De sistemas a series estocasticas : aplicações a analise de vibrações no dominio do tempo

AUTOR(ES)
DATA DE PUBLICAÇÃO

1992

RESUMO

The present work describes the use of time domain identification techniques in the modal analysis of mechanical structures. The fundamentals of continuous system discretization and modal analysis are presented in a suitable maner for the study of dynamic systems in such a domain. Time domain identification models employed in other areas (electric engineering, statistics and economics) are related to the modal model with the objective of identification of natural frequencies and mode shapes. Stochastic series algorithms (ARMA,ARlMA) are presented and the Prony series is introduced as a tool for the identification of systems subjected to different kinds of excitation. There are also shown examples of using the algorithms in numerical simulations and a pratical modal analysis of a beam subjected to an impulsive forcing. The aim of the work is to introduce to the dynamic analist some of the most important techniques of time domain modal identification. The examples shown illustrate the potential of the algorithms to the identification of systems even with coincident natural frequencies and also in pratical modal analysis. The work suggests further developments in the field of computational poles determination, more efficient numerical methods for the solution of Prony s problem and the identification of non-linearities in the time domain

ASSUNTO(S)

vibração sistemas estocasticos

ACESSO AO ARTIGO

Documentos Relacionados