Recursive Least-Squares Time Domain Identification of Structural Parameters

Abstract

A method of identifying structural parameters such as damping and stiffness of a building from its time response under dynamic excitation is presented. A least-squares recursive computer algorithm which requires no matrix inversion is developed and tested with the response of a two-degree-of-freedom structure including Gaussian white noise. The algorithm provides means to account for both the model uncertainty and the investigators’ confidence in the initial guess of the parameters. These statistical quantities can be updated with passage of time. The study indicates that rapid convergence to the correct values of the parameters takes place even under severe noise in the response data.