A Hybrid Method for Structural Modal Parameter Identification Based on IEMD/ARMA: A Numerical Study and Experimental Model Validation

This article presents a hybrid method of structural modal parameter identification, based on improved empirical mode decomposition (EMD) and autoregressive moving average (ARMA). Special attention is given to some implementation issues, such as the mixing, false modes, judgment real intrinsic function (IMF) classical EMD, difficulty fixing order ARMA. To resolve existing defects an EMD (IEMD) that combines frequency band filtering cluster analysis proposed in this paper, where divides signal into several narrowband signals before process, used determine IMFs. Euclidean distance results, with no need adjust any indexes or thresholds, only by means using nearest efficiently IMF. Moreover, IEMD pre-processing tool for ARMA, its order. The capabilities were compared assessed numerical simulation experimental model. results showed could mixing problems process automatically identified IMFs, while was combined ARMA successfully identify shape structure. Additionally, since each IMF single component signal, it easy