Author ORCID Identifier

Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Mechanical Engineering

First Advisor

Dr. David W. Herrin


Panel contribution analysis (PCA) can be used to predict machinery noise emissions, component contributions, and to assess the impact of sound reduction treatments. PCA is a measurement approach that is advantageous for complex machinery that is not easily modeled using conventional numerical analysis approaches. In this research, PCA is combined with scale modeling in order to speed up the necessary measurement work. Moreover, the method can be applied to much larger machinery and noise emissions can be assessed prior to locating and installing the equipment. This eliminates the necessity to use voluminous anechoic chambers.

The machinery is first discretized into a collection of panels or patches. Volume velocities are measured for each patch with the machinery operating, and transfer functions are measured between panels and receiver locations with the machinery turned off. It is shown that transfer functions may be measured using a scale model. Then, the sound pressure level produced by the machinery is predicted. The method is first applied to a generator set and a 1/2 scale model is used to measure the acoustic transfer functions. It is demonstrated that PCA can be used to predict sound pressure levels in the far-field of a source even using a relatively small hemi-anechoic chamber. PCA was then used to assess the efficacy of barrier treatments.

The PCA and scale modeling combination were then applied to an interior acoustics scenario. The acoustic emissions from three similar air handlers positioned throughout a bakery were predicted at two locations. Transfer functions were measured between the panels and three different customer locations using a 1/10th scale model. Transfer functions were corrected to account for air attenuation and predicted sound pressure levels compare well with measurement. The described approach may be used to determine the sound pressure levels in large interior spaces before they are constructed so long as volume velocities on the source can be measured a priori. In addition, strategies, such as barriers and sound absorption, to reduce the noise by modifications to the acoustic path were accurately assessed prior to equipment installation.

PCA was then applied to a small unmanned aerial vehicle (UAV) and the sound pressure level was predicted 5.5 m away. In this case, both the panel volume velocities and sound pressures must be measured because the boundary encompassing the source is no longer semi-rigid. Measurements were performed on six measurement surfaces forming an imaginary box encompassing the UAV. A P-U Probe was utilized to measure both sound pressure and particle velocity on the imaginary surfaces. Acoustic transfer functions between the source and a receiver point were measured reciprocally. The noise level was predicted from measurements close to the UAV assuming both correlated and uncorrelated sources at the receiver point. The sound pressure level calculated by the correlated model compared well with direct measurement.

Digital Object Identifier (DOI)

Funding Information

This study was supported by Mechanical Department's Research Assistantship from 2012 to 2019.