Özet:
Two separate methods are proposed and evaluated in this study for the structural health monitoring and damage detection of mid -to-high rise reinforced concrete buildings. The first method discussed herein, aims to predict the response of the structure at non-instrumented floors. A numerical structural model of the real-life structure is constructed and the modal response quantities are retrieved. The measured mode shapes of the real structure are considered as linear combination of the calculated mode shapes of the numerical model. For each mode shape to be superimposed, a weighting coefficient is obtained through minimizing the error between the actual and superimposed mode shapes and the final response is estimated by using the relation between the physical and modal coordinates. The second method is based on wave propagation in buildings and reverse nonlinear structural analysis. The building floors are first scanned with limited number of sensors and the impulse response function (IRF) for each story is identified by deconvolution. Once the IRFs are determined, the story responses are obtained via convolution of the reference signal and the story IRFs at any given instance; and finally, the interstory drifts are approximated. The internal forces and deformations of the structural members are computed by the reverse inelastic structural analysis utilizing nonlinear fiber models, kinematic, and joint equilibrium. The extent and limits of convolution operation as for inelastic structural response is investigated. The change in the shear wave velocity of the damaged are also evaluated.
