Brillouin Optical Frequency-domain Analysis Research Articles

A collection of machine learning assisted distributed fiber optic sensors for infrastructure monitoring

Abstract In this paper, we present a collection of machine learning assisted distributed fiber optic sensors (DFOS) for applications in the field of infrastructure monitoring. We employ advanced signal processing based on artificial neural networks (ANNs) to enhance the performance of the dynamic DFOS for strain and vibration sensing. Specifically, ANNs in comparison to conventional and computationally expensive correlation and linearization algorithms, deliver lower strain errors and speed up the signal processing allowing real time strain monitoring. Furthermore, convolutional neural networks (CNNs) are used to denoise the dynamic DFOS signal and enable useable sensing lengths of up to 100 km. Applications of the machine learning assisted dynamic DFOS in road traffic and railway infrastructure monitoring are demonstrated. In the field of static DFOS, machine learning is applied to the well-known Brillouin optical frequency domain analysis (BOFDA) system. Specifically, CNN are shown to be very tolerant against noisy spectra and contribute towards significantly shorter measurement times. Furthermore, different machine learning algorithms (linear and polynomial regression, decision trees, ANNs) are applied to solve the well-known problem of cross-sensitivity in cases when temperature and humidity are measured simultaneously. The presented machine learning assisted DFOS can potentially contribute towards enhanced, cost effective and reliable monitoring of infrastructures.

Multi-taper array for dynamic strain measurements by a Brillouin optical frequency-domain analysis

In this manuscript, we demonstrate that an array of tapers, consecutively formed along a single-mode optical fiber, can be used to realize a localized, dynamic strain sensing position with a Brillouin Optical Frequency-Domain Analysis (BOFDA) sensor. The multi-taper realizes a periodic modulation of the Brillouin frequency shift, which appears as a spectral resonance in the baseband transfer function acquired by the BOFDA sensor. Any strain acting along the multi-taper affects the magnitude of this resonance, offering a means to perform dynamic strain measurements in that position. Fabricating multi-taper arrays with a different period, more dynamic sensing positions can be realized along the same fiber. Measurements on two vibrating cantilevers were performed at a sampling rate of 46 Hz, using two multi-taper arrays with a length of 72 mm and 75 mm, respectively.

An Integrated Intelligent Approach for Monitoring and Management of a Deep Foundation Pit in a Subway Station.

As the scale of foundation pit projects of subway stations in Shenzhen becomes larger, and the construction constraints become more and more complex, there is an urgent need for intelligent monitoring and safety management of foundation pits. In this study, an integrated intelligent approach for monitoring and management of a deep foundation pit in a subway station was proposed and a case study based on the Waterlands Resort East Station Project of Shenzhen Metro Line 12 was used for validation. The present study first proposed the path of intelligent foundation pit engineering. Based on geotechnical survey and building information modeling, a three-dimensional transparent geological model of foundation pit was constructed. Multi-source sensing technologies were integrated, including micro electromechanical system sensing technology, Brillouin optical frequency domain analysis sensing technology, an unmanned aerial vehicle and machine vision for real-time high-precision wireless monitoring of the foundation pit. Moreover, machine learning models were developed for predicting key parameters of foundation pits. Finally, a digital twin integrated platform was developed for the management of the subway foundation pit in both construction and maintenance phases. This typical case study is expected to improve the construction, maintenance and management level of foundation pits in subway stations.

Deformation monitoring and remote analysis of ultra-deep underground space excavation

The development and utilization of urban underground space is producing many deep foundation pits. On-site monitoring is a key issue to ensure the safety of the foundation excavation, especially the monitoring of the lateral displacement of the structure. The ultra-deep circular foundation excavation project (58.65 m deep, the deepest foundation pit in Shanghai) in the test section of the Suzhou River deep tunnel is one of the most challenging engineering around the world. The distributed optical fiber monitoring technology of Brillouin optical frequency domain analysis (BOFDA) was employed to monitor the lateral displacement of the deep foundation. The monitoring process lasted for 14 months, and a total of 19 times of monitoring was carried out following the progress of the excavation of the foundation pit. This paper introduces the process, methods and precautions of optical fiber monitoring from early installation, equipment debugging to data acquisition, data processing and data analysis. By comparing the fiber monitoring data with the manual monitoring data, the deformation of the deep foundation excavation is analysed. The analysis results not only obtain the deformation law of ultra-deep foundation excavation, but also prove the advantages of optical fiber monitoring in deep foundation excavation. Finally, with the support of the Smart Infrastructure Service System (iS3), the research combines optical fiber monitoring with an information platform. Under the framework of the iS3 platform, an integrated data processing and analysis module for fiber monitoring is developed, and the feasibility of remote real-time collection and analysis of monitoring data is discussed. With the support of the information platform, the monitoring data can be transmitted back to the computer terminal to realize remote monitoring.

