Nonlinear Model Updating Using Ambient Responses For Damage Diagnosis In Concrete Structures

点击翻译按钮获取中文摘要

Abstract: The focus of the research is on the development of a reliable method to update nonlinear finite element models of concrete structures to represent the damaged state of the structure. Ambient responses, combined with an understanding of the structure?s fundamental hysteretic behavior, will be used to generate the updated nonlinear finite element model. Such updated models are essential for determining the remaining lifetime of these structures and identifying repair/replacement priorities. Current global, vibration-based techniques have the ability to accurately identify linear elastic stiffness parameters for an appropriately defined identification model through ambient testing. However, existing damage diagnosis techniques are primarily limited to linear models in which the stiffness of some of the elements is reduced to represent damage by relating it to the change in the modal characteristics of the structure. The primary goals of the research efforts include: gaining an understanding of the fundamental behaviors of this class of structures; developing experimentally-verified, improved nonlinear constitutive models for civil engineering structures that can capture the salient features of the hysteretic response; developing a method for the assessment of the condition of a damaged structure when the elastic stiffness has deteriorated; validating the proposed methodology to relate structural damage to changes in elastic stiffness, resulting in a model that will be able to predict the capacity. Degradation in the elastic stiffness of the structure will be related to a reduced set of damage parameters that identify the state of the constitutive models in the nonlinear finite element model. Parametric identification models will be constructed appropriately to relate the damage parameters to the reduced initial stiffness values based on ambient response measurements. Currently available modal identification and parameter identification techniques based on ambient vibration responses will be employed and improved for determination of elastic stiffness parameters.