PSI - Issue 78

Carlo Rainieri et al. / Procedia Structural Integrity 78 (2026) 426–432

431

= − "

(2) Finally, the novelty index (NI) has been computed as the L2-norm of the residuals E, as EOV compensated data matrix. In the context of this paper, the training period spans between November 4 th , 2024 and January 31 st , 2025, while the remaining data have been considered for validation and anomaly detection. The threshold has been set considering a 5% probability of exceedance of the NI in the training period. The NI time history after the training period is shown in Fig. 3. The results show that, apart from a few outliers overcoming the threshold, no significant changes in the structural response occurred as a result of the earthquakes which hit the structure. This is consistent with the results of visual survey and inspections carried out on the structure, confirming the absence of any damage.

Fig. 3. NI time series after the training period.

3. Conclusions The paper described the modal-based SHM system recently installed at the School of Engineering Main Building of the University of Naples “Federico II”. The building is characterized by a relevant architectural value, and it is located nearby the Campi Flegrei, a large and active volcanic area generating frequent seismic sequences. After a description of the technological solutions adopted for the system, the measured ambient vibration response of the structure has been first manually processed in order to check the monitoring system functionality and to get reference estimates of the fundamental modal parameters to validate the automatic outcomes of the SHM system. Consistent results have been obtained by applying two well-established OMA procedures. Moreover, they are also consistent with the results obtained from a previous monitoring system installed on the structure about two decades ago and decommissioned a few years later. Results of continuous monitoring of the fundamental natural frequencies have been also presented and discussed, highlighting the EOV influence on the natural frequency patterns. Moreover, no changes in the dynamic properties have been detected, that could have been interpreted as a result of possible earthquake-induced damage. Thus, the paper confirmed the key role of OMA in the development of effective SHM technologies. The SHM system is still active and collecting data about the operational and seismic response of the structure that will be the object of further studies. Acknowledgments The present study is part of the research activities developed by the authors within the framework of the PNRR Program, “Multi-Risk sciEnce for resilienT commUnities undeR a changiNg climate (RETURN - PE00000005)” Research Project (CUP E63C22002000002) and the associated cascade funding calls: PE00000005_1 “MITIGATE - Monitoring buIlT-up envIronment throuGh dinsAr Time series” (CUP E63C22002000002) and “ATTENTIONS - multiscAle strucTural, geoTechnical and Exposure data fusioN for healTh monItOring aNd risk assesSment –” PE00000005, (CUP H93C22000610002). Additional support from the STRIVE – INOSTRI FOE Project is also gratefully acknowledged by the first author. References

Carden, E.P., Fanning, P., 2004. Vibration based condition monitoring: a review, Structural Health Monitoring 3, 355–377.