Damage assessment using flexibility and flexibility-based curvature for structural health monitoring

As a result of the recent advances in sensors, information technologies and material science, a considerable amount of research has been conducted in the area of smart infrastructures. While there are many important components of a smart infrastructure, an automated and continuous structural health monitoring (SHM) system is a critical one. SHM is typically used to track and evaluate the performance of a structure, symptoms of operational incidents, anomalies due to deterioration and damage during regular operation as well as after an extreme event. Successful health monitoring applications can be achieved by integrating experimental, analytical and information technologies on real-life operating structures. However, real-life investigations must be backed up by laboratory benchmark studies for validating theory, concepts, and new technologies. For this reason, a physical bridge model is developed to implement SHM methods and technologies. In this study, different aspects of model development are outlined in terms of design considerations, instrumentation, finite element modeling, and simulating damage scenarios. Different damage detection methods are evaluated using the numerical and the physical models. Modal parameter estimation studies are carried out to reliably identify the eigenvalues, eigenvectors and modal scaling from the measurement data. To assess the simulated damage, modal flexibility-based displacements and curvatures are employed. Structural behavior after damage is evaluated by inspecting the deflected shapes obtained using modal flexibility. More localized damage simulations such as stiffness reduction at a joint yield a very subtle stiffness decrease. In this case, the writers use a baseline to identify damage and also investigate the use of curvature as a complementary index. Curvature is advantageous for certain cases where the displacement results do not provide substantial changes. Issues related to using curvature as a damage identification index are also addressed.