Dear Rostislav, I have encountered the same problem, so you are not alone. Unfortunately, I have yet to find a solution for this : (
Best, Marc On Monday 02 November 2009 09:06:29 am Rostislav Khlebnikov wrote: > Dear Deal.II developers and users, > > I am using Cubit software for generating hexahedral meshes for use in > deal.ii-based solver. > > The problem I am working on now is the radio-frequency ablation of liver > tumors which includes both electromagnetic and thermal processes. > The electromagnetic part is a simple Laplace equation with time and > space-dependent electrical conductivity. > > Up until now the geometry of the needle that acts as an antenna during the > ablation, was simply subtracted from the domain and thus prescribing the > boundary condition (which is setting a certain voltage on the electrode > boundary) was not a problem. But excluding the needle from the domain is > obviously not a correct way of calculating EM field. Therefore right now > I'm trying to use multiple materials for 'liver tissue' and 'electrode'. It > is not a problem to create a UCD file such that these subdomains can be > distinguished within deal.ii code. However so far I didn't find a way to > pass the information about the parts of inter-material boundaries and/or > nodes at these parts of boundaries to deal.ii. The boundary marking doesn't > fit with deal.ii because it throws exception stating that internal quads > cannot be boundaries. I need this information to set the prescribed voltage > values at these nodes/boundaries (by the way - do I understand correctly > that for nodes within domain such constrains should be done using > ConstraintMatrix objects?). > > Is there any better way of finding these nodes other than creating a > special material id and then searching for nodes that belong to cells with > a certain combination of material id's? > > Best regards, > Rostislav.
_______________________________________________ dealii mailing list http://poisson.dealii.org/mailman/listinfo/dealii
