80

Development of Prototype Lattice-Boltzmann Code for Computational Fluid Dynamics of Fusion Systems--MetaHeuristics, LLC, 209 W. Alamar Avenue, Suite A, Santa Barbara, CA  93105-3701; 805-886-1555, www.metah.com 

Dr. Martin J. Pattison, Principal Investigator, martin@metah.com 

Dr. Sanjoy Banerjee, Business Official, sanjoy@metah.com

DOE Grant No.  DE-FG02-03ER83715

Amount:  $749,632  

There is a need to predict the behavior of fluid systems in the extremely demanding conditions of very high heat and neutron fluxes, magnetohydrodynamic forces, and complex geometries found in fusion facilities.  Reliable simulations of this nature would reduce development time and costs, and allow rapid, cost-effective assessment of design concepts.  This project will develop a highly parallel code based on lattice Boltzmann (LB) methods to simulate the behavior of fluid systems in fusion applications, with emphasis on high heat fluxes, free-surface flows, magnetohydrodynamic effects, complex geometries, and impurity chemistry.  In Phase I, a fully three-dimensional lattice Boltzmann code was written and used to test the performance of the LB methodology and investigate critical modelling issues for fusion applications.  These issues included multicomponent flows and free surfaces, flow over complex boundaries, heat and impurity transport, and turbulence.  In Phase II, the lattice Boltzmann code will be enhanced to include magnetohydrodynamic effects, surface and volumetric chemical reactions, and pre- and post-processing packages.  Emphasis will be put on parallelisation, so the code can be used on distributed computing systems for large-scale fluid dynamic simulations of fusion problems.

Commercial Applications and Other Benefits as described by awardee:  In addition to the fusion application, the LB code should have applicability in numerous sectors where the flow of complex fluids plays an important role.  These sectors include the oil industry, chemical process plants, and thermal management systems (e.g., for electronics cooling).  Advantages of the lattice Boltzmann method over more conventional computational fluid dynamics (CFD) approaches are that it is very easy to parallelise, execution speed scales almost linearly with the number of processors used, and complex geometries can be handled easily.  This makes lattice Boltzmann methods key contenders for next generation codes for the rapidly growing (8-10% per year) CFD market, currently estimated to be $2 billion/year.