Model Order Reduction of Cerebrovascular Hemodynamics Using POD_Galerkin and Reservoir Computing_based Approach

TL;DR

This paper compares physics-based and data-driven model order reduction techniques for cerebrovascular hemodynamics, demonstrating significant computational speed-ups and accurate flow predictions using POD-Galerkin and Reservoir Computing methods.

Contribution

It introduces a combined POD-Galerkin and reservoir computing framework for efficient cerebrovascular flow simulation, highlighting the advantages of data-driven approaches.

Findings

Both models achieve 100-1000x speed-up over full CFD simulations.

The POD-RC model effectively learns temporal dynamics of flow.

Training with multi-harmonic signals improves model accuracy.

Abstract

We investigate model order reduction (MOR) strategies for simulating unsteady hemodynamics within cerebrovascular systems, contrasting a physics-based intrusive approach with a data-driven non-intrusive framework. High-fidelity 3D Computational Fluid Dynamics (CFD) snapshots of an idealised basilar artery bifurcation are first compressed into a low-dimensional latent space using Proper Orthogonal Decomposition (POD). We evaluate the performance of a POD-Galerkin (POD-G) model, which projects the Navier-Stokes equations onto the reduced basis, against a POD-Reservoir Computing (POD-RC) model that learns the temporal evolution of coefficients through a recurrent architecture. A multi-harmonic and multi-amplitude training signal is introduced to improve training efficiency. Both methodologies achieve computational speed-ups on the order of 10^2 to 10^3 compared to full-order simulations, demonstrating their potential as efficient and accurate surrogates for predicting flow quantities such as wall shear stress.