<p style="text-align: justify">The continuous Non-Destructive Evaluation of assets for long-term assurance of performance has led to several developments over the deployment of an online Structural Health Monitoring (SHM) system. Considering the challenges involved under the implementation of an SHM system for the applications working under harsh environmental conditions with limited access to power sources, this work is anticipated to contribute towards the development of a Passive Ultrasonic SHM system that works without an input power source and also overcome the detrimental effects of the harsh environmental conditions. This kind of system is found to be realized using the cross-correlation technique of a fully diffused noise field, with the noise field being generated within the structure due to structure’s machinery or any ambient source and with the application of Fibre Bragg Grating (FBG) sensors for sensing the wave field. Hence, having the structure interrogation without any power source and the FBG sensors providing the applicability under harsh environments. Working over this notion, initial understandings on the principle, parameters, suitability, and the favorable conditions of the technique are obtained through finite element simulations using the commercial Abaqus/Explicit software package. It is observed that in the directions having a diffused noise field there is a reconstruction of a wavefield, which would have been obtained conventionally by placing a transmitter and a receiver at the passive sensing locations. Thus, showing the potentiality of the technique on the identification of the defect and its localization. The understandings obtained from the results of simulations are further proposed to be applied towards effective universal deployment of the technique over defect identification and localization using a Deep Learning based methodology. Since the discussed passive sensing system gives the cross-correlated output representing the actual signal, which would have been obtained conventionally using a transmitter and a receiver. Hence, it is anticipated that the cross-correlated outputs obtained from a particular arrangement of an array of sensors can be analyzed for detecting a distinct similarity among them, occurring due to the presence of a defect in their traversing paths. To this end, a Long-term Recurrent Convolutional Network (LRCN) model is proposed, to detect the outputs containing the distinct similarity representing a particular defect in the structure. The outputs obtained from the LRCN model can be further traced to localize the defect. Further, the model is aimed towards overcoming the shortcomings of present conventional imaging techniques using passive acquisitions and hence improve the defect analysis towards an effective SHM system.</p>

Development of a Convolutional Lstm Network Assisted Fibre Optics Based Passive Structural Health Monitoring System

Paper Proceedings: https://asmedigitalcollection.asme.org/qnde/QNDE2021/volume/85529

ASME 2021 48th Annual Review of Progress in Quantitative Nondestructive Evaluation (QNDE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.