Abstract:We propose in this paper a Proper Generalized Decomposition (PGD) solver for reduced-order modeling of linear elastodynamic problems. It primarily focuses on enhancing the computational efficiency of a previously introduced PGD solver based on the Hamiltonian formalism. The novelty of this work lies in the implementation of a solver that is halfway between Modal Decomposition and the conventional PGD framework, so as to accelerate the fixed-point iteration algorithm. Additional procedures such that Aitken's delta-squared process and mode-orthogonalization are incorporated to ensure convergence and stability of the algorithm. Numerical results regarding the ROM accuracy, time complexity, and scalability are provided to demonstrate the performance of the new solver when applied to dynamic simulation of a three-dimensional structure.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to improve the computational efficiency of the Reduced - Order Model (ROM) solver based on the Proper Generalized Decomposition (PGD) method in structural dynamics applications. Specifically, the author proposes a new PGD solver, aiming to improve the previous PGD solver based on the Hamiltonian form and accelerate the convergence speed of the fixed - point iteration algorithm. ### Main Problems and Challenges 1. **Computational Efficiency**: When dealing with transient structural dynamics problems, traditional PGD solvers have low computational efficiency due to the poor convergence of the fixed - point algorithm. The goal of this paper is to develop a more efficient solver to reduce the computational time and resource consumption. 2. **Convergence and Stability**: To ensure the stability and convergence of the algorithm, the author introduces additional procedures, such as Aitken's delta - squared process and modal orthogonalization. These methods help accelerate the convergence of the fixed - point iteration algorithm and ensure the stability of the algorithm. 3. **Applicability of Space - Time Separation**: The effectiveness of using the space - time separation method to solve wave equations or second - order hyperbolic problems has always been controversial. This paper explores how to overcome these problems through an improved PGD solver framework. ### Solutions 1. **Combining Modal Decomposition and the Traditional PGD Framework**: The newly proposed solver is between modal decomposition and the traditional PGD framework, taking advantage of both to accelerate the fixed - point iteration algorithm. 2. **Aitken Acceleration**: Aitken's delta - squared process is introduced to accelerate the convergence of the fixed - point iteration algorithm. This significantly reduces the number of iterations required to reach convergence. 3. **Modal Orthogonalization**: A new orthogonal projection of spatial modes is carried out to ensure their linear independence and enhance the stability of the algorithm. 4. **Ritz Vector Pre - processing**: To avoid repeated matrix factorizations, the author develops an original method, that is, pre - processing the approximate eigenpairs (Ritz pairs) of the operator, so that the spatial part of the problem remains diagonalized throughout the fixed - point iteration process. This greatly reduces the computational burden. ### Numerical Results The paper provides numerical results on the accuracy, time complexity, and scalability of the ROM to demonstrate the performance of the new solver in three - dimensional structural dynamic simulations. In conclusion, this paper provides a more efficient method for reduced - order modeling in the field of structural dynamics by improving the PGD solver, especially in terms of computational efficiency and algorithm stability.

An Efficient PGD Solver for Structural Dynamics Applications

Multi-temporal decomposition for elastoplastic ratcheting solids

Nonintrusive proper generalised decomposition for parametrised incompressible flow problems in OpenFOAM

An ultraweak DPG method for viscoelastic fluids

Modular parametric PGD enabling online solution of partial differential equations

An energy-efficient GMRES–multigrid solver for space-time finite element computation of dynamic poroelasticity

A hybrid numerical methodology coupling Reduced Order Modeling and Graph Neural Networks for non-parametric geometries: applications to structural dynamics problems

The LATIN-PGD methodology to nonlinear dynamics and quasi-brittle materials for future earthquake engineering applications

A Multi‐layer Solver for XPBD

CFD Stability Improvement Using Dynamic Mode Decomposition of Solution Update Vectors

An Efficient Backward Euler Time-Integration Method for Nonlinear Dynamic Analysis of Structures

Accelerating structural dynamics simulations with localised phenomena through matrix compression and projection‐based model order reduction

Recycling Krylov Subspaces for Efficient Partitioned Solution of Aerostructural Adjoint Systems

Two-level dynamic load-balanced p-adaptive discontinuous Galerkin methods for compressible CFD simulations

A cell-centered Lagrangian ADER-MOOD finite volume scheme on unstructured meshes for a class of hyper-elasticity models

Concurrent multi-domain simulations in linear structural dynamics using multiple grid multiple time-scale (MGMT) method

High-order elements in position-based dynamics

Solving large numerical substructures in real‐time hybrid simulations using proper orthogonal decomposition

A nonintrusive distributed reduced‐order modeling framework for nonlinear structural mechanics—Application to elastoviscoplastic computations