Detection and estimation of damage to steel frames through shaking table tests

In this paper, we show the results of two damage detection tests using steel frames. The first test structure is a five-storey steel frame with simulated damage, and the other one is a three-storey large-scale steel frame with cementitious devices. Two method identifying damage, one using changes in flexibility and one using shifts in natural frequency, are applied to the frame with simulated damage. Both methods can pinpoint the storey that has damage. To the frame with cementitious devices, earthquake excitations, white noise shaking and microtremor observations are carried out. Identification method by the ARX (auto regressive exogenous) model is applied to all three cases. Natural frequency decreases, damping ratio increases and storey stiffness decreases as the experienced amplitude or input amplitude increases. A model using stick–slip elements is proposed. Natural frequency, damping ratio and storey stiffness, described by this model, are consistent with experimental results.