Development of Multi-Frequency Multi-Scale Radiation Transport Modeling
Agency / Branch:
DOD / USAF
The objective of this proposal is to develop advanced radiation transport modeling techniques that accurately and efficiently treat transport in media having widely varying optical properties; in particular, hot gases and plasmas with optical depths ranging from the optically thin to the optically thick regimes. We will develop a hybrid diffusion-Monte Carlo (HDMC) model that efficiently transports multi-frequency radiation on multi-dimensional grids. During Phase I, we have successfully developed algorithms for the HDMC package, and demonstrated their accuracy and efficiency on simple 1-D grids. We have also investigated variance reduction, escape probability, and domain decomposition techniques for improving the efficiency and accuracy of the HDMC modeling. Algorithms developed during Phase I will be extended to support simulations on multi-dimensional grids during Phase II. Advanced techniques for treating: the interfacing between the diffusion and Monte Carlo models on non-orthogonal grids; domain decomposition for large-scale 3-D grids; and escape probability-based line transport will be developed and implemented. The models will be benchmarked against known solutions, and tested for efficiency and scalability to many-processor systems. Successful completion of this work will result in an efficient multi-scale multi-dimensional radiation transport package that accurately treats radiation flow in materials with realistic frequency-dependent radiative properties. BENEFIT: This project will develop state-of-the-art tools that numerically simulate the radiative properties of hot gases and plasmas. This capability is of substantial interest to a number of government laboratory research programs, including those supported by DOD and DOE. In addition, this project will lead to the development of tools that accurately simulate plasma spectral and radiative properties. Such tools are of significant value in developing commercial systems related to laser-induced breakdown spectroscopy (LIBS) and plasma radiation sources, such as those used in EUV lithography and medical research and technology.
Small Business Information at Submission:
Joseph J. MacFarlane
Research Institution Information:
Prism Computational Sciences, Inc.
455 Science Drive Suite 140 Madison, WI 53711
Number of Employees:
University of Wisconsin
21 North Park Street
Madison, WI 53715
E. Diane Barrett