NSPOD: Accelerating Krylov solvers via DeepONet-learned POD subspaces

TL;DR

This paper introduces NSPOD, a deep operator network-based preconditioner that significantly accelerates Krylov solvers for parametric PDEs, outperforming existing methods in unstructured geometries.

Contribution

We propose NSPOD, a novel deep operator network-based preconditioner that dramatically reduces Krylov solver iterations for complex PDEs on unstructured domains.

Findings

NSPOD reduces iteration counts compared to algebraic multigrid preconditioners.

Numerical experiments show NSPOD's efficiency on solid mechanics PDEs.

NSPOD outperforms previous hybrid approaches in convergence speed.

Abstract

The convergence of Krylov-based linear iterative solvers applied to parametric partial differential equations (PDEs) is often highly sensitive to the domain, its discretization, the location/values of the applied Dirichlet/Neumann boundary conditions, body forces and material properties, among others. We have previously introduced hybridization of classical linear iterative solvers with neural operators for specific geometries, but they tend to not perform well on geometries not previously seen during training. We partially addressed this challenge by introducing the deep operator network Geo-DeepONet and hybridizing it with Krylov-based iterative linear solvers, which, despite learning effectively across arbitrary unstructured meshes without requiring retraining, led to only modest reductions in iterations compared to state-of-the-art preconditioners. In this study we introduce Neural Subspace Proper Orthogonal Decomposition (NSPOD), a multigrid-like deep operator network-based preconditioner which can dramatically reduce the number of iterations needed for convergence in Krylov-based linear iterative solvers, even when compared to state-of-the-art methods such as algebraic multigrid preconditioners. We demonstrate its efficiency via numerical experiments on a linearized version of solid mechanics PDEs applied to unstructured domains obtained from complex CAD geometries. We expect that the findings in this study lead to more efficient hybrid preconditioners that can match, or possibly even surpass, the convergence properties of the current gold standard preconditioning methods for solid mechanics PDEs.