PSI - Issue 44

Federico Germano et al. / Procedia Structural Integrity 44 (2023) 902–909 F. Germano et al./ Structural Integrity Procedia 00 (2022) 000 – 000

906

5

• The need of a system who can acquire different types of sensors, with different electrical connections and properties, conditioning them properly and sufficiently flexible to adapt itself to different scenarios as each structure is different from the others; • The need of an acceleration sensor with low noise floor, very good phase performance, robust against temperature variation, for OMA applications; • A flexible software suite, able to adapt to the difference of each structure, and open, able to receive and give data from and to other digital platforms (Asset Management, third party proprietary analysis, …). Of course, those requirements derived from technical need have to be coupled with general market requirement, so system at a correct price range and proper availability on the market of subcomponents. Technical requirements are anyway often not fully met by today’s monitoring system installed base. Some common mistakes: • The necessity of a close-to-zero phase error is completely failed if the accelerometers run via wi-fi network, due to network latency that cannot be avoided. No OMA can be run if phase shift error is relevant, and this can be the case with wifi sensors. • Accelerometer sensor with higher noise floor can measure correctly a modeshape in case of EMA or even giving sufficient data to identify modes in OMA if the structure is sufficiently excited, but for SHM we are looking for differences , for instance in the MAC matrix, overtime. It is demonstrated that there is very strict correlation between noise floor of sensors and time of SHM parameters giving a warning signal -such warnings be identified much earlier with low noise floor sensors; • Quite often the acceleration signals appear dirtied with temperature variation influence, rising the complexity to fix proper trigger level and often crossing them without a real issue or -even worse- not triggering when an acceleration peak occurs; more in general, temperature influence not fully controlled put analysis at risk of misinterpretation of data; • A close software suite cannot be fully exploited when the numbers of artifact monitored is growing -even worse, the characteristic of some platforms to hold the propriety of the data inside a closed suite hampers the possibility to broaden the knowledge outside the boundaries of the monitoring software, limiting it as a prerogative of few experts and not giving resonance at enterprise level. 1.4. MEMS Capacitive technology One of the main technologies for acquiring acceleration signals from a structure, as per 2022, is the MEMS capacitive. While piezoelectric technology still has dominance in some areas with his high dynamic capabilities, MEMS technology appears very promising in this context as can more easily catch low-frequency phenomena, having simpler network connection, in correct configuration showing good noise floor level. Variable capacitance accelerometers utilize a micro electro-mechanical system (MEMS) that is fabricated from silicon using a bulk micro machining process. Within the structure is a small proof mass suspended by a beam acting as a spring that deflects as a function of applied acceleration (Figure 1). The deflection causes a capacitance change between the proof mass and two electrodes that are fixed on either side separated by a small air gap.