کد مقاله کد نشریه سال انتشار مقاله انگلیسی نسخه تمام متن
287447 509566 2014 19 صفحه PDF دانلود رایگان
عنوان انگلیسی مقاله ISI
Nonparametric identification of nonlinear dynamic systems using a synchronisation-based method
ترجمه فارسی عنوان
شناسایی غیرپارامتری از سیستم های پویا غیر خطی با استفاده از یک روش مبتنی بر هماهنگ سازی
موضوعات مرتبط
مهندسی و علوم پایه سایر رشته های مهندسی مهندسی عمران و سازه
چکیده انگلیسی

The present study proposes an identification method for highly nonlinear mechanical systems that does not require a priori knowledge of the underlying nonlinearities to reconstruct arbitrary restoring force surfaces between degrees of freedom. This approach is based on the master–slave synchronisation between a dynamic model of the system as the slave and the real system as the master using measurements of the latter. As the model synchronises to the measurements, it becomes an observer of the real system. The optimal observer algorithm in a least-squares sense is given by the Kalman filter. Using the well-known state augmentation technique, the Kalman filter can be turned into a dual state and parameter estimator to identify parameters of a priori characterised nonlinearities. The paper proposes an extension of this technique towards nonparametric identification. A general system model is introduced by describing the restoring forces as bilateral spring-dampers with time-variant coefficients, which are estimated as augmented states. The estimation procedure is followed by an a posteriori statistical analysis to reconstruct noise-free restoring force characteristics using the estimated states and their estimated variances. Observability is provided using only one measured mechanical quantity per degree of freedom, which makes this approach less demanding in the number of necessary measurement signals compared with truly nonparametric solutions, which typically require displacement, velocity and acceleration signals. Additionally, due to the statistical rigour of the procedure, it successfully addresses signals corrupted by significant measurement noise. In the present paper, the method is described in detail, which is followed by numerical examples of one degree of freedom (1DoF) and 2DoF mechanical systems with strong nonlinearities of vibro-impact type to demonstrate the effectiveness of the proposed technique.

ناشر
Database: Elsevier - ScienceDirect (ساینس دایرکت)
Journal: Journal of Sound and Vibration - Volume 333, Issue 24, 2 December 2014, Pages 6405–6423
نویسندگان
, ,