Files in this item



application/pdf9416337.pdf (11MB)Restricted to U of Illinois
(no description provided)PDF


Title:Identification of structural systems from measured response
Author(s):Banan, Mohammad Reza
Doctoral Committee Chair(s):Hjelmstad, Keith D.
Department / Program:Civil and Environmental Engineering
Discipline:Civil Engineering
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Engineering, Civil
Abstract:This research study presents an approach to the problem of parameter estimation of finite element models of complex structural systems using measured response. The developed method computes the element, constitutive parameters of a finite element model of a structure by finding a constrained nonlinear minimum of the difference between the measured response of the real structure and the computed response of a finite element model of the structure. Constraints are used to bound the constitutive parameters.
The proposed constrained nonlinear optimization problem is solved using a recursive quadratic programming (RQP) method. The constructed recursive quadratic programming algorithm is an attractive local optimization method that applies directly to problems with inequality as well as equality constraints, it is globally convergent, and it is amenable to large-scale computation. The RQP method is an iterative gradient search method that needs the gradient and Hessian of the loss function with respect to the unknown variables. A simple and straightforward method is developed to compute these sensitivities of the loss function with respect to the unknown variables. All the computations are in the element level.
The method presented herein has been implemented as a general purpose parameter estimation program for structural systems. This program has all the flexibilities of a general purpose finite element package in that it can treat complex structures with different topologies, geometries and element types. It is suited to static, modal dynamic, or forced dynamic response, sparsely sampled in space, time, and state. It has a grouping scheme for considering the elements with the same constitutive parameters. Another feature of the developed package is to implement a priori knowledge about some of the constitutive parameters. Furthermore, it has strategies to compute initial values of the unknown variables for starting the iteration.
Monte Carlo simulation is used to study the behavior of the developed parameter estimation method in the presence of noise in measurements. Extensive simulations are carried out for both the static and force dynamic cases. A case study, using real field measurements is presented for the modal dynamic case.
Issue Date:1994
Rights Information:Copyright 1994 Banan, Mohammad Reza
Date Available in IDEALS:2011-05-07
Identifier in Online Catalog:AAI9416337
OCLC Identifier:(UMI)AAI9416337

This item appears in the following Collection(s)

Item Statistics