A new approach to fuzzy modeling of nonlinear dynamic systems with noise: relevance vector learning mechanism

2006 ◽  
Vol 14 (2) ◽  
pp. 222-231 ◽  
Author(s):  
Jongcheol Kim ◽  
Y. Suga ◽  
S. Won
2020 ◽  
Vol 20 (06) ◽  
pp. 2040012
Author(s):  
Changzhao Qian ◽  
Changping Chen

A new approach is proposed for diagnosing the nonlinear characteristics of vibratory systems in engineering applications. The correlation coefficient of two periods’ response wave is used in this approach and a new concept named similarity degree of waveform (SDW) is proposed to describe the correlations of the responses and therefore the characteristics of the nonlinear dynamic systems. A Duffing system is used in this research as an example to demonstrate the application and efficiency of the proposed approach. The approach has shown effectiveness in diagnosing periodic, multi-periodic, quasi-periodic and chaotic motions. The characterization with implementing the proposed approach is compared with that of the Lyapunov exponent method. The advantages of the proposed approach are demonstrated.


Author(s):  
James Kapinski ◽  
Alexandre Donze ◽  
Flavio Lerda ◽  
Hitashyam Maka ◽  
Edmund Clarke ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document