Interface learning of multiphysics and multiscale systems

Abstract

Complex natural or engineered systems comprise multiple characteristic scales, multiple spatiotemporal domains, and even multiple physical closure laws. To address such challenges, we introduce an interface learning paradigm and put forth a data-driven closure approach based on memory embedding to provide physically correct boundary conditions at the interface. To enable the interface learning for hyperbolic systems by considering the domain of influence and wave structures into account, we put forth the concept of upwind learning towards a physics-informed domain decomposition. The promise of the proposed approach is shown for a set of canonical illustrative problems. We highlight that high-performance computing environments can benefit from this methodology to reduce communication costs among processing units in emerging machine learning ready heterogeneous platforms toward exascale era.

References in corpus (9)

• Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
• A non-intrusive reduced order modeling framework for quasi-geostrophic turbulence
• Sub-grid scale model classification and blending through deep learning
• Unified approach to discretization of flow in fractured porous media
• An improved model for reduced-order physiological fluid flows
• Interface learning in fluid dynamics: statistical inference of closures within micro-macro coupling models
• Reduced Order Modeling Framework for Combustor Instabilities Using Truncated Domain Training
• 'Proxy-equation' paradigm - A novel strategy for massively-parallel asynchronous computations
• Boundary conditions for macroscale waves in an elastic system with microscale heterogeneity

Cited by in corpus (2)

• Multiscale modeling of inelastic materials with Thermodynamics-based Artificial Neural Networks (TANN)
• Physics guided machine learning using simplified theories