Modellbildung und Fluid-Struktur-Interaktion in der Biomechanik am Beispiel der menschlichen Phonation
Gömmel, Andreas; Meskouris, Konstantin (Thesis advisor)
Aachen : Klinkenberg (2010, 2011)
Dissertation / PhD Thesis
In: Mitteilungen des Lehrstuhls für Baustatik und Baudynamik, Fakultät für Bauingenieur- und Vermessungswesen, Rheinisch-Westfälische Technische Hochschule Aachen 18 = 10/1
Page(s)/Article-Nr.: VII, 142 S. : Ill., graph. Darst.
In human phonation, fluid-structure interaction between the air from the lungs (fluid) and the laryngeal tissues (structure) leads to a self-sustained oscillation of the vocal folds (consisting of muscle, mucosa, and ligament). The oscillation modulates the supraglottal pressure by interrupting the airflow which produces the sound of the voice. The presented model consists of a finite-element model of the structure and a finite-volume model of the fluid. Both of the models are coupled by the Arbitrary Lagrangian Eulerian method. The closure of the gap between the vocal folds is not directly possible to model and was for the first time considered by the combination of a contact problem and a distance dependent loss coefficient. Different versions of quasi two-dimensional and one three-dimensional model were set up. For the 2D models, two shapes were analyzed in detail. Only one of them showed self-sustained oscillations. This shape had a characteristic change of positive and negative pressure in the glottal gap. The other shape was deflected by the flow, showed some oscillations with decreasing amplitude and reached a static value of displacement. It was found that a precondition for self-sustained oscillations is a parallel channel between the vocal folds and an early torsional eigenform. Furthermore model shapes were derived from 3D magnet resonance imaging. A method for the use of more realistic material parameters was developed which calculates the Young's and shear moduli as well as the Poisson's ratios by an optimization algorithm using a large number of force-deflection-measurements of excised larynges as an input.