The estimation of modal parameters through Goal Programming, an optimisation technique, is presented. The estimation, performed in the frequency domain, seeks to minimise the total squared error between experimental and estimated values of the frequency response functions. Its purpose is to obtain g
COUPLING OF NON-LINEAR SUBSTRUCTURES USING VARIABLE MODAL PARAMETERS
โ Scribed by Y.H. CHONG; M. IMREGUN
- Publisher
- Elsevier Science
- Year
- 2000
- Tongue
- English
- Weight
- 227 KB
- Volume
- 14
- Category
- Article
- ISSN
- 0888-3270
No coin nor oath required. For personal study only.
โฆ Synopsis
This paper presents a general methodology for the coupling analysis of systems with relatively weak non-linearities by assuming that the response remains harmonic under harmonic excitation. Standard coupling methods and their current shortcomings were discussed "rst. Two ways of obtaining non-linear modal parameter variations, namely pro"le constructing and parameter extracting, were presented next. The pro"le constructing method uses the system's spatial data directly, while the parameter extracting method is based on a non-linear modal analysis of measured response data. Through numerical test cases, it was shown that both methods yielded virtually identical results. An iterative algorithm for the coupling of non-linear subsystems was presented in a form compatible with pro"le building. A six-degree-of-freedom system with cubic sti!ness non-linearity was chosen for a detailed numerical study. Two subsystems, one linear and the other non-linear, were coupled to obtain the modal parameter variations of the coupled system. Using the non-linear modal parameters, the response of the coupled system was predicted at various force levels and the "ndings were checked via direct simulations using the harmonic balance method. Finally, the methodology was validated by coupling experimentally derived nonlinear modal models of two substructures. As for the numerical study, the response of the coupled non-linear structure was predicted at various force levels and the "ndings were checked against direct measurements. Very good agreement was obtained in all cases studied.
๐ SIMILAR VOLUMES
Conventional sensors used for structural measurement are usually discrete-type add-on devices. Lightweight distributed neurons fully integrated (laminated or embedded) with structural components can serve as in situ sensors monitoring structure's dynamic state and health status. Thin-"lm lightweight