Urban environments face numerous challenging vibration-related issues that are difficult to resolve due to the complex nature of urban infrastructure and the diverse sources of vibrations. These issues can stem from various factors, including traffic, construction activities, industrial operations, and even natural phenomena like earthquakes.

In this study, we employed Distributed Acoustic Sensing (DAS) as a means to monitor vibrations generated by vehicle-induced active sources, with the aim of mitigating vibration-related issues in urban environments. For DAS data processing, we proposed a novel semi-supervised model that enhances detection accuracy and mitigates false detections. Compared to traditional vehicular trace detection methods, our approach achieves the highest accuracy. Moreover, our model incorporates a self-updating mechanism that allows for continuous refinement based on newly collected data. This approach aims to provide a more reliable and adaptive solution for vibration monitoring in complex urban settings, ultimately contributing to the development of effective strategies for addressing vibration-related problems in cities.


Additional information: Mr. Xi WANG,