St. Paolo Hospital – Naples

The described application concerns the monitoring of the structural health of a reinforced concrete healthcare facility subjected to frequent earthquakes through the analysis of time variations in its modal parameters.

The structure under monitoring is the San Paolo Hospital, located in the Campi Flegrei area of Naples, an area that since spring 2024 has been affected by numerous seismic events related to the phenomenon of bradyseism. The earthquakes that have occurred over the past two years have been numerous and, in some cases, of strong intensity, reaching and exceeding a magnitude of 4.0.

S2X was commissioned to supply and install a monitoring system with the aim of promptly identifying any changes or damage to the structure following seismic events. Indeed, this is a strategic facility for which it is not possible to interrupt operations as a preventive measure.

The Building is a reinforced concrete structure with planar frames, consisting of two building units, B1 and B2, which rise to five above-ground floors and one basement level. The two structural units have plan dimensions of approximately 50 × 15 m and 60 × 15 m, respectively, and are separated, along the shorter side, by an expansion joint.

In general, a structural monitoring system (Structural Health Monitoring, SHM) is based on the installation—over relatively long periods of time (several months or years) or for the entire operational life of a structure—of sensor networks managed by hardware/software systems that allow data from the sensors to be collected and processed automatically, identifying, through appropriate algorithms, the presence of structural malfunctions.

In this way, the system composed of the structure and the SHM setup forms an “smart” system, capable of self-diagnosis functions and of transmitting “messages” to a human operator. The primary capability of SHM systems is to detect structural damage or malfunctions of auxiliary equipment early and more reliably than traditional inspection operations, which are primarily based on visual observation of defects and local checks.

The continuous monitoring system installed at the aforementioned structure consists of 8 single-axis accelerometers, of which 5 are installed at the roof level and 3 are installed along three orthogonal directions at the foundation level. The system also includes an 8-channel dynamic acquisition unit with 24-bit resolution and a dynamic range of 102 dB, as well as an industrial PC for remote management of the monitoring system. Each sensor is connected to the acquisition system via appropriate cabling.

The continuous acquisition and processing (24/7) of data are managed by the industrial PC installed in the building. Through an internet connection, it is also possible to access and manage the monitoring system remotely. The monitoring system continuously records the structural response to ambient vibrations 24/7 and automatically estimates the dynamic properties of the structure in terms of vibration frequencies, damping ratios, and modal shapes at the measurement points. These parameters are further processed to compensate for the effects of environmental and operational variables and to highlight possible anomalies in the structural response.

S2-DDA is the software dedicated to the continuous acquisition and management of accelerometric data and their local storage. The response to ambient vibrations is acquired at the desired sampling frequency (100 Hz in the present case). The collected data are stored locally every 30 minutes, ensuring continuity between consecutive files.
In addition to continuous acquisition, a trigger-based acquisition has been set up, which writes data to a dedicated file whenever the acceleration at the base of the structure exceeds a preset threshold of 0.005 g. This threshold, useful for the automatic selection of data related to the structural response to seismic events, can be modified according to user requirements.
The software, in addition to displaying the acquired data, allows for online estimation of the Power Spectral Density associated with the time series. It features a user-friendly interface, and plug-and-play hardware recognition allows for immediate operation. Finally, the WatchDog functionality enables automatic restarting of the application in the event of unexpected stops, such as power outages.

The S2-SHM software, on the other hand, is the true core of the system. It continuously and fully automatically processes the data and extracts the modal parameters in order to monitor their evolution over time. Using machine learning algorithms to compensate for environmental and operational effects, it identifies potential anomalies or malfunctions in the structure and sends alerts via email.

The software used for this case study were S2-SHM and S2-DDA.