This thesis is motivated by the need for noise control in aircraft engine with orifices and perforated liner. The presence of high-level acoustic excitation, different flow situations either bias flow, grazing flow or any combination in the aircraft engine, makes the acoustic behavior complex due to the interaction between sound and flow over the lined wall. Both systematic acoustic prediction of aircraft engines and liner optimization necessitate progress in impedance measurement methods by including the effect of the complex flow situations. The aim of the present thesis is to experimentally study the change in acoustic properties of orifices and perforated liners under bias or grazing flow. In order to study the effect of different combinations of bias flow and high-level acoustic excitation, an in-duct orifice has been investigated with finely controlled acoustic excitation levels and bias flow speeds. This provides a detailed study of the transition from cases when high-level acoustic excitation causes flow reversal in the orifice to cases when the bias flow maintains the flow direction. Nonlinear impedance is measured and compared, and a scattering matrix and its eigenvalues are investigated to study the potentiality of acoustic energy dissipation or production. A harmonic method is proposed for modelling the impedance, especially the resistance, which captures the change in impedance results at low frequencies compared with experimental results. The presence of grazing flow can increase the resistance of acoustic liners and shift their resonator frequency. So-called impedance eduction technology has been widely studied during the past decades, but with a limited confidence due to the interaction of grazing flow and acoustic waves. A comparison has been performed with different test rigs and methods from the German Aerospace Center (DLR). Numerical work has been performed to investigate the effect of shear flow and viscosity. Our study indicates that the impedance eduction process should be consistent with that of the code of wave propagation computation, for example with the same assumption regarding shear flow and viscosity. A systematic analysis for measurement uncertainties is proposed in order to understand the essentials for data quality assessment and model validation. The idea of using different Mach numbers for wave dispersion and in the Ingard-Myers boundary condition has been tested regarding their effect on impedance eduction. In conclusion, a local Mach number based on friction velocity is introduced and validated using both our own experimental results and those of previous studies. / <p>QC 20150522</p>

Sound generation and propagation in circular ducts for frequencies beyond the cut-on frequencies of several higher order acoustic modes is investigated. To achieve this, experimental and numerical set-ups are designed and used to research aeroacoustic interactions between in-duct components and to conceive noise mitigation strategies. Describing in-duct sound for frequencies with a moderate number of propagating modes is important, for example, for improving the noise emission from mid-size ventilation systems. Challenges that are largely unacknowledged in the literature involve efficient test rig design, quantification of limits in the methods, numerical modelling, and development of effective noise mitigation strategies for higher order modes. In this thesis, in-duct sound is mapped on a set of propagating pressure eigenmodes to describe aeroacoustic components as multi-ports with sound scattering (passive properties) and a source strength (active properties). The presented analysis includes genetic algorithms and Monte Carlo Methods for test rig enhancement and evaluation, multi-port network predictions to identify model limitations, and scale resolving (IDDES) and Linearized Navier Stokes computations for numerical multi-port eduction and the silencer design. It is first shown that test rig optimization improves the quality of multi-port data significantly. Subsequently, measurements on orifice plates are used to test the network prediction model. The model works with high accuracy for two components that are sufficiently separated. For small separations, strong coupling effects are observed for the source strength but not for the scattering of sound. The measurements are used for numerical validation, which gives reliable results for coupled and uncoupled systems. The total acoustic power of tandem orifices is predicted with less than 2 dB deviation and the passive properties for most frequencies with less than 5 % difference from the measurement. The numerical (FEM) models are also used to design a completely integrated silencer for spinning modes that is based on micro-perforated plates and gives broadband attenuation of 3-6 dB per duct diameter silencer length. The multi-port method is a powerful tool when describing aerodynamically decoupled in-duct components in the low- to mid-frequency range. Due to a robust passive network prediction, multi-port methods are particular interesting for the design of silencer stages. Furthermore, the demonstrated applicability to numerical data opens novel application areas. / <p>QC 20170522</p> / IdealVent