved in three parts(i.e unfrozen, frozen and freezing phases). All three phases have different values of conductivity and specific heat. Also, the ice front will be continuously changing position.
The variational form of governing equation will be :

a = (c + omega*dt)*T*v*dx + dt*k*inner(nabla_grad(T), nabla_grad(v))*dx
L = (c*T_1 + dt*omega*tb + qmet*dt)*v*dx

with different values of "c", "omega", "k", "tb" and "qmet" depending on the temperature at that particular x.

Is there a way to solve this type of problem. I am new to fenics. Please help..!!
Thanks

asked Feb 10, 2014 by iitmayush FEniCS Novice (280 points)
...