| dc.description |
Noise and vibration are important design issues for many types of
vehicles such as ships, cars, and aeroplanes. Structure borne sound,
which may be of relatively high frequency, usually emanates from an
engine or some other type of localised source and propagates through the
vehicle. Excessive vibration levels, and thus structural damage, may
occur while structural acoustic interactions may lead to unacceptable
interior noise.
In the analysis of energy transmission between plate structures, it is
common practice to consider only bending modes (or waves) of the
structure. However if the concern is with high frequency vibration
analysis, then due allowance may need to be made for the presence of inplane
shear and longitudinal modes.
Due to the infeasibility of the industry standard technique, the Finite
Element Method, at high frequencies, almost all of the studies that have
investigated the importance of in-plane energy transmission have used
Statistical Energy Analysis (SEA).
In this study an existing dynamic stiffness method is extended to include
in-plane effects, and used as a benchmark against which SEA is assessed.
Additionally the Wave Intensity Analysis (WIA) technique, which is an
improved form of SEA, is extended to in-plane vibrations, and used to
identify some of the reasons for the poor performance of SEA in certain
applications. All three methods are applied to a wide range of plate
structures within the frequency range of 600 Hz to 20 kHz. While the
response levels as predicted by the WIA are generally quite close to
exact results, it has been found that although all of the requirements
which are usually postulated for the successful application of SEA are
fulfilled, SEA severely underpredicts the energy transmission in large
structures because of the diffuse wave field assumption. It is also shown
that the exclusion of in-plane modes may lead to sizeable errors in
energy predictions unless the structure is very simple. |
|