A Galerkin discretisation-based identification for parameters in nonlinear mechanical systems

In the paper, a new parameter identification method is proposed for mechanical systems. Based on the idea of Galerkin finite-element method, the displacement over time history is approximated by piecewise linear functions, and the second-order terms in model equation are eliminated by integrating by...

Full description

Saved in:
Bibliographic Details
Published inInternational journal of systems science Vol. 49; no. 5; pp. 908 - 919
Main Authors Liu, Zuolin, Xu, Jian
Format Journal Article
LanguageEnglish
Published London Taylor & Francis 04.04.2018
Taylor & Francis Ltd
Subjects
Computer simulation
continuous-time system
Finite element method
Galerkin method
Iterative methods
Linear functions
low sampling frequency
Mechanical systems
Nonlinear systems
Parameter identification
System effectiveness
Online AccessGet full text

Cover

More Information
Summary:In the paper, a new parameter identification method is proposed for mechanical systems. Based on the idea of Galerkin finite-element method, the displacement over time history is approximated by piecewise linear functions, and the second-order terms in model equation are eliminated by integrating by parts. In this way, the lost function of integration form is derived. Being different with the existing methods, the lost function actually is a quadratic sum of integration over the whole time history. Then for linear or nonlinear systems, the optimisation of the lost function can be applied with traditional least-squares algorithm or the iterative one, respectively. Such method could be used to effectively identify parameters in linear and arbitrary nonlinear mechanical systems. Simulation results show that even under the condition of sparse data or low sampling frequency, this method could still guarantee high accuracy in identifying linear and nonlinear parameters.
ISSN:0020-7721
1464-5319
DOI:10.1080/00207721.2018.1432781