Development of modal analysis technique using bivariate time series

Abstract

The Extended Autoregressive Moving Average(EARMA) modeling program in the Dynamic Data System(DDS) methodology has been more improved for the modal analysis of mechanical structures. The method is composed of the following 3 steps; (1) Fitting of EARMA models for the experimental data (2) Computation of natural frequencies and damping ratios from the characteristic roots of the model (3) Computation of mode shapes from the model parameters based on the characteristic roots derived in step (2) To examine the applicability, the method has been applied to the vibration data of a 3 DOF lumped mass system simulated by a digital computer. The modal parameters obtained by the proposed method have shown good agreement with the analytical solutions. A milling machine has also been analyzed by the method. The results have been compared with those obtained by the conventional modal testing method. With these successful applications, the method is believed to be a useful tool in the experimental modal analysis of mechanical structure.