Characterization of Fiber Bragg Grating Accelerometer for Underwater Vibration Detection

: In this paper, the results of the characterization of a Fiber Bragg Grating (FBG) Accelerometer for detecting vibrations underwater are reported. The FBG Accelerometer, consisting of three FBGs, is utilized to detect underwater vibrations in three-dimensional directions. A water pump, with positions varied from 0 to 10 cm, is employed as the vibration source. Furthermore, the experimental results are presented in the form of the peak wavelength shift reflected by the FBG (Δλ B ) and frequency. From the experiment results, it is shown that with increasing distance, Δλ B decreases linearly with successive gradients of 0.0058 nm/cm; 0.0059 nm/cm; and 0.0045 nm/cm. for FBG X, Y, and Z. It is also shown that with increasing distance, there is a decrease in frequency from 50 Hz for FBGs X, Y, and Z to 39 Hz; 38 Hz; and 40 Hz for FBGs X, Y, and Z respectively.


Introduction
Vibration detection plays a crucial role in observing and analyzing seismic activities on the Earth's crust (Hou et al., 2023) and has various applications such as detecting vibrations on the ocean surface (Lan et al., 2022) and identifying potential tsunamis and earthquakes (Duggal et al., 2022).
In Indonesia, natural events such as earthquakes occur quite frequently on several islands, including Jawa, (Supendi et al., 2022), Sulawesi (Jena et al., 2020), Kalimantan, Maluku (Sianipar et al., 2022), and other smaller islands (Ramdani et al., 2019).This is due to Indonesia's location surrounded by three tectonic plates: the Eurasian Plate, the Indo-Australian Plate, and the Pacific Plate (Wibowo et al., 2023).Various methods have been developed to create sensor systems that detect underwater earthquakes and swiftly transmit information to land (Z.Li, 2021).One of the developments implemented is Indonesia Tsunami Early Warning System (Ina-TEWS) in the form of buoys.Buoy is a floating device on the sea surface that records changes in water level (Martínez-Osuna et al., 2021).Although this system is relatively new, it has its weaknesses, namely the high cost of procuring and maintaining buoys, and its inability to provide information regarding the scale of the impending tsunami (Made, 2018).
Therefore, due to these issues, FBG can become an alternative sensor to detect underwater earthquake vibrations.Currently, Badan Riset dan Inovasi Nasional (BRIN) is developing an FBG sensor system called the Indonesia Cable-Based Tsunameter (Ina-CBT), which can be integrated into the existing Ina-TEWS system.The FBGs are installed on the seabed with lengths reaching hundreds of kilometers and connected to a base station located on the coast.FBG has a very long operational lifespan because it does not rely on batteries and can transmit data in larger quantities and shorter timeframes.Research on the utilization of FBG in water has been conducted, which examines the effects of temperature on the Bragg wavelength shift (Kustianto et al., 2023).
As a continuation of that research, this study characterizes an FBG accelerometer for detecting underwater vibrations as a basis for developing underwater sensors.The FBG accelerometer consists of three FBGs that will detect vibrations in threedimensional directions: X, Y, and Z. Subsequently, an analysis of the changes in Bragg wavelength and frequency for each FBG is conducted.

Method
The FBG sensor operates based on the principle of interference that occurs between the transmitted (Yan & Liang, 2020) and reflected light (L.Zhang et al., 2022) by the grating structure within the optical fiber (L.Wang et al., 2022).
This results in a specific wavelength being reflected by the FBG (H.Wang et al., 2021) while other wavelengths pass through the grating.When light passes through the grating structure, a specific wavelength, known as the Bragg wavelength, will be reflected back (Qian et al., 2021).Bragg wavelength (λB) can be determined as Formula 1: where neff is the effective core index of refraction (Le et al., 2022), and Λ is the spatial period of the grating (Shen et al., 2023).When the FBG is subjected to disturbances such as vibration, pressure, or temperature, the value of λB changes correspondingly (Nguyen et al., 2023).The change in Bragg wavelength (∆  ) can be described as Formula 2: where ∆ is the longitudinal strain variation (J.Li et al., 2022), ∆ is the temperature increment,  is the elastooptical coefficient of the fiber (Dong et al., 2023),  is the the coefficient of thermos-optic (H.Zhang et al., 2018), dan  is the coefficient of thermal expansion of the fiber (Pan et al., 2019).When the FBG experiences vibration, a change in strain occurs within the FBG.This change in strain affects λB, resulting in changes of ∆  (T.Yang et al., 2021).
Figure 1 shows the FBG Accelerometer consisting of 3 FBGs: FBG X, FBG Y, and FBG Z. FBG X has λB of 1544 nm, FBG Y has λB of 1529 nm, and FBG Z has λB of 1540 nm. Figure 2 shows the experimental setup for vibration detection measurements.In this setup, the FBG is connected to the optical interrogator (3S-BS01).The function of the interrogator is to send light to the FBG and demodulate the optical signal reflected back from the FBG (Guo et al., 2023).After the demodulation process is completed, the information is displayed on a laptop (Y.Li et al., 2022).
A regulated DC power supply (PAN 35-20A) provides 5 V and 0.3A to the interrogator processing system for optical signal demodulation.For the vibration source, a 60 W water pump (PW-105) is used to generate vibrations underwater (J.Li et al., 2023).A container box with a capacity of 120 liters serves as a small pool where the FBG accelerometer is placed for underwater experiments.The water level in the container box reaches 17 cm.This height is sufficient to submerge both the FBG and the water motor.For each measurement, different distance values are used for each FBG, starting from 0, 2, 4, 6, 8, and 10 cm (Le et al., 2023).Figures 4 to 6 show an experimental setup measuring vibrations from the vibration source with FBG X, Y, and Z at the distance of 10 cm.The software used to display the demodulation results is Ibsen Photonics I-MON 512 USB Evaluation Software.Figures 7 and 8 show two graphs: a spectrum graph and a wavelength graph.During the experiment, the spectrum graph needs to be kept constant at all times to ensure that the results of each measurement remain consistent and accurate (X.Yang et al., 2020).Wavelength graph shows λB of FBG X, Y, and Z.

