PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
March 14, 2026Computational Methods in Applied Mathematics0 citations

Vectorized 3D Mesh Refinement and Implementation of Primal Hybrid FEM in MATLAB

View Full Paper
HMHarish Nagula MalleshamSGSharat GaddamJVJan Valdman

Key Points

  • This research aims to develop a MATLAB software package for solving 3D elliptic problems using primal hybrid finite element method.
  • Implemented a fast 3D uniform finite element mesh refinement technique in MATLAB.
  • Established Nodes-to-Edge and Faces-to-Tetrahedron connectivity systematically.
  • Developed an efficient assembly procedure for primal hybrid finite element matrices.
  • Created a vectorized Schur complement solver to enhance computational performance.
  • Achieved improved run-time performance compared to MATLAB’s default solver.
  • Reduced original block system to a smaller Schur complement system, optimizing processing efficiency.
  • Demonstrated effectiveness through various numerical experiments.

Abstract

Abstract This article presents a MATLAB software package for solving a three-dimensional (3D) second-order elliptic problem with mixed boundary conditions using the primal hybrid finite element method (FEM). First, we introduce a novel fast 3D uniform finite element mesh refinement technique implemented in MATLAB and establish the Nodes-to-Edge and Faces-to-Tetrahedron connectivity through an efficient and systematic approach. We then describe an efficient MATLAB assembly procedure for the 3D lowest-order primal hybrid finite element matrices. Furthermore, we develop a vectorized Schur complement solver, where the computational improvement over MATLAB’s default direct solver ( mldivide ) arises from reducing the original block system to a substantially smaller Schur complement system. The run-time performance of the software is demonstrated through numerical experiments.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Mallesham et al. (2026) studied this question.

synapsesocial.com/papers/69b4fc7fb39f7826a300d601https://doi.org/10.1515/cmam-2025-0171
Ask AI
Helpful
Bookmark
Share
View Full Paper