Crowdsensing-based Indirect Bridge Structural Health Monitoring using smartphones: Application to a footbridge in Bologna

Bridge Structural Health Monitoring (SHM) is becoming a pressing concern due to the aging and degradation of the population of long- and short-span bridges and viaducts built in Italy over 40 years ago along more than 3,000 km of the national road network. The knowledge of the modal parameters and their eventual change in time can be used as indicators of the presence of potential structural damage, allowing also for the prediction of the remaining operational life of the bridge itself. This can increase the efficiency of the infrastructure management framework. In this context, traditional vibration-based SHM is implemented through the use of a costly network of fixed sensors on the structure, such as accelerometers, to acquire vibration data under operating conditions, falling into the Operational Modal Analysis framework. In the past years, researchers worldwide have studied the feasibility of the application of indirect Bridge Structural Health Monitoring (iBSHM), or Drive-by monitoring, where data are collected by sensors installed on a moving vehicle. This alternative method potentially offers a more cost-effective solution for bridge management. Moreover, recent studies have shown that smartphones can be employed as low-cost and reliable sensors to be used to assess bridges' health conditions, considering both single-sensor solutions and crowdsensing-based ones. These latter ones consist of collecting a large number of data from all the vehicles passing over the infrastructure. This approach has already been proven efficient in increasing road safety and traffic monitoring. In this thesis, a crowdsensing-based iBSHM approach is presented, exploiting the increasing precision and connectivity of modern smartphones installed on light vehicles, a commercial bicycle in this case. This study offered the possibility to simulate the application of this new method to shared micromobility vehicles, whose standard geometries and properties make them suitable for crowdsensing applications. In this study, the Covariance-driven Stochastic Subspace Identification (SSI-COV), one of the most known Operational Modal Analysis algorithms, is used to identify natural frequencies and damping ratios of a real footbridge in Bologna from acceleration data acquired during several passages of a bicycle with two smartphones installed. An approach based on the Extended Kalman Filter and the Hilbert Transform has been implemented to complete the bridge modal identification and retrieve the first mode shape.