Comprehensive research on the failure evolution of the floor in upper mining of deep and thick coal seam

The Group A coal seam represents one of the most important coal deposits of the Huainan coalfield. It poses a series of dynamic geohazards during mining operations due to its thickness, stratification, and depth location. One of the major issues is the susceptibility of the coal floor to failure under the combined action of mining and pressurized water. At greater depths, the stress becomes concentrated at the tip of the primary fractures under the influence of the high surrounding pressure. The exploitation of coal has caused new fissures to develop along the primary fissures, with energy regrouping occurring at the tips of the new fissures. To better understand the rock deformation and failure of the coal floor, uniaxial loading strain tests were performed on a standard rock specimen based on the Brillouin optical frequency domain analysis (BOFDA) to capture the deformation characteristics on the surface of the rock specimen with high accuracy. During loading, the strain of the fiber optic at peak strength was equal to 4281 με. To further study the evolution of the failure of the coal floor rock stratum during mining, electrodes, and distributed fiber optic were placed in the coal floor monitoring borehole of the coal seam to record the changes in the current and strain. The monitoring results indicate that during the initial stage, when the working face was close to the monitoring borehole, the abutment pressure of the surrounding rock mass gradually increased, the strain increased, and the current value steadily rose. During the middle stage of the monitoring process, the working face was located above the monitoring hole, the upper part of the hole was under compressive strain, the current value increased, the bottom of the hole was under tensile strain, and the electrode current showed fluctuating values. At the late stage of the monitoring process, the working face was located in proximity to the monitoring hole and the mudstone layer. The current value dropped below 1 mA, and the peak tensile strain reached 8589 με, which was followed by an abrupt change, showing the characteristics of rock failure. Comprehensive diagnosis indicated that the maximum floor failure depth was at the 21.0 m mark. During mining, the electrical current and the strain values had a positive correlation when the rock at the bottom was not failure. However, these values suddenly changed after the rock failed. The obtained results can serve as guidance for borehole multi-field monitoring in mining engineering.

Multi-physical field joint monitoring of buried gas pipeline leakage based on BOFDA

Determining the leakage location in an underground gas pipeline is a serious technical problem. Distributed fiber optic testing technology based on Brillouin optical frequency domain analysis (BOFDA) can be used for monitoring the change in the strain and temperature of the surrounding media caused by the gas leakage. In this study, a point source leakage model was built. The distributed fiber optic strain and actively heated fiber optic were installed around the pipeline to monitor the gas pipeline leakage. Variations in the strain and temperature characteristics of the pipeline under different gas pressure conditions were found to be in the range of −392–402 μϵ and 3.5 °C during the 1.0 MPa gas leakage, respectively. When the soil model was loaded with a gas leakage pressure of 0.3–1.0 MPa, a significant variation in strain and temperature was noticed. The copper mesh actively heated fiber optic can effectively identify the location of the leak. Meanwhile, the heating rate of the actively heated fiber optic cables in different media environments were obtained and ranked from fast to slow: air, followed by general soil, followed by leakage soil. This led to the generation of new technology to detect the leakage location of buried gas pipelines through actively heated optical cables.

Quasi-Distributed Refractive Index Sensing by Stimulated Brillouin Scattering in Tapered Optical Fibers

In this paper, we demonstrate that multiple tapers in optical fibers allow for quasi-distributed refractive index sensing via a high spatial resolution Brillouin Optical Frequency-Domain Analysis (BOFDA) configuration. We first characterize, theoretically and experimentally, the variation of the Brillouin frequency shift (BFS) with the diameter of the tapered fiber. Then, we characterize the dependence of the BFS from the outer refractive index in an optical fiber taper with a waist diameter of 10 µm. Finally, we show that more tapers can be realized along the same fiber, in order to provide multi-point refractive index sensing.

Performance monitoring of a curved shield tunnel during adjacent excavations using a fiber optic nervous sensing system

With the accelerated urbanization and population growth in the Yangtze River Delta of China, numerous metro tunnels are in operation or under construction, and adjacent deep excavation activities are frequently encountered. For curved shield tunnels, the impact of excavation-induced ground movements is more complicated due to their asymmetric shape. This paper presents the monitoring results of a curved shield tunnel in clayey soil in Suzhou, China, which were captured by a fiber optic nervous sensing system. This system utilized Brillouin optical frequency domain analysis technology to monitor the distribution of longitudinal and circumferential strains of tunnel linings induced by adjacent excavation. The results show that the tunnel linings were mainly subjected to bending deformations along the tunnel alignment. Maximum compressive strains were observed below the tunnel springline, and their absolute values were higher than those of maximum tensile strains measured at the tunnel crown, distorting the circular tunnel into a rotated oval/ellipsoid. Based on the monitoring results, two kinematic models reflecting the spatial relationship between movements of tunnel segments and measurements of strain sensing cables are proposed. Furthermore, a mechanical method is proposed to convert strain measurements into radial displacements of tunnel linings and the structural health condition of the tunnel is evaluated using longitudinal and circumferential risk indexes. The conclusions drawn in this study provide improved insight into the deformation pattern and health condition of curved shield tunnels subjected to adjacent excavations.

