High-order numerical investigations into landing gear wheel noise
High-order numerical investigations into landing gear wheel noise
Wheels are significant landing gear noise sources. In this project, high-order numerical simulations were conducted to investigate the landing gear wheel noise generation mechanisms and noise reduction treatments. The high-order solver solves the Navier-Stokes equations on multiblock structured grids. In this work, a modified solver was developed based on cell-centred formulation, which can provide more accurate solutions than cell-vertex space. This solver applies a finite-difference scheme at interior control points and at block interfaces with smooth grid metrics. At discontinuous block interfaces, a finite-volume method is employed as an interface condition. Two sets of interpolation schemes were developed to apply the finite volume method. This cell-centred high-order solver is accurate and robust for aeroacoustic simulations of complex geometries.
The numerical solver was applied to investigate the major noise sources of a 33% scaled isolated landing gear wheel by simulating three different wheel configurations using a hybrid CFD/FW-H approach. The configurations simulated include a baseline wheel with a hub cavity and two rim cavities. Two additional simulations were performed; one with the hub cavity covered and the other with both the hub cavity and rim cavities covered. The tyre is the main low frequency noise source and shows a lift dipole and side force dipole pattern depending on the frequency. The hub cavity is identified as the dominant middle frequency noise source and radiates in a frequency range centred around the first and second depth modes of the cylindrical hub cavity. The rim cavities are the main high-frequency noise sources. The largest noise reduction is achieved by covering both hub and rim cavities in the hub side direction.
Simulations of two wheels in tandem were also performed to study the wheel interaction noise at different angles of attack. The interaction noise is greatest at zero angle of attack, radiating towards the two sideline directions with a spectral peak at StW = 0.19, based on the width of the wheel. The dominant interaction noise source is the upstream shoulder of the downstream wheel. The wheel interaction noise is reduced at positive angles of attack, as less of the downstream wheel is immersed in the wake of the upstream wheel. A gap fairing was simulated, and it can significantly reduce the interaction noise by eliminating large-scale turbulent structures in the gap region. The downstream wheel hub and rim cavities do not have large contributions to the far-field acoustics.
University of Southampton
Wang, Meng
05c44cf0-f7d9-4ecc-9827-102ef72d7642
March 2017
Wang, Meng
05c44cf0-f7d9-4ecc-9827-102ef72d7642
Angland, David
b86880c6-31fa-452b-ada8-4bbd83cda47f
Wang, Meng
(2017)
High-order numerical investigations into landing gear wheel noise.
University of Southampton, Doctoral Thesis, 243pp.
Record type:
Thesis
(Doctoral)
Abstract
Wheels are significant landing gear noise sources. In this project, high-order numerical simulations were conducted to investigate the landing gear wheel noise generation mechanisms and noise reduction treatments. The high-order solver solves the Navier-Stokes equations on multiblock structured grids. In this work, a modified solver was developed based on cell-centred formulation, which can provide more accurate solutions than cell-vertex space. This solver applies a finite-difference scheme at interior control points and at block interfaces with smooth grid metrics. At discontinuous block interfaces, a finite-volume method is employed as an interface condition. Two sets of interpolation schemes were developed to apply the finite volume method. This cell-centred high-order solver is accurate and robust for aeroacoustic simulations of complex geometries.
The numerical solver was applied to investigate the major noise sources of a 33% scaled isolated landing gear wheel by simulating three different wheel configurations using a hybrid CFD/FW-H approach. The configurations simulated include a baseline wheel with a hub cavity and two rim cavities. Two additional simulations were performed; one with the hub cavity covered and the other with both the hub cavity and rim cavities covered. The tyre is the main low frequency noise source and shows a lift dipole and side force dipole pattern depending on the frequency. The hub cavity is identified as the dominant middle frequency noise source and radiates in a frequency range centred around the first and second depth modes of the cylindrical hub cavity. The rim cavities are the main high-frequency noise sources. The largest noise reduction is achieved by covering both hub and rim cavities in the hub side direction.
Simulations of two wheels in tandem were also performed to study the wheel interaction noise at different angles of attack. The interaction noise is greatest at zero angle of attack, radiating towards the two sideline directions with a spectral peak at StW = 0.19, based on the width of the wheel. The dominant interaction noise source is the upstream shoulder of the downstream wheel. The wheel interaction noise is reduced at positive angles of attack, as less of the downstream wheel is immersed in the wake of the upstream wheel. A gap fairing was simulated, and it can significantly reduce the interaction noise by eliminating large-scale turbulent structures in the gap region. The downstream wheel hub and rim cavities do not have large contributions to the far-field acoustics.
Text
Final e-thesis for e-Prints 26189461
- Version of Record
More information
Published date: March 2017
Identifiers
Local EPrints ID: 416431
URI: http://eprints.soton.ac.uk/id/eprint/416431
PURE UUID: 74cf0084-e4b7-4500-8c35-9060d534a9ea
Catalogue record
Date deposited: 15 Dec 2017 17:31
Last modified: 16 Mar 2024 05:58
Export record
Contributors
Author:
Meng Wang
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
Loading...
View more statistics