Structural Damage Identification Based On Symplectic Geometry Spectrum Analysis

With the increasing amount of bridges, dams and other important infrastructures.and the construction of the high-level, high-rise structures, the safety of these structures during the service time has drawn more and more attention of engineering construction personnel and researchers. In order to avoid the great loss of peoples life and property caused by the sudden destroy of these important structures, structural health monitoring(SHM) subject is developed. Damage identification algorithm is the core issue of structural health monitoring, having a direct impact on the results of structural health monitoring. In recent years, the damage identification method based on modern signal processing technology and the structure vibration test has been widely researched and developed. In this paper, the following work has been done in terms of structural damage identification:To demonstrate the importance of the study of the structural health monitoring, the practical engineering background is introduced. The damage identification algorithms that based on structural vibration characteristics are summarized. The effects of environmental factors and operation conditions on structural vibration characteristics are introduced.A nonlinear time series decomposition method called Symplectic Geometry Spectrum Analysis(SGSA) is introduced in detail. The theoretical derivation and implementation steps are shown. The SGSA method can decomposes a signal into a set of components that with different scales, and these components can be used to reconstruct the original system and signal denoising, a synthetic sine signal is decomposed to demonstrate the method. The ability of the SGSA method in terms of the trend extraction of a signal is shown through the decomposition of a measured temperature data.The SGSA can decomposed a signal into a set of independent components that with specific physical meaning, and the components can represent the local characteristics of the original signal in frequency domain. Referring to the idea that the HHT method in structural characteristic parameter identification, combining the SGSA method with the Hilbert transform to identifies structural characteristic parameters, the analyze of a numerical simulation and laboratory test data demonstrates the effectiveness of this method.For the structure that excited in a single point, according to the theory of the structural frequency respond function, a new damage index is derived only using the acceleration response data. Through the decomposition of the index by the SGSA method, the first decomposed component is used to identify the structural damage, and through the analyze of a laboratory structural model by this method, the effectiveness is demonstrated.Environmental factors and structural operation conditions have great influence on structure vibration characteristic parameters. The SGSA method can extract the trend of a signal, a computer-simulated date of the first frequency of a simply supported beam is analyzed by this method to remove the influence of environmental factors on the first frequency change. The result shows that the damage is identified by this method successfully.