Distributed Liquid Level Sensor Based on Brillouin Optical Frequency-Domain Analysis

We propose and demonstrate a distributed liquid level sensor based on high-resolution Brillouin Optical Frequency-Domain Analysis and an actively heated cobalt-doped optical fiber. The liquid level position is determined by localizing the abrupt temperature change occurring along the active fiber at the liquid/air interface, through distributed temperature sensing. Differently from previous works based on the use of optically heated fibers, in the proposed configuration no separate laser source was employed to heat the fiber. In practice, a single laser source was adopted for both optical heating and distributed temperature sensing. The experiments report the measurement of the liquid level position with 5 mm level resolution along a 9-cm long heated fiber. The paper also discusses the limitations of the proposed method. In particular, the negative impact of spontaneous Brillouin scattering, in terms of measurement noise, is revealed by the measurements. We show that the spontaneous Brillouin scattering induces spurious components in the frequency-domain, affecting the accuracy of temperature measurements.

Distributed humidity fiber-optic sensor based on BOFDA using a simple machine learning approach.

We report, to our knowledge for the first time, on distributed relative humidity sensing in silica polyimide-coated optical fibers using Brillouin optical frequency domain analysis (BOFDA). Linear regression, which is a simple and well-interpretable algorithm in machine learning and statistics, is utilized. The algorithm is trained using as features the Brillouin frequency shifts and linewidths of the fiber's multipeak Brillouin spectrum. To assess and improve the effectiveness of the regression algorithm, we make use of machine learning concepts to estimate the model's uncertainties and select the features that contribute most to the model's performance. In addition to relative humidity, the model is also able to simultaneously provide distributed temperature information addressing the well-known cross-sensitivity effects.

Deformation monitoring of ultra-deep foundation excavation using distributed fiber optic sensors

The development and utilization of urban underground space is producing a large number of deep foundation pits. On-site monitoring is a key issue to ensure the safety of the foundation excavation, especially the monitoring of the horizontal displacement of the structure. The rapid development of distributed fiber optic sensors technology in recent years could avoid the deficiencies of manual monitoring methods, which can quickly, real-time and continuously monitor the horizontal displacement of deep soils. At the same time, with the support of the information platform, the monitoring data can be transmitted back to the computer terminal to realize remote monitoring. This paper relies on the ultra-deep circular foundation excavation project (58.65m deep, the deepest foundation pit in Shanghai) in the test section of the Suzhou River deep tunnel, and uses the distributed optical fiber monitoring technology of Brillouin optical frequency domain analysis (BOFDA) to monitor the horizontal displacement of the deep foundation. This paper introduces the process, methods and precautions of optical fiber monitoring from early installation, equipment debugging to data acquisition, data processing and data analysis. The analysis results not only obtained the deformation law of ultra-deep foundation excavation, but also proved the advantages of optical fiber monitoring in deep foundation excavation. Finally, the data has been uploaded to the digital platform for remote monitoring in the future.

Time-Efficient Convolutional Neural Network-Assisted Brillouin Optical Frequency Domain Analysis.

To our knowledge, this is the first report on a machine-learning-assisted Brillouin optical frequency domain analysis (BOFDA) for time-efficient temperature measurements. We propose a convolutional neural network (CNN)-based signal post-processing method that, compared to the conventional Lorentzian curve fitting approach, facilitates temperature extraction. Due to its robustness against noise, it can enhance the performance of the system. The CNN-assisted BOFDA is expected to shorten the measurement time by more than nine times and open the way for applications, where faster monitoring is essential.

Performance monitoring of offshore PHC pipe pile using BOFDA-based distributed fiber optic sensing system

Brillouin Optical Frequency Domain Analysis (BOFDA) is a distributed fiber optic sensing (DFOS) technique that has unique advantages for performance monitoring of piles. However, the complicated production process and harsh operating environment of offshore PHC pipe piles make it difficult to apply this method to pile load testing. In this study, sensing cables were successfully pre-installed into an offshore PHC pipe pile directly for the first time and the BOFDA technique was used for in-situ monitoring of the pile under axial load. High-resolution strain and internal force distributions along the pile were obtained by the BOFDA sensing system. A finite element analysis incorporating the Degradation and Hardening Hyperbolic Model (DHHM) was carried out to evaluate and predict the performance of the pile, which provides an improved insight into the offshore pile-soil interaction mechanism.

Monitoring Method of Longitudinal Land Subsidence and Deformation in Seismic Geological Disasters

Monitoring of longitudinal land subsidence and deformation in seismic and geological hazards plays an important role in preventing and curing land collapse, land subsidence, and ground cracks. In this paper, a distributed monitoring model experiment on vertical land subsidence and deformation of seismic and geological hazards is carried out by Brillouin optical frequency-domain analysis technology (BOTDA). By using the self-made indoor longitudinal land subsidence and deformation simulation box, different intensity seismic ground is simulated by air bag. Distributed optical fibers are used to monitor the longitudinal land subsidence and deformation during different