Damage identification of a flag - shaped hysteresis structure subject to seismic excitation

This research investigates the structural health monitoring of nonlinear structures after a major seismic event. It considers the identification of flag-shaped or pinched hysteretic behaviour in response to structures as a special, perhaps more general, case of a normal hysteretic curve without pinching. The method is based on the overall least squares solution and the asymptotic distribution theory of log likelihood ratio. In particular, the structural response is divided into different loading and unloading half-cycles. The overall least squares analysis is first implemented to obtain the minimum residual mean square of each half-cycle regression model with the number of segments assumed. Then the log likelihood ratio test is proposed to assess the likelihood of these nonlinear segments in the presence of noise. The least square fit is finally used for identified segmented regression models to obtain elastic stiffness, plastic stiffness, yielding deformation and energy dissipation parameters. The performance of the proposed method is illustrated using a single degree freedom system and a suite 20 earthquake records. The simulations for this proof of concept include 10% added noise. The proposed method is computationally efficient and accurate in identifying nonlinear hysteretic structures. These parameters are within 6% average (standard deviation of 10%) of the known values. These results indicate that the system is able to capture highly nonlinear behaviour and structural parameters directly relevant to damage and performance using a computationally efficient and simple method. Finally, the method requires no user input and could thus be automated and performed in real-time for each half cycle.