At present, the buildings under damage assessment are actually with minor or serious damages, so the damage indices evaluated accordingly may not be adequate. To solve the problem, the experimental method is used to simulate different states of damage, followed by identifying the parameters of the associated damaged structure. Then the damage indices describing the damage degree can be computed using the identified parameters, and the threshold value of different damage state can be reached. In this paper, the technique of system identification is used to analyze the Benchmark Model D. The model is three-dimensional steel structure without diagonal bracings, which is 3m in the longitudinal (x axis) direction and 2m in the transverse (y axis) direction. A series of experiments have been conducted on the structure which can be divided into three stages, including the first-stage linear test, the second-stage linear test and the third-stage nonlinear test. The first-stage linear test is a series of linear shaking table tests; the second-stage linear test is also a series of linear shaking table tests, but the bottom of the two columns of the first floor were tapered to simulate the weak section; the model of the third-stage non-linear test is the same model as that of the second-stage linear test, and a series of non-linear shaking table tests with earthquake of high intensity were conducted. The analysis of this paper is first based on the identified results of first-stage linear test, and then the degree of damage is determined by the identified results of the second-stage linear test. Finally, a comparison is made between the first-stage linear test and the third-stage nonlinear test to determine the degree of damage and the location or floor of damage.
|