Date of Award

2026

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy in Mechanical Engineering-Engineering Mechanics (PhD)

Administrative Home Department

Department of Mechanical and Aerospace Engineering

Advisor 1

Bhisham Sharma

Committee Member 1

Sriram Malladi

Committee Member 2

Kazuya Tajiri

Committee Member 3

Quang Tran

Abstract

Acoustic liners are the principal treatment for turbofan noise, but conventional perforate-over-honeycomb designs limit bandwidth and geometric freedom. This dissertation develops methods to characterize, model, and design additively manufactured porous liners and tests whether the pointwise scalar impedance used in liner engineering adequately represents volumetrically open structures. Gyroid and diamond triply periodic minimal surface (TPMS) geometries provide the test class, with prescribed relative density, unit-cell size, and layering.

Printed specimens are characterized at normal incidence in their as-fabricated state. Porosity is estimated from density, airflow resistivity is measured, and the remaining Johnson–Champoux–Allard parameters are identified by inverse analysis as effective model coordinates. Transfer-matrix predictions reproduce measured absorption of layered configurations withheld from calibration. Homogenized ideal gyroids yield porosity master curves over 0.10–0.90 with exact cell-size scaling. Comparison at four common points captures the combined effects of fabrication, identification, and model form.

At grazing incidence, open and partitioned liners are modeled as volumetric equivalent fluids. Impedance educed from duct measurements under a pointwise, locally reacting wall assumption is compared with simulated interface impedance. For the open liner, the impedances disagree across much of the band, including regions where the model reproduces the measured field. Partitioning improves agreement, with the closest agreement at the smallest tested spacing. An imposed-impedance positive control verifies the boundary implementation. The results establish an inconsistency among the volumetric model, the educed impedance, and the pointwise scalar representation for the open liner, but do not identify one as its source. Measured and modeled level differences show that partitioning concentrates attenuation near the treatment-depth resonance under the tested no-flow conditions.

The master curves and transfer matrix provide a fast forward model for layered-porosity searches. At fixed treatment depth, searches at unit-cell sizes of 5, 3, and 1 mm produce nonmonotonic profiles whose calculated absorption exceeds the best-uniform baselines over most of the 500–6000 Hz band. The calculations address ideal geometry at normal incidence without manufacturability constraints.

The central scientific contribution is that internal architecture governs both liner performance and whether pointwise scalar impedance adequately represents the liner.

Download

Available for download on Friday, July 30, 2027

Share

COinS