|
|
High Performance Transport for Reactor Simulations |
Todd Urbatsch, Jim Warsa, Jae Chang, Kent Budge, Jon Dahl,
Randy Baker, Rob Ward, Jeff Densmore, Scott Mosher, and
Tom Evans
Los Alamos National Laboratory
The reactor simulation community has enjoyed a stable foundation for three decades. With the current nuclear renaissance, new reactor designs and especially new fuel designs will shake that foundation. With fuel recycling and proliferation concerns, accurate estimates of material are required. Even steady-state neutronics need improvements in axial resolution and in pin power accuracy. Beyond steady-state, coupled physics are required to model the swelling of fuel and the bowing of control and fuel rods.
We present transport methods and software capabilities at Los Alamos National Laboratory (LANL) that will contribute to the nation’s reactor simulation requirements. PARTISN is a discrete ordinates transport code that has been in use for nearly 40 years for reactor and shielding simulation. It has been applied successfully to the OEC/NEA MOX Benchmark. PARTISN is a good complement to LANL’s widely used Monte Carlo code, MCNP. LANL also has two thermal radiation transfer research and software efforts: the Capsaicin Project, a discrete ordinates, massively parallel, Krylov-solved transport code on unstructured meshes; and the Jayenne Implicit Monte Carlo Project, a massively parallel, multiple geometry/dimension, Monte Carlo code. These capabilities are designed to couple to hydrodynamics physics and are thus well-suited to be extended to neutronics and fully coupled neutronics, fluid dynamics, and thermo-mechanics, which are required to model and predict the dynamic nature of nuclear reactors.
These two projects utilize effective software quality design and engineering, which are imperative for collaborating and coupling software with other laboratories in this national effort to accurately simulate nuclear reactors.
Contact
Todd Urbatsch
Computational Physics and Methods Group
Computer, Computational, and Statistical Sciences Division
Los Alamos National Laboratory
P.O. Box 1663, MS D409
Los Alamos, NM 87545
tmonster@lanl.gov