Efficient fluid structure interaction simulation of vocal fold oscillations using a homogenized Lattice Boltzmann Method

Read full text
Authors: Adrian Kummerländer, Bogac Tur, Maik Haase, Fedor Bukreev, Michael Döllinger, Mathias J. Krause, Stefan Kniesburges
Cited by: 0
References: 42

Abstract

Computational modeling of human phonation is a uniquely challenging fluid-structure interaction (FSI) problem. It is constrained both by the large computational cost and numerical restrictions of existing methods, particularly in handling vocal fold (VF) contact. This is why new methods are required that combine physiological fidelity with high computational efficiency.This study presents a novel, efficient FSI solver that addresses this challenge by coupling a homogenized Lattice Boltzmann Method (HLBM) for the fluid dynamics with a Six-Mass-Model (6MM) for the VF structure. The fluid is modeled using the filtered Brinkman-Navier-Stokes equations, platform-transparently discretized by an HLBM variant in the OpenLB framework. The moving VFs are represented as a time-dependent porosity field, implicitly handling complex geometry changes and contact without remeshing. Their dynamics are represented by two-way coupling the 6MM, capturing both elastic motion and collision. Simulations in a 3D laryngeal channel produced stable, self-sustained oscillations at physiologically plausible values: fundamental frequency (f_0) ~248 Hz, mean subglottal pressure ~1.01 kPa, and an open quotient ~0.64. The model robustly achieved complete glottal closure in each cycle, yielding credible peak contact forces (~27 mN).The proposed HLBM-6MM solver provides both high computational efficiency (up to 60 oscillation cycles/hour on a NVIDIA RTX 2000 Ada laptop GPU) and physiological results, enabling systematic parameter studies of phonation previously infeasible with traditional FSI approaches.

Fields of Study & Subjects

Lattice Boltzmann Simulation Studies Voice and Speech Disorders Fluid Dynamics and Vibration Analysis

Publication details

Journal / source
Computer Methods in Applied Mechanics and Engineering
Volume
457
Pages
119009–119009
Publication date
25 April 2026
Language
EN