Damage Identification for Structures Subjected to Severe Earthquakes Using Wavelet Transform

Abstract eng:
The availability of methods for damage identification of civil infrastructures under sever earthquake is crucial for safety assessment and repair decision of structures post earthquake. Two novel damage identification methods based on wavelet transform are proposed in this paper. In the first method, the structural seismic responses are dealt with by the continuous wavelet transform directly. The time-varying frequency of structure is identified, and then the occurrence time and degree of structural damage can be observed. However, this method can not work well for multi-degree of freedom system because of the complicated relationship between damage and structural frequencies. Then the second method is presented subsequently to identify nonlinear hysteresis curves of structures during strong earthquake. By this method, not only the location but the degree of structural damage can be identified. This method alternately uses the extended Kalman filter (EKF) and wavelet (W) multiresolution analysis. Then it is entitled as the EKF-W method. In each time step, the full structural states are first estimated by using the EKF based on limited observations; then the tangent stiffness as well as hysteresis curves of structures is identified by using the wavelet multiresolution analysis based on the estimated full states. Simulation results from two structures are presented to illustrate the power and efficiency of the proposed two methods.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket