I just noticed that my previous response to your question seems to have gone missing. What I mean is that I would create an observation area that cuts through your structure and extends to the boundaries you are concerned with. Then after running a simulation you can view that area and determine if the fields get close to your boundaries or not.
Scott