Work

Evaluation of effects on a viaduct affected by a snowmelt-induced landslide using AI-Driven Pattern Recognition

Year: 2024

Type: article

Abstract: Over recent years, the worldwide increase in infrastructure failures due to natural hazards such as landslides and earthquake has raised significant concern.Aggravated by the effects of climate change, this issue has had a considerable impact, particularly in Italy.This study focuses specifically on the "Villa Ilii" Viaduct located on the A24 highway to understand how landslides interact with bridge structures.The methodology adopted in this study is thorough, starting with an in-depth analysis of the geological context and the specific characteristics of the landslides in the area.Comprehensive field surveys, along with detailed morphometric and geophysical analyses, form an integral part of this research.These efforts are key to identifying and understanding the significant signs of deformation, which are critical in assessing the potential impacts of landslides on the "Villa Ilii" Viaduct's structural integrity.A novel element in this study is the application of Gradient Boosting Machines (GBM), an advanced machine learning technique, for automated pattern recognition within SAR data processed using the Differential Interferometry SAR technique.This approach leverages two distinct datasets: one detailing snow melting patterns in the AoI and another outlining the cumulative deformation of radar targets.GBMs are employed to uncover intricate, non-linear relationships between snow melting dynamics and ground deformation patterns.The integration of GBMs into this research represents a significant advancement in the field of landslide hazard assessment.By effectively analysing complex and multi-dimensional datasets, this approach offers a more nuanced and predictive understanding of the interplay between natural processes and infrastructure stability, thereby contributing to more effective hazard management strategies and infrastructure design.

Source: Research and Review Journal of Nondestructive Testing

DOI: 10.58286/30489