Comparative Study of Output-Based Modal Identification Methods Using Measured Signals From an Offshore Platform

Abstract

Research on vibration based damage identification using changes of modal parameters has been expanding rapidly over the last decades. And modal parameter identification methods are essentially important since these parameters directly affect the damage detection results. Many modal identification methods have been proposed and further verified based on numerical and laboratory data. However, few papers are reported on comparative investigation of several output-only modal identification methods, especially based on measured signals from offshore platform in service. The main objective of the present paper is to investigate the effectiveness and applicability of several output-only modal identification methods, including Ibrahim Time Domain(ITD), Eigensystem Realization Algorithm(ERA), Auto-Regressive Moving Average method(ARMA) and Stochastic Subspace Identification(SSI) methods. First the four modal identification methods are briefly reviewed. Field measurements are conducted and vibration signals are recorded for modal identification. The effectiveness is compared for the four identification methods. Useful results could be obtained.