Aingura IIoT, Spain
Industrial Internet of Things (IIoT) and machine learning technologies can provide useful real-world applications on the structural health monitoring (SHM) sector giving useful actionable insights towards bridge maintenance. Therefore, the combination of data acquisition, pre-processing and machine learning techniques, working within embedded computing nodes located at bridges, can provide indicators to help on their maintenance. These types of solutions can help to prioritize bridges depending on their health using objective indicators that can help to improve current inspections based on subjective parameters. Therefore, this keynote describes state-of-the-art advances in each IIoT phase towards SHM, in terms of bridge instrumentation, new feature generation, dynamic machine learning approaches and edge computing. As an example of these applications, a use-cases involving four real deployments at Massachusetts (U.S.A) area are described. Effectiveness of these methodologies in terms of estimating the modal parameters of the bridge and monitoring their trend as an indicator of bridge health is shown. Specifically, the keynote will showcase how the bridge behavior pattern is developed and monitored based on normal traffic, without complex approaches such as finite elements models or expensive instrumentation, enabling the solution the be potentially applied to conventional bridges.