EXPERIMENTAL STATISTICAL ENERGY ANALYSIS OF SATELLITE STRUCTURE USING ENERGY SENSITIVITY ANALYSIS METHOD

Hongliang Zhang,Xianren Kong,Jiang Qin
2018-01-01
Abstract:Statistical Energy Analysis (SEA) is an efficient method to predict wide-band-width noise and vibration for complex spacecraft structure. However, an analytical SEA model often exhibits errors due to unreasonable estimation of SEA parameters, such as damping loss factors (DLF) and coupling loss factors (CLF). Experimental SEA is practically used to update the model or to identify the DLFs and CLFs in a SEA prediction work. This paper proposes a hybrid SEA modeling method based on energy sensitivity analysis for experimental SEA, which can update analytical SEA model with errors using test measured data. The proposed method can minimize residuals between measured and analytically predicted energy levels by solving a nonlinear constrained optimization problem iteratively, and the SEA parameters of the analytical model are updated directly in more physical sense. A reduction technique is utilized to match the analytical subsystems with the experimental ones. In addition, the problem of ill conditioned sensitivity matrix is investigated based on regularization method to improve the numerical stability of the algorithm. The proposed methods are validated through several numerical simulations, which show that the analytical SEA model with error parameters can be updated successfully even with a very limited experimental data set. And then an acoustic test investigation of a satellite structure is carried out and a procedure to perform an analytically predictive SEA model updating with the measured data is summarized. It is found that the predictions of updated model of the satellite match the measured result more closely, which indicates that the hybrid SEA modeling method can enhance the ability to model the vibroacoustic behavior of complex systems.
What problem does this paper attempt to address?