Robust Incomplete Factorization Preconditioning Based on A-orthogonalization Process for Parallel Finite Element Analysis

Abstract

Robust incomplete factorization (RIF) based on A-otrhogonalization process is an effective preconditioning technique for the conjugate gradient (CG) method to solve highly ill-conditioned linear systems. This research aims to apply the same sparsity pattern as that of coefficient matrix and parallelization of localized strategy to the process of RIF preconditioning for acceleration of convergence in finite element analysis of shell structures. As numerical results, the present RIF preconditioner leads to better convergence than the incomplete Cholesky preconditioner.