Modal parameter identification based on robust blind source separation using temporal predictability

YANG Yan-long;CHENG Wei

Journal of Vibration and Shock ›› 2012, Vol. 31 ›› Issue (10) : 9-12,2.

PDF(1172 KB)
PDF(1172 KB)
Journal of Vibration and Shock ›› 2012, Vol. 31 ›› Issue (10) : 9-12,2.
论文

Modal parameter identification based on robust blind source separation using temporal predictability

  • YANG Yan-long1, CHENG Wei1
Author information +
History +

Abstract

Abstract: In this paper, a robust blind source separation using temporal predictability is adapted for use in modal parameter identification. In order to improve noise immunity of the algorithm, a robust whitening process is performed to reduce the noise in the measured vibration signals. In this method, the (inverse) modal shapes and the modal responses are regarded as a mixed matrix and source signals, respectively. The objective function is constructed under the assumption that the temporal predictability of any physical response is less than (or equal to) that of any of its modal responses. Then, all the modal shapes are recovered directly from free responses by optimizing the objective function. Finally, frequency and damping ratios can be obtained from the modal responses by simple single degree of freedom methods. Numerical evaluations demonstrate that the mode shapes could be extracted exactly even under low signal to noise rate (SNR) case or closely spaced modes.

Key words

Robust TPBSS / Modal parameter identification / Blind source separation / Robustness / Closely spaced modes

Cite this article

Download Citations
YANG Yan-long;CHENG Wei . Modal parameter identification based on robust blind source separation using temporal predictability[J]. Journal of Vibration and Shock, 2012, 31(10): 9-12,2
PDF(1172 KB)

Accesses

Citation

Detail

Sections
Recommended

/