Physically-Based Data-Driven Approach for Anomaly Detection in Long-Term Continuous Monitoring Systems: Application to a Case Study

Abstract

Smart management of cultural heritage structures benefits from long-term continuous monitoring. Continuous monitoring collects large dataset of measurements that significantly improve the possibility to provide a correct interpretation of the structure’s behaviour and to detect evolutionary trends and anomalies. By referring to structural features which are systematically extracted from the monitored dataset and identified as Reference Quantities (RQ), a physically-based data-driven approach is developed in this research work and applied to the case study of the Two Towers in Bologna (Italy). The proposed approach streamlines data processing, correlates structural behaviour, accounts for environmental effects, and supports the establishment of alert thresholds for an effective structural monitoring for the considered case study.