Model updating of nonlinear structures from measured FRFs

Canbaloglu G., ÖZGÜVEN H. N.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING, vol.80, pp.282-301, 2016 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 80
  • Publication Date: 2016
  • Doi Number: 10.1016/j.ymssp.2016.05.001
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.282-301
  • Keywords: Nonlinear model updating, Nonlinearity, Nonlinear identification, Friction nonlinearity, Nonlinear structures, Multiple nonlinearity, FINITE-ELEMENT MODEL, RESPONSE FUNCTION DATA, SYSTEM IDENTIFICATION, HARMONIC RESPONSE, FE MODEL, STIFFNESS, PARAMETERS, SELECTION
  • Middle East Technical University Affiliated: Yes


There are always certain discrepancies between modal and response data of a structure obtained from its mathematical model and experimentally measured ones. Therefore it is a general practice to update the theoretical model by using experimental measurements in order to have a more accurate model. Most of the model updating methods used in structural dynamics are for linear systems. However, in real life applications most of the structures have nonlinearities, which restrict us applying model updating techniques available for linear structures, unless they work in linear range. Well-established frequency response function (FRF) based model updating methods would easily be extended to a nonlinear system if the FRFs of the underlying linear system (linear FRF5) could be experimentally measured. When frictional type of nonlinearity co-exists with other types of nonlinearities, it is not possible to obtain linear FRF5 experimentally by using low level forcing. In this study a method (named as Pseudo Receptance Difference (PRD) method) is presented to obtain linear FRFs of a nonlinear structure having multiple nonlinearities including friction type of nonlinearity. PRD method, calculates linear FRFs of a nonlinear structure by using FRF5 measured at various forcing levels, and simultaneously identifies all nonlinearities in the system. Then, any model updating method can be used to update the linear part of the mathematical model. In this present work, PRD method is used to predict the linear FRFs from measured nonlinear FRF5, and the inverse eigensensitivity method is employed to update the linear finite element (FE) model of the nonlinear structure. The proposed method is validated with different case studies using nonlinear lumped single-degree of freedom system, as well as a continuous system. Finally, a real nonlinear T-beam test structure is used to show the application and the accuracy of the proposed method. The accuracy of the updated nonlinear model of the test structure is demonstrated by comparing the calculated and measured nonlinear FRFs of the test structure at several different forcing levels. (C) 2016 Elsevier Ltd. All rights reserved.