A complex phase change and fluid flow situation exists in nuclear reactors, particularly boiling water reactors, when the fuel element surfaces become highly fouled as has been the case in several instances. Perhaps the worst case to date was at the River Bend station in Louisiana several years ago. Question: How may CFD be employed to accurately describe the flow and heat transfer processes? Our great federal government has spent well over a billion dollars on CFD codes such as the TRAC and RELAP series. However, these codes are useless in producing realistic evaluations of phase change heat transer and fluid flow in the world's fleet of over 400 nuclear power reactors.

I think the fouling question goes far beyond CFD.

If you knew exactly the thermal resistance added by the fouling - and you knew the thickness of the fouling so you could factor that into the changes in the flow in that portion of the tube - and if you knew how to model the boiling at the fouled surface quantitatively - and if you knew the proper turbulence modeling to use in the tubes near the fouling, you would then be ready to pick a CFD technique/code to try to simulate the fouling process and (perhaps) estimate the impact on heat transfer. In other words, this is a lot more than a CFD problem. The fundamental physical processes are more or less quantifiable, depending on the particular subsystem.

So, "How may CFD be employed to accurately describe the flow and heat transfer processes?" Start by covering the physics, then ask how to obtain solutions to the resulting simulation problem. At that point, CFD techniques may be considered. Probably, considering the age of the TRAC and RELAP systems, their physics submodels are not up to date.

Anyway, entire engineering/scientific meetings have been devoted to your question. Evidently there are no answers yet.

It's not clear how the last part of your post relates to your technical problem. It is likely true that the US Government spends enormous sums on CFD work. (By the way, large sums are spent on items of much less value as well!) The cost of the nuclear weapons design codes has to be huge, with maintenance and development going back to the 50's. NASA's aerodynamic development work is carried on in a big way at at least three centers, with lessor efforts at least half a dozen other locations. And NOAA, the Department of Energy, and the Nuclear Regulatory Commission [probably the place your question should be considered] all do at least some physical modeling (and CFD) from time to time.