Identification of Linear Structural Dynamic Systems

Abstract

This paper studies methods of parameter estimation for linear multi-degree-of-freedom structural dynamic systems, based on observed records of the external forces and the structural responses. The auto-regressive and moving-average (ARMA) model is used for this purpose. It is found that the ARMA model is a convenient model representing linear multi-degree-of-freedom structural dynamic systems and that the model is highly compatible with the instrumental variable method and the maximum likelihood method of identification. In order to check the accuracy of the estimation methods, analytical simulation studies are performed on the basis of simulated data dealing with the aerodynamic coefficient matrices that appear in the equations of motion of a two-dimensional model of a suspension bridge. Then, these methods are applied to the same equations to identify the coefficient matrices using the field measurement data yielding good estimates of the system parameters even under large output noise conditions.