Abstract

AbstractApp4SHM is a smartphone application for structural health monitoring (SHM). It can be applied to perform SHM of bridges or other special structures to assess their condition after a catastrophic event or when required by authorities. The application interrogates the phone’s internal accelerometer to measure accelerations, then estimates the natural frequencies, and compares them with a reference data set through a machine learning algorithm properly trained to detect damage. The machine learning is fundamental to take into account the effects of operational and environmental variability on the damage detection. A server is accessed and used by the application to run most of the computational operations and store the data sets. As a customized SHM process, App4SHM follows four main steps: (i) structure identification; (ii) data acquisition; (iii) feature extraction, which calls the server to estimate the first three natural frequencies and stores them into a feature vector (observation); and (iv) damage detection, where a damage indicator is computed for each new observation, based on the Mahalanobis-squared distance. The damage indicator of the new observation is plotted, and a flag is raised green if the structure is undamaged and raised red if structural damage is suspected. To test the robustness of the application, the damage detection capability was tested on real data sets from two twin post-tensioned concrete bridges in Brazil under traffic and temperature variability. The natural frequencies obtained from the application were also compared with the ones estimated using data sets from a traditional data acquisition system.KeywordsStructural health monitoringDamage detectionBridgesDynamics of structuresMachine learningApp

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call