Modal identification of concrete arch dam by fully automated operational modal identification

Abstract

Modal identification is a type of system identification, which studies on the modal parameters of systems by using modal test. In the case of using operational or ambient modal analysis, there is no need to measure excitation, and the system output data are adequate for identification purposes. These modal parameters of system are of great importance from the engineering point of view particularly in the area of system identification, damage detection, and condition monitoring. In a fully-automated identification approach, the modal parameters are extracted without intervention of a specialized user. In this study, a Fully Automated Operational Modal Identification algorithm is developed to identify modal parameters of an arch dam. For this purpose, Morrow point arch dam is selected, and a 3-D finite element of the dam is established. In the first step, the computed natural frequencies of the dam based on the forced vibration tests are utilized to verify the developed finite-element model of the dam. Afterwards, the dam is excited by El-Centro earthquake, and its seismic response is recorded. The collected signals are presented to the proposed identification algorithm. The proposed algorithm presents a methodology to classify the physical mode shapes from spurious modes using the Stochastic Subspace framework. The results indicate acceptable identification via the proposed algorithm using a dataset and non-repetitive data acquisition. Moreover, the identification process is automated and it does not require any user interaction. It is noteworthy to state that the algorithm can identify any dam with any number of mode shapes, and it is practical and straightforward to apply.