Integration of system identification and finite element modelling of nonlinear vibrating structures. (1st March 2018)

• Record Type: Journal Article
• Authors: Cooper, Samson B.; DiMaio, Dario; Ewins, David J.
• Is Part Of: Mechanical systems and signal processing. Volume 102(2018)
• Page Start: 401
• Page End: 430
• Publication Date: 2018-03-01
• Subjects: Nonlinear identification -- Finite element -- Numerical simulation -- System integration; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2017.09.031
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: Integration of system identification and FE modelling of nonlinear structures is presented. Identification of Polynomial-type nonlinearities from experimental data. A finite element application of the nonlinearities is described. Implementation of nonlinear damping elements for improved nonlinear response prediction. Successful validation of the nonlinear FE model with experimental test data is shown. Abstract: The Finite Element Method (FEM), Experimental modal analysis (EMA) and other linear analysis techniques have been established as reliable tools for the dynamic analysis of engineering structures. They are often used to provide solutions to small and large structures and other variety of cases in structural dynamics, even those exhibiting a certain degree of nonlinearity. Unfortunately, when the nonlinear effects are substantial or the accuracy of the predicted response is of vital importance, a linear finite element model will generally prove to be unsatisfactory. As a result, the validated linear FE model requires further enhancement so that it can represent and predict the nonlinear behaviour exhibited by the structure. In this paper, a pragmatic approach to integrating test-based system identification and FE modelling of a nonlinear structure is presented. This integration is based on three different phases: the first phase involves the derivation of an Underlying Linear Model (ULM) of the structure, the second phase includes experiment-based nonlinear identification using measured time series and the third phase covers augmenting the linear FE model and experimental validation of the nonlinear FE model. The proposed case study is demonstrated on a twin cantilever beam assembly coupled with a flexible arch shaped beam. In this case, polynomial-type nonlinearities are identified and validated with force-controlled stepped-sine test data at several excitation levels.