A Generalized Optimization Framework for High-Order, Metric-Based Mesh Adaptation of High-Order Finite-Element Discretization

Abstract

High-order computational fluid dynamics (CFD) simulations using the finite-element method (FEM) can achieve high accuracy at a low cost, but their efficiency deteriorates when the mesh is not optimal. The strict mesh requirements to realize the full potential of high-order CFD make meshing a significant bottleneck in the CFD workflow and hinder the widespread adoption of high-order CFD. accelerated convergence by as much as a full order. Metric-based mesh adaptation is appealing for optimization due to the continuous nature of the mesh framework. This project will create the theoretical underpinnings for high-order, metric-based mesh adaptation and use a higher-order Riemannian metric field to build an accurate continuous mesh framework for high-order meshes. This project will also develop a robust error model to quantify the effect of elemental splitting and high-order geometry node movements on the elemental error. This error model will guide the metric updates and hqr-adaptation. This project will also investigate efficient approaches to drive the mesh-implied metric toward the desired metric through hqr-adaptation.

Document Details

Document Type
DoD Grant Award
Publication Date
Feb 06, 2025
Source ID
FA95502410232

Entities

People

  • Devina Sanjaya

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Tennessee

Tags

Fields of Study

  • Engineering

Readers

  • Computational Fluid Dynamics (CFD)
  • Organizational Process Management (OPM).