Browse by author
Lookup NU author(s): Dr Wenxian YangORCiD
This is the authors' accepted manuscript of an article that has been published in its final definitive form by ASME, 2016.
For re-use rights please refer to the publisher's terms and conditions.
The dynamic response of a nonlinear system can be characterized by its instantaneous amplitude (IA) and instantaneous frequency (IF) features, which dependent on the physical properties of the system. Accordingly, the system properties can be inferred from the IA and IF features if they can be identified accurately. To fulfil such an idea, a nonlinear system parameter identification method is proposed in this paper with the aid of Polynomial Chirplet Transform (PCT), which has been proved a powerful tool for processing non-stationary signals. The procedure of the proposed system identification method is summarized as: firstly, the instantaneous characteristics, IA and IF, are extracted by using the PCT from the nonlinear responses of the system; secondly, calculate the instantaneous modal parameters and present them to backbone and damping curves, which characterize the inherent nonlinearities of the system; and finally, estimate the physical property parameters of the system through fitting the identified average nonlinear characteristic curves. The proposed nonlinear system identification method is experimentally validated in the paper. The experimental results have shown that the proposed method is superior to two existing Hilbert transform based methods, particularly on the robust performance against noise. In other words, the proposed method can work very well in system identification under any noise contamination condition, while the HT based methods can work only in the absence of noise.
Author(s): Deng Y, Cheng CM, Yang Y, Peng ZK, Yang W, Zhang WM
Publication type: Article
Publication status: Published
Journal: ASME Journal of Vibration and Acoustics
Print publication date: 01/10/2016
Online publication date: 23/06/2016
Acceptance date: 13/05/2016
ISSN (print): 1048-9002
ISSN (electronic): 1528-8927
Altmetrics provided by Altmetric