Papers by Author: Giacomo V. Demarie

Abstract: The identification of non-linear systems is an important topic in structural health monitoring of structures undergoing non-stationary behavior. In general, a non-linear or hysteretic response is typical for buildings, bridges, dampers and structural elements not only as a consequence of strong excitations (i. e. earthquake), but also for low to medium loading levels, due to the constitutive behavior of structural elements or joints. This paper focuses on the non-linear identification of a RC beam-column joint, modeled as a SDoF system, subjected to non-stationary loading: the technique used entails the definition of proper instantaneous estimators of the system dynamic properties by using a linear time-varying approximation of the actual system dynamics and representing the structural response in the joint time-frequency domain.

Abstract: This paper presents a technique for identification of non-linear hysteretic systems subjected to non-stationary loading. In the numerical simulations, a Bouc-Wen model was chosen for its ability to represent the properties of a wide class of real hysteretic systems. The parameters of the model are computed instantaneously by approximating the internal restoring force surface through an “ad hoc” polynomial basis. Instantaneous estimates result from time-varying spectra of the response signals. A numerical application of interest to earthquake engineering is finally reported.

Abstract: In structural engineering applications a sufficient quantity of experimental data to be able to achieve a consistent estimate of nonlinear quantities is seldom available: this applies in particular when the structures are to be tested in situ. This report discusses the definition of instantaneous estimators to be used in the dynamic identification of invariant nonlinear systems on the basis of Short-Time Fourier Transform representation of excitation and system’s response and within the framework of a Volterra series representation of the input/output relationship. An estimation of the parameters of a dynamic system can be worked out from the evolution of such instantaneous estimators.