Posted Mon, 04 Jun 2007 05:23:45 GMT by qzhou
Hi, Peter,

I checked the flow and mass transport condition for a pumping well, and found there is no way to specify the mass condition with unknown concentration but known flow pumping rate. For flow, I specified the pumping rate at the nodes in 3D, and it's OK. For mass transport, because the concentration is part of the solution, we don't know the concentration in the pumped water. If we don't specify concentration at the pumping well, the concentration in the groundwater at the nodes of pumping wells is higher and higher because no solute mass is removed with pumped water. I think we have to specify a kind of boundary condition for transport (using the simulated concentration as mass concentration at the pumping-well nodes, automatically with the specified flow condition for pumping wells), do you know the easy way to implement?

Thanks a lot.

Posted Thu, 07 Jun 2007 09:00:42 GMT by Kerstin Kernbach
Dear Quanlin,

normally it should not be a problem to simulate the solute mass outflow only via a pumping well. There is no additionally mass transport boundary condition necessary.
I assume the reason why the results are not plausible is that you are using the divergence form for the transport.
In case of pumping wells it is better to use the convective form. Maybe you take a look into the FEFLOW online help. Here some explanations are given.
You can compare the settings also with the demo data, where an similar example is contained (demo_transport_3d.fem).


You must be signed in to post in this forum.