PSI - Issue 64

Luca Schenato et al. / Procedia Structural Integrity 64 (2024) 1636–1641 Luca Schenato / Structural Integrity Procedia 00 (2019) 000 – 000

1637

2

1. Introduction Optical Frequency Domain Reflectometry (OFDR) is a sensing technique based on Rayleigh backscattering. It relies on the coherent measurement, taken from one end of the fiber, of the permanent arrangement of scattering centers randomly distributed along the fiber. This measurement reveals the so-called fiber signature, representing a unique fiber fingerprint permanently frozen during production. The fingerprint is paramount because when the fiber experiences a perturbation, such as a local temperature or strain change, it results in a quantifiable shift in its signature or, more precisely, in its spectrum. Commercial platforms, e.g., LUNA OBR 4600 ® , exploit this feature to achieve distributed strain and temperature measurements with high spatial resolution. Practically, the procedure consists of taking a first reference measurement of the Rayleigh fingerprint to be compared with the following measurement using proprietary software, which can cross-correlate the spectra of the two aligned traces. All the following traces are processed to determine the variation of strain or temperature with respect to the reference trace. In doing so, these measurements must be collected using the same patch cords from the interrogator to the sensing fiber initially used for the reference trace. Moreover, the same interrogator must be used. The reference is lost whenever the interrogator or different patch cords are replaced, and a new intermediate reference trace must be collected. This also means that if, for any reason, the intermediate reference trace cannot be referred to the original one, the continuity of the monitoring gets irremediably lost. This issue severely limits long-term measurement campaigns: hardware failure or damage to the optical sensing network over the years hampers stakeholders' capability to track the evolution of the phenomena under investigation reliably. Nonetheless, this paper shows that this limitation can be overcome. In fact, our measurements confirm that the fiber's Rayleigh signature is persistent despite being collected with different setups and interrogators, regardless of the time elapsed since the reference was taken. Of course, a proper spectral correlation data analysis procedure, not available when relying on standard approaches implemented by commercial interrogators, is mandatory in the case of a change in the optical setup (i.e., interrogator and patch cord). This work presents a selection of measurements taken from fibers operating in challenging conditions and collected over a long period. Due to the failure of the original interrogator, the last measurements were collected by a new device. This fact, along with the change of the optical patchcord between the interrogator and the sensing cables, would have normally impeded the use of the original reference trace, with a consequent interruption of the continuity in the monitoring data. However, by correctly applying the correlation analysis, we show that the Rayleigh fingerprint, collected in different operative conditions, is persistent and can still correlate with the one collected years before, with high quality, sufficient to track the monitored phenomena' evolution accurately. 2. Methods The basic configuration of an OFDR system involves using an interferometer and swept-frequency light to generate beating signals. The beating signal encodes the so-called roundtrip impulse response of the fiber under test (FUT). For the interested reader, Palmieri et al. (2022) describe the optical setup needed to measure the beating signal and, from this, extract the roundtrip impulse response describing the local Rayleigh reflectivity of the fiber. It is essential to clarify that commercial systems, like the one employed in this study, provide this roundtrip impulse response (“reflectivity”) as an output signal. Therefore, in the following, we will refer to this signal as the raw optical signal. Typically, this signal is expressed in terms of dB/m as a position along the FUT. Still, the position is half of the roundtrip propagation time multiplied by the speed of light in the fiber. The standard procedure for analyzing the raw signal for distributed sensing purposes has been known since 1998 (Froggat and More (1998)). Here, it will be only briefly summarized. First, a reference measurement of the FUT's round-trip impulse response is taken: this measurement is intended to take a “snapshot” of the FUT under unperturbed conditions. This measurement is used as a reference to determine the change the FUT will undergo in terms of strain or temperature in the subsequent measurement. Then, repeated measurements are taken over time, with the FUT possibly at different temperatures or strain fields. As highlighted above, all these measurements should be normally taken with the same interrogator and optical setup.