Road infrastructures systems are critical in many regions of Italy, counting thousands of bridges and viaducts that were built over several decades. A monitoring system is therefore necessary to monitor the health of these bridges and to indicate whether they need maintenance. Different parameters affect the health of an infrastructure, but it would be very difficult to install a network of sensors of various kinds on each viaduct. For this purpose, we want to finalize the use of geomatics technologies to monitor infrastructures for early warning issues and introducing automations in the data acquisition and processing phases. This study describes an experimental sensor network system, based on long term monitoring in real-time while an adaptive neuro-fuzzy system is used to predict the deformations of GPS-bridge monitoring points. The proposed system integrates different data (used to describe the various behaviour scenarios on the structural model), and then it reworks them through machine learning techniques, in order to train the network so that, once only the monitored parameters (displacements) have been entered as input data, it can return an alert parameter. So, the purpose is to develop a real-time risk predictive system that can replicate various scenarios and capable to alert, in case of imminent hazards. The experimentation conducted in relation to the possibility of transmitting an alert parameter in real time (transmitted through the help of an experimental control unit) obtained by predicting the behavior of the structure using only displacement data during monitoring is particularly interesting.
Road Infrastructure Monitoring: An Experimental Geomatic Integrated System / Barrile, V.; Fotia, A.; Bernardo, E.; Bilotta, G.; Modafferi, A.. - 12252:(2020), pp. 634-648. (Intervento presentato al convegno 20th International Conference on Computational Science and Its Applications, ICCSA 2020 tenutosi a ita nel 2020) [10.1007/978-3-030-58811-3_46].
Road Infrastructure Monitoring: An Experimental Geomatic Integrated System
Barrile V.;Fotia A.
;
2020-01-01
Abstract
Road infrastructures systems are critical in many regions of Italy, counting thousands of bridges and viaducts that were built over several decades. A monitoring system is therefore necessary to monitor the health of these bridges and to indicate whether they need maintenance. Different parameters affect the health of an infrastructure, but it would be very difficult to install a network of sensors of various kinds on each viaduct. For this purpose, we want to finalize the use of geomatics technologies to monitor infrastructures for early warning issues and introducing automations in the data acquisition and processing phases. This study describes an experimental sensor network system, based on long term monitoring in real-time while an adaptive neuro-fuzzy system is used to predict the deformations of GPS-bridge monitoring points. The proposed system integrates different data (used to describe the various behaviour scenarios on the structural model), and then it reworks them through machine learning techniques, in order to train the network so that, once only the monitored parameters (displacements) have been entered as input data, it can return an alert parameter. So, the purpose is to develop a real-time risk predictive system that can replicate various scenarios and capable to alert, in case of imminent hazards. The experimentation conducted in relation to the possibility of transmitting an alert parameter in real time (transmitted through the help of an experimental control unit) obtained by predicting the behavior of the structure using only displacement data during monitoring is particularly interesting.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
