ITR/AP: Multiscale Models for Microstructure Simulation and Process Design

University Of Illinois At Urbana-Champaign, Urbana IL

Investigators

Abstract

This grant is the result of a proposal submitted to the Information Technology Research (ITR) Initiative. The award is co-funded equally by the Divisions of Materials Research and Advanced Computation Infrastructure and Research. The research is an interdisciplinary effort to simulate the evolution of microstructure during materials processing, including the effects of microstructure on bulk material properties. The project advances a set of mutually-informative models that, collectively, span atomistic to macroscopic length scales and that couple thermal, chemical and mechanical response. It will lead to improved predictive capabilities for optimizing existing materials systems, and to the prospect of new engineered materials and processes. The rich physical basis of the models makes them computationally intensive, so a closely-coupled program of information technology research is planned to support applications research. The project brings together engineers, materials scientists, computer scientists and mathematicians. It links two existing NSF-sponsored research centeres at Illinois: the Center for Process Simulation and Design (CPSD), and the Materials Computation Center (MCC). This link is strategic, because it reflects the significant coupling between the meso- and macroscopic phenomena that CPSD studies and the largely atomistic behavior that MCC investigates. Three particular applications, listed in order of ascending scale, are (i) coupled quantum and continuum models of material interfaces; (ii) dendritic solidification with fluid flow in metallic microstructures; and (iii) microstructure evolution and process optimization in extrusion and quench processes. The target applications are, in many respects, distinct. Yet, from the perspective of computational science and information technology, they pose a number of common challenges. These include problems with moving boundaries and interfaces; coupled and heterogeneous physical models at highly disparate length scales; and, computational complexity that calls for massively parallel and adaptive analysis techniques, as well as improved iterative solution methods. The group structure of this grant allows for sharing solutions across disciplines and the possibility of investigating more alternative solutions for each application. The associated information technology includes: domain-specific abstraction frameworks that enhance programmer productivity in parallel applications development; run-time load-balancing strategies for heterogeneous parallel applications exhibiting dynamic behavior; formulation and analysis of space-time discontinuous Galerkin methods, space-time mesh generation and 4-D visualization; and, new techniques for interface tracking and variable-topology shape optimization. %%% This grant is the result of a proposal submitted to the Information Technology Research (ITR) Initiative. The award is co-funded equally by the Divisions of Materials Research and Advanced Computation Infrastructure and Research. The research is an interdisciplinary effort to simulate the evolution of microstructure during materials processing, including the effects of microstructure on bulk material properties. The project advances a set of mutually-informative models that, collectively, span atomistic to macroscopic length scales and that couple thermal, chemical and mechanical response. It will lead to improved predictive capabilities for optimizing existing materials systems, and to the prospect of new engineered materials and processes. The rich physical basis of the models makes them computationally intensive, so a closely-coupled program of information technology research is planned to support applications research. The project brings together engineers, materials scientists, computer scientists and mathematicians. It links two existing NSF-sponsored research centeres at Illinois: the Center for Process Simulation and Design (CPSD), and the Materials Computation Center (MCC). This link is strategic, because it reflects the significant coupling between the meso- and macroscopic phenomena that CPSD studies and the largely atomistic behavior that MCC investigates. Three particular applications, listed in order of ascending scale, are (i) coupled quantum and continuum models of material interfaces; (ii) dendritic solidification with fluid flow in metallic microstructures; and (iii) microstructure evolution and process optimization in extrusion and quench processes. ***