Parameter Estimation of Structural Systems Possessing One or Two Nonlinear Normal Modes

Files
fahey.pdf (1.35 MB)
Downloads: 218
TR Number
Date
2000-10-30
Journal Title
Journal ISSN
Volume Title
Publisher
Virginia Tech
Abstract

In this Dissertation, we develop, and provide proof of principle for, parameter identification techniques for structural systems that can be described in terms of one or two nonlinear normal modes. We model the dynamics of these modes by second-order ordinary-differential equations based on the principles of mechanics, past experience, and engineering judgment. We perform a number of separate experiments on a two-mass structure using several different types of excitation. For the linear tests, the theoretical system response is known in closed-form. For the nonlinear test, we use the method of multiple scales to determine second-order uniform expansions of the model equations and hence determine the approximations to responses of the structure. Then, we estimate the linear and nonlinear parameters by regressive fits between the theoretically and experimentally obtained response relations. We report deviations and agreements between model and experiment.

Description
Keywords
signal processing, line spectrum, system identification, modal analysis, Nonlinear dynamics, building dynamics, vibration, structural dynamics
Citation