Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4977056 | Mechanical Systems and Signal Processing | 2017 | 13 Pages |
Abstract
The effect of parametric uncertainties on the dispersion of Statistical Energy Analysis (SEA) models of structural-acoustic coupled systems is studied with the Fourier analysis sensitivity test (FAST) method. The method is firstly applied to an academic example representing a transmission suite, then to a more complex industrial structure from the space industry. Two sets of parameters are considered, namely error on the SEA model's coefficients, or directly the engineering parameters. The first case is an intrusive approach, but enables to identify the dominant phenomena taking place in a given configuration. The second is non-intrusive and appeals more to engineering considerations, by studying the effect of input parameters such as geometry or material characteristics on the SEA outputs. A study of the distribution of results in each frequency band with the same sampling shows some interesting features, such as bimodal repartitions in some ranges.
Related Topics
Physical Sciences and Engineering
Computer Science
Signal Processing
Authors
Jean-Loup Christen, Mohamed Ichchou, Bernard Troclet, Olivier Bareille, Morvan Ouisse,