PSI - Issue 72

V.P. Matveenko et al. / Procedia Structural Integrity 72 (2025) 229–234

231

Among the key parameters influencing the measurement process, the gauge length (Δx) plays a crucial role. It defines the portion of the optical fiber over which a single measurement point is determined. In contrast to point FOS based on fiber Bragg gratings (FBG), where the gauge length is physically fixed, the gauge length in DFOS based on Rayleigh scattering could be adjusted programmatically. Choosing the optimal gauge length is essential for ensuring accurate strain measurement, particularly in cases of gradient strain distribution. In the provided study the value of gauge length was set to 5 mm to match the length of previously studied FBG sensors. Like all sensors, strain measurement using DFOS based on Rayleigh scattering is subject to noise in the reflectometer-FOS system. However, unlike point sensors, where measurement noise is observed for consecutive measurements over time, distributed sensors introduce an additional spatial dimension — the length of the measurement section of the optical fiber — resulting in variations in readings both over time and along the fiber length. This study provides a statistical analysis of noise level in DFOS based on Rayleigh scattering readings. Assessing the dispersion of DFOS readings under conditions free from external mechanical loads and ambient temperature variations is crucial for determining the sensor's sensitivity. Such analysis establishes the minimum strain level that can be reliably detected, which is essential for applications requiring high precision. The conducted experiment consisted of recording DFOS measurements under isothermal conditions, ensuring no mechanical loading on the measurement section of the optical fiber for a predetermined time interval (approximately 10 hours). Strain measurements were performed using Luna Innovations OBR 4600 reflectometer. A standard single-mode optical fiber with a polyimide protective coating was utilized in the experiment. The wavelength scan range of the reflectometer tunable laser source for this study was set to 40 nm. The analyzed signal, which represents strain variations along the optical fiber section over time, is obtained from the spectral shift  between the current and reference Rayleigh scattering configurations using the following relation: T K K T      , (1) where T K and K  are the temperature and strain coefficients. For most silica glass optical fibers 6 1 6.45 10 С T K     , 0.78 K   . For isothermal conditions, the equation takes the form:

cK     ,

(2)

where  is the spectral shift;  is the central wavelength of the scan; c is the speed of light. 3. Results and discussion

The strain measurement results over a 10-hour period, recorded at a frequency of once per minute, are presented in Fig. 2. The collected data is visualized through a set of graphs. At the center, a two-dimensional heat map illustrates the strain distribution along the sample's length (x-axis) over time (y-axis). The color scale corresponds to the strain level. In this graph, two dotted lines mark the slices of strain distribution over time (red dotted line) and length of the sample (blue dotted line), the graphs of which are located to the right and bottom of the central graph, respectively. To evaluate the noise characteristics in FOS measurements, the standard deviation (σ) and the amplitude range ( R ) of each measurement signal were calculated:

1

N

1   1 i 

2 ) ,

(

,

x

R x

x





max  

(3)

i

min

N