A new data analysis method, referred to as the adaptive quadratic sum-squares error(AQSSE) method,has been investigated for the online identification of time-varying system parameters of structures. Experimental study was conducted and presented to verify the capability of the AQSSE approach for identifying and tracking the variations of structural parameters. A base-isolated building model was tested in the laboratory. A stiffness element device was proposed to simulate the abrupt reduction of model stiffness due to damage during the tests. Both El Centro and Kobe earthquake excitations were used to drive the test model. Different damage scenarios were simulated and tested. Then the AQSSE approach was used to track the variation of the stiffness during the test based on the measured acceleration response data. The tracking results for the stiffness variations correlate well with that of the referenced values obtained by finite element method. It is concluded that the AQSSE approach is effective and accurate in tracking the variations of structural parameters leading to the detection of structural damages, including the time, location and severity of the damage.
Yin Q,Zhou L,Wang X.Parameter identification of hysteretic rubber-bearing based on sequential nonlinear least-square estimation [J].Journal of Earthquake Engineering and Engineering Vibration.2010,9(3):375-383