Damage Assessment of a Building Under An Earthquake Via the Tvarx Model and Bayesian Approach

Abstract eng:
The work proposes a procedure for the damage assessment of a building under a strong earthquake. The Cauchy wavelet transform is applied to the measured acceleration responses of a building and a time-varying autoregressive with exogenous input (TVARX) model with time-variant coefficients expanded as piecewise polynomial functions. The time-variant modal parameters of the building are directly identified from the coefficients of the appropriately established TVARX model. A Bayesian approach for dynamical model updating is employed to identify the variations of stiffness components with time using those identified time-varying modal parameters. Consequently, the damaged stories of the building are located and the severity of the damage can be determined. The proposed approach is validated through processing the numerically simulated acceleration responses of a six-story shear building with time varying stiffness and damping under base excitation with considering the effects of noise and incomplete measurement. Finally, the proposed approach is applied to handle the acceleration responses of a five-story steel frame in a shaking table test with a strong earthquake input. The frame is 3 m long, 2 m wide and 6.5 m high, and its columns of the first story yield during the test.

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