Premium
A parametric identification scheme for non‐deteriorating and deteriorating non‐linear hysteretic behaviour
Author(s) -
Ashrafi Seyed Ali,
Smyth Andrew W.,
Betti Raimondo
Publication year - 2006
Publication title -
structural control and health monitoring
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.587
H-Index - 62
eISSN - 1545-2263
pISSN - 1545-2255
DOI - 10.1002/stc.142
Subject(s) - weibull distribution , parametric statistics , identification (biology) , displacement (psychology) , parametric model , mathematics , identification scheme , system identification , engineering , computer science , process (computing) , statistics , data modeling , psychology , botany , software engineering , psychotherapist , biology , operating system
The identification of deteriorating and non‐deteriorating non‐linear rate‐independent hysteretic phenomena is considered using a recently developed model form. The model type, developed by authors Ashrafi and Smyth is a generalized Masing model and is based on the observed memory behaviour of distributed element models. The model's direct analytical link between the force–displacement response functions and the distribution of the distributed element yield displacements together with the ductility permits a parametric identification to be performed using non‐linear optimization techniques for arbitrary response time histories. The Weibull and Rayleigh distributions are considered in this study. The identified parameters accurately determine the physical model and are robust when the simulation model assumed distribution was the same as that used for identification, and also in the over‐parametrized case when the model was more complex than necessary. In the under‐parametrized case, where the assumed model lacked the complexity of the underlying phenomenon, relatively accurate response agreement was obtained without accurate parameter convergence. Such a result was not robust in the validation phase with other arbitrary excitations. Overall however, assuming sufficient complexity in the identified model, the approach provides a means of identifying deterioration with a fixed‐parameter model suitable for simulation with arbitrary excitations. Copyright © 2005 John Wiley & Sons, Ltd.