The DOE Accelerated Strategic Computing Initiative: Enabling the tools for predictive materials modeling and simulation
Abstract
The objective of the DOE Science-Based Stockpile Stewardship (SBSS) program is to ensure confidence in the performance, safety, and reliability of the U.S. nuclear stockpile on the basis of a vigorous science-based approach without nuclear testing, in compliance with the comprehensive test-ban treaty. A critical element of this approach is the development of predictive, first-principles, full-physics computer simulation tools. In support of the SBSS program, the DOE has launched the Accelerated Strategic Computing Initiative (ASCI) to enable these computational developments and to promptly shift from an \underlineempirical test-based methodology to a \underlinepredictive simulation-based approach. In particular, the development of advanced materials simulation capabilities to predict the effects of materials properties -- as these properties change as a result of aging and/or re-manufacturing -- on stockpile performance has explicitly been identified as one of the most critical component of the SBSS program. Consequently, the emerging SBSS program at the national laboratories presents unprecedented opportunities and challenges for solving important materials physics problems of significance to national security. A key element in the development of predictive materials simulation capabilities is the establishment of rigorous theoretical links between ab initio quantum-based descriptions at the electronic and atomic levels and engineering continuum-based treatments at the macroscopic scale. These links can be established through the identification of the appropriate degrees of freedom which determine the materials response. Applications which illustrate the use of advanced materials simulation methods for the prediction of the thermodynamical and mechanical properties of materials as they afford to bridge the length-scale gap between different levels of descriptions will be presented.
- Publication:
-
APS Meeting Abstracts
- Pub Date:
- August 1997
- Bibcode:
- 1997APS..CPC...F01M