K Wienen
Fast transfer path analysis using controlled electric motors as in-situ blocked force vibration sources
Wienen, K
Authors
Contributors
AT Moorhouse A.T.Moorhouse@salford.ac.uk
Supervisor
Dr Joshua Meggitt J.W.R.Meggitt1@salford.ac.uk
Supervisor
AS Elliott A.S.Elliott@salford.ac.uk
Supervisor
M Sturm
Supervisor
Abstract
Under certain driving and road conditions, steering manoeuvres may provoke audible noise inside the vehicle compartment induced by the electric power steering (EPS) system. Regarding noise, vibration and harshness (NVH), engineers require practical tools to assess, design and troubleshoot steering noise in vehicles. For acoustic development and refinement, test-based methodologies such as Transfer Path Analysis (TPA) are used to analyse noise and vibration propagation in complex systems. In response to the evolving demands in acoustic engineering, TPA approaches have been continuously improved to provide reliable diagnostic information, simplify the procedure or accelerate development time. Although an established tool in many industries, TPA is a rather complex and time-intensive procedure. Furthermore, state-of-the-art TPA approaches also tend to suffer from a variety of practical limitations such as impracticality to include particular paths for structure-borne sound transmission (e.g. in-plane), inability to measure rotational dynamics (e.g. moment excitation), insufficient signal-to-noise ratios (SNR), or simply restricted access. These inherent challenges have led to near-routine neglect of transmission paths, potentially providing an engineer with unrealistic diagnoses to make informed design changes. This thesis addresses most of the above challenges by proposing a variety of novel experimental techniques to augment state-of-the-art TPA methods with the aim to increase reliability of in-vehicle and bench-based NVH system development while significantly reducing measurement time and effort.
A key step, also the most time-consuming part, of any TPA is the measurement of frequency response functions (FRFs). These measurements may be difficult or impossible if the measurement locations are inaccessible or difficult to excite (e.g. in-plane directions). Therefore, a framework for indirect measurement is proposed, by invoking the round-trip identity, providing experimentalists with the ability to relocate measurements to more convenient positions on the test structure. A generalisation of this identity is proposed to reconstruct FRFs between an interface and some selected points using only remote measurement positions. Within this generalised concept, direct measurement of rotations or inaccessible points is avoided altogether to reduce complexity in the measurements commonly involved in TPA. Manipulation of the identity yields a formulation for long distance transfer FRFs, expressed by multiple shorter paths with a better SNR, to facilitate TPA in heavyweight or extensively large structures.
Classical TPA measurements require each transmission path to be characterised by the dynamic output (e.g. vibration or sound pressure) in response to a known input load (e.g. force or moment excitation). More recently, measurements using output quantities only have been developed, but the reduced measurement time they allow comes at the expense of clarity and accuracy. This thesis proposes an alternative approach in which in-situ system identification of all physically existing transfer paths is performed simultaneously by converting any vibration source into a multiple degree of freedom (DoF) blocked force exciter. The concept is to exploit the invariance of the source's blocked force, so that the same blocked force can be assumed to act irrespective of the receiver to which the source is attached. This source excitation has to be characterised on a test bench in the so-called calibration stage prior to installing the same (calibrated) source in the target assembly. In the subsequent system identification stage, exact structural and vibro-acoustic transmission paths can then be characterised by simple operational response measurements in the installation due to the known blocked force excitation.
This two-stage system identification method with a controlled source provides a convenient alternative to conventional FRF measurements with an instrumented hammer or shaker. Especially in the context of TPA, in which such FRFs are required for inverse force identification and forward response prediction, significant time savings can be achieved. As a diagnostic tool, this process is denoted as `fastTPA'. It is shown that concepts adopted from control theory, more specifically controllability and observability, are strongly related to fastTPA and provide practical guidelines.
The thesis concludes with an experimental case study utilising the introduced methods to analyse steering induced sound and vibration in a fully assembled vehicle. It is demonstrated that a steering system, calibrated as a controlled blocked force exciter (multi-DoF shaker), can be used to obtain high-accuracy structural and vibro-acoustic FRFs in a multi-kHz range. For a time-efficient yet precise estimation of path contributions, these FRFs are used for fastTPA and to construct a realistic Virtual Acoustic Prototype capable of predicting the operational pressure response in the vehicle compartment. When benchmarked against the latest advancements in component-based TPA approaches, fastTPA is found to be significantly faster with at least the same accuracy and reliability.
Citation
Wienen, K. (2021). Fast transfer path analysis using controlled electric motors as in-situ blocked force vibration sources. (Thesis). University of Salford
Thesis Type | Thesis |
---|---|
Deposit Date | Dec 14, 2021 |
Publicly Available Date | Aug 1, 2024 |
Award Date | Jul 31, 2021 |
Files
FastTPA_PhD_K.Wienen.pdf
(76.1 Mb)
PDF
You might also like
Power flow sensitivity analysis for optimal structural modification
(2023)
Journal Article
Interval-based identification of response-critical joints: A tool for model refinement
(2022)
Journal Article
An experimental exploration of the properties of random frequency response functions
(2020)
Journal Article
Downloadable Citations
About USIR
Administrator e-mail: library-research@salford.ac.uk
This application uses the following open-source libraries:
SheetJS Community Edition
Apache License Version 2.0 (http://www.apache.org/licenses/)
PDF.js
Apache License Version 2.0 (http://www.apache.org/licenses/)
Font Awesome
SIL OFL 1.1 (http://scripts.sil.org/OFL)
MIT License (http://opensource.org/licenses/mit-license.html)
CC BY 3.0 ( http://creativecommons.org/licenses/by/3.0/)
Powered by Worktribe © 2024
Advanced Search