Deep learning for structural health monitoring: An application to heritage structures

TL;DR

This paper applies deep learning, specifically a Temporal Fusion Transformer, to analyze seismic ambient noise data for structural health monitoring of heritage buildings, enabling anomaly detection without supervision.

Contribution

It introduces an unsupervised deep learning approach using a Temporal Fusion Transformer for anomaly detection in long-term structural health monitoring data.

Findings

Successful detection of anomalies in the San Frediano bell tower data

Demonstrated effectiveness of deep learning in heritage structure monitoring

Provided a new methodology for non-invasive structural health assessment

Abstract

Thanks to recent advancements in numerical methods, computer power, and monitoring technology, seismic ambient noise provides precious information about the structural behavior of old buildings. The measurement of the vibrations produced by anthropic and environmental sources and their use for dynamic identification and structural health monitoring of buildings initiated an emerging, cross-disciplinary field engaging seismologists, engineers, mathematicians, and computer scientists. In this work, we employ recent deep learning techniques for time-series forecasting to inspect and detect anomalies in the large dataset recorded during a long-term monitoring campaign conducted on the San Frediano bell tower in Lucca. We frame the problem as an unsupervised anomaly detection task and train a Temporal Fusion Transformer to learn the normal dynamics of the structure. We then detect the anomalies by looking at the differences between the predicted and observed frequencies.