Combining Machine Learning and Adaptive Coarse Spaces---A Hybrid Approach for Robust FETI-DP Methods in Three Dimensions

DOI10.1137/20M1344913zbMath1476.65326OpenAlexW3037002849MaRDI QIDQ3382792

Alexander Heinlein, Axel Klawonn, Martin Lanser, Janine Weber

Publication date: 22 September 2021

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1137/20m1344913

zbMATH Keywords

finite elementsdomain decomposition methodsmachine learningFETI-DPadaptive coarse spacesML-FETI-DP

Mathematics Subject Classification ID

Multigrid methods; domain decomposition for boundary value problems involving PDEs (65N55) Learning and adaptive systems in artificial intelligence (68T05) Finite element, Rayleigh-Ritz and Galerkin methods for boundary value problems involving PDEs (65N30) Iterative numerical methods for linear systems (65F10)

Related Items

FROSch Preconditioners for Land Ice Simulations of Greenland and Antarctica ⋮ A surrogate model for the prediction of permeabilities and flow through porous media: a machine learning approach based on stochastic Brownian motion ⋮ Accelerating algebraic multigrid methods via artificial neural networks ⋮ On the use of graph centralities to compute generalized inverse of singular finite element operators: Applications to the analysis of floating substructures ⋮ Learning adaptive coarse basis functions of FETI-DP

Uses Software

• METIS
• Adam

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• A comparison of adaptive coarse spaces for iterative substructuring in two dimensions
• Triangular and quadrilateral surface mesh quality optimization using local parametrization
• Adaptive selection of face coarse degrees of freedom in the BDDC and the FETI-DP iterative substructuring methods
• Robust FETI-DP methods for heterogeneous three dimensional elasticity problems
• Adaptive BDDC in three dimensions
• Local spectra of adaptive domain decomposition methods
• Machine learning in adaptive FETI-DP -- a comparison of smart and random training data
• A frugal FETI-DP and BDDC coarse space for heterogeneous problems
• FETI-DP: a dual-primal unified FETI method?part I: A faster alternative to the two-level FETI method
• Adaptive Coarse Spaces for FETI-DP in Three Dimensions
• Deflation, Projector Preconditioning, and Balancing in Iterative Substructuring Methods: Connections and New Results
• Dual-primal FETI methods for linear elasticity
• A Preconditioner for Substructuring Based on Constrained Energy Minimization
• Machine Learning in Adaptive FETI-DP: Reducing the Effort in Sampling
• Machine Learning in Adaptive Domain Decomposition Methods---Predicting the Geometric Location of Constraints
• Highly scalable parallel domain decomposition methods with an application to biomechanics