Manifold learning for coarse-graining atomistic simulations: Application to amorphous solids

Abstract

We introduce a generalized machine learning framework to probabilistically parameterize upper-scale models in the form of nonlinear PDEs consistent with a continuum theory, based on coarse-grained atomistic simulation data of mechanical deformation and flow processes. The proposed framework utilizes a hypothesized coarse-graining methodology with manifold learning and surrogate-based optimization techniques. Coarse-grained high-dimensional data describing quantities of interest of the multiscale models are projected onto a nonlinear manifold whose geometric and topological structure is exploited for measuring behavioral discrepancies in the form of manifold distances. A surrogate model is constructed using Gaussian process regression to identify a mapping between stochastic parameters and distances. Derivative-free optimization is employed to adaptively identify a unique set of parameters of the upper-scale model capable of rapidly reproducing the system's behavior while maintaining consistency with coarse-grained atomic-level simulations. The proposed method is applied to learn the parameters of the shear transformation zone (STZ) theory of plasticity that describes plastic deformation in amorphous solids as well as coarse-graining parameters needed to translate between atomistic and continuum representations. We show that the methodology is able to successfully link coarse-grained microscale simulations to macroscale observables and achieve a high-level of parity between the models across scales.

arXiv comment34 pages, 12 figures, references added, Section 4 added, Section 2.1 updated

References in corpus (23)

• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• Dynamics of Viscoplastic Deformation in Amorphous Solids
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Vibrational modes identify soft spots in a sheared disordered packing
• Deformation and Failure of Amorphous Solidlike Materials
• Deep Potential: a general representation of a many-body potential energy surface
• Predicting plasticity in disordered solids from structural indicators
• DeePCG: constructing coarse-grained models via deep neural networks
• Evaluation of the "disorder temperature" and "free volume" formalisms via simulations of shear banding in amorphous solids
• Dynamics of Shear-Transformation Zones in Amorphous Plasticity: Formulation in Terms of an Effective Disorder Temperature
• Nonequilibrium Thermodynamics of Amorphous Materials III: Shear-Transformation-Zone Plasticity
• Coarse-scale PDEs from fine-scale observations via machine learning
• Adversarial-Residual-Coarse-Graining: Applying machine learning theory to systematic molecular coarse-graining
• Unsupervised Machine Learning Based on Non-Negative Tensor Factorization for Analyzing Reactive-Mixing
• Predictive Coarse-Graining
• Grassmannian Learning: Embedding Geometry Awareness in Shallow and Deep Learning
• Notch fracture toughness of glasses: Rate, age and geometry dependence
• A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables
• Shear Band Broadening in Simulated Glasses
• An Eulerian projection method for quasi-static elastoplasticity
• On orthogonal projections for dimension reduction and applications in augmented target loss functions for learning problems
• Parallel three-dimensional simulations of quasi-static elastoplastic solids
• Coordinate transformation methodology for simulating quasi-static elastoplastic solids