چكيده به لاتين
Modal identification of Structural systems is a procedure to estimate the modal parameters of a structure, which is an important part of structural health monitoring. The main challenge of identification in this domain is the inaccessibility of the input data, which restricts us to use of output-only identification methods. The modal identification methods can be classified in different types, based on the excitation source. In recent years the main part of output-only system identification methods has focused on ambient Excitation.
In order to identify the modal parameters of infrastructures, it is essential to have different output channels, moreover there are large amount of data in each channel and it is important to improve the accuracy of the identification by using the data-fusion methods. The identification of the structural systems is computationally complicated, hence, automated modal identification play an important role in modal identification of infrastructures. In this study, the ambient input modal identification and automated identification are examined. The main topics which introduced in this thesis can be organized as:
• Reviewing the Vector Backward Auto-Regressive (VBAR) identification method and extracting an interval for sampling frequency in this method.
•Reviewing the Subspace Identification (SSI) method.
•Introducing the FKF data-fusion method to improve the accuracy of the identification.
•Automatic system identification by using the main stabilization criteria in eliminating the spurious modes from the set of identified modes.
•Implementing the introduced methods on simulated and experimental data to verify the performance of the proposed algorithms.
Keywords: Output-only system identification, Vector Backward Auto-Regressive (VBAR), Stochastic Subspace Identification (SSI), Data-Fusion, Automated identification
