Multiscale Modeling for Non-Equilibrium Material Behavior

Catalog description

This course will discuss a wide range of strategies aimed at integrating modeling and simulation techniques at various length and time scales with the goal of achieving a compromise between physical fidelity and computational cost. Our focus will lie on the bridge between atomistic simulations, described by Hamiltonian or Langevin dynamics, and continuum material descriptions, as well as the coarse-graining of mesoscopic continuum models. The course will review equilibrium statistical mechanics and thermodynamics, which are well-established theories, and discuss extensions to the non-equilibrium realm. We remark that non-equilibrium coarse-graining remains a long-standing challenge and a controversial quest in multiscale modeling, for which many approaches currently exist. Due to time limitations, this course can only give an overview of some state-of-the-art techniques. Although the course is aimed at being self-contained, it is assumed that students have a solid background in either elasticity/continuum mechanics or statistical physics, and are comfortable in Matlab, Python or some other programming language.