18SC006 - Mechanistic Data-driven Multiscale Analysis and Applications

Instructors:

Prof. Wing Kam Liu, w-liu@northwestern.edu, Northwestern University, Director of Global Center on Advanced Material Systems and Simulation; and President of International Association for Computational Mechanics

Dr. C.T. Wu, ctwu@lstc.com, Livermore Software Technology Company (LSTC), USA

Zeliang Liu, zlliu@lstc.com, Livermore Software Technology Company (LSTC), USA

Course description

An open frontier in mechanics is the efficient and accurate description of highly nonlinear or heterogeneous material behavior that strongly depends on complex microstructure. This one-day short course will introduce participants to the latest efforts to expand this frontier: mechanistic data-driven methods for material homogenization and concurrent multiscale analysis. Higher order homogenization theories complete concurrent multiscale methods with a theoretical framework enabling the modeling of failure processes across multiple length scales. Concurrent multiscale methods were developed to avoid the limitations of phenomenological models by directly relating material macro-response to the microstructure. Benchmarks on metal alloys, polymer matrix composites will be utilized to demonstrate the effectiveness of those numerical methods through the commercial code LS-DYNA. The class will also provide the detailed description of LS-DYNA keywords and their use in the multiscale analysis.

Preliminary Course schedule