Result and Discussion
For each measurement, data was collected for 1 minute, obtaining 600 data points with a sampling time of 0.1 seconds for FBG X, Y, and Z.When FBG X is stationary and not subjected to any vibration disturbances, λB of FBG X is 1544.1399nm, λB of FBG Y is 1529.1390nm, and λB FBG Z is 1540.0208nm.
When water pump is running, vibrations are generated and causing changes to λB of FBG X, Y, dan Z. Figure 9 shows the changes in λB values of FBG X, Y and Z at the distance of 0, 2, 4, 6, 8, and 10 cm.When the water pump is at 0 cm from FBG, λB of FBG X is 1544.2520nm, λB of FBG Y is 1529.5455nm, and λB of FBG Z is 1529.5455nm.When water pump is 10 cm from FBG, all three values of λB decrease.The λB of FBG X is 1544.1878nm, λB of FBG Y is 1529.4825nm, and λB of FBG Z is 1540.0658nm.The measurement results indicate that as the distance increases, the changes in λB for FBG X, Y, and Z decreases.
For frequency measurements, the FFT function of the I-MON software was used to gather frequency data received by each FBG from distances of 0, 2, 4, 6, 8, and 10 cm. Figure 10 shows the changes in distance against the frequency values received by FBG X, Y, and Z.At the distance of 0 cm, frequency received by FBG X, Y, and Z is 50 Hz.At the distance of 10 cm, frequency received by FBG X is 39 Hz, FBG Y is 38 Hz, and FBG Z is 40 Hz.The measurement results indicate that as the distance increases, the frequency values received by FBG X, Y, and Z decrease.A linear regressions analysis is conducted on the change in λB or ΔλB of FBG X, Y, and Z with respect to the distance.Figure 11 shows a graph of ΔλB for FBG X.From linear regression analysis, there's gradient of decline with a value of 0.0058 nm/cm and R-squared value of 92.31%.Figure 12 shows a graph of ΔλB for FBG Y. From linear regression analysis, there's gradient of decline with a value of 0.0059 nm/cm and R-squared value of 96.47%.Figure 13 shows a graph of ΔλB for FBG Z. From linear regression analysis, there's gradient of decline with a value of 0.0045 nm/cm and R-squared value of 98.82%.(Feng et al., 2015;Parida et al., 2019;X. Wang et al., 2018;Teng et al., 2024Xu et al., 2024).These experiments used one or two FBGs and were conducted on air with shaking table to simulate vibrations produced by a source.The novelty of this specific experiment are 3 FBGs were used and conducted underwater with a water pump as vibration source that was located underwater.The results show that FBG Accelerometer can be utilized to detect a vibration source located underwater and not limited to air only.

Conclusion
In this paper, FBG Accelerometer characterization has been conducted to detect vibrations underwater as a foundation for developing underwater vibration sensors.The experimental results show that the changes in the peak reflected wavelength (λB) of FBG X, Y, and Z decrease with increasing distance.Meanwhile, the experimental results also indicate that as the distance increases, the frequency received by the FBGs decreases as well.Based on linear regressions analysis, ΔλB of FBG X decreases linearly with gradient of 0.0058 nm/cm and R-squared value of 92.31%.While ΔλB of FBG Y decreases linearly with the gradient of 0.0059 nm/cm and Rsquared value of 96.47%.Next, ΔλB of FBG Z decreases linearly with the gradient of 0.0045 nm/cm and Rsquared value of 98.82%.The results show that FBG Accelerometer can be utilized to detect a vibration source located underwater.

Figure 2 .
Figure 2. Schematic of experiment setup

Figure 9 .Figure 10 .Figure 11 .
Figure 9.The change in λB of FBG X, Y, and Z due to distance change from 0 to 10 cm

Figure 12 .Figure 13 .
Figure 12.The change in ΔλB of FBG Y due to distance change from 0 to 10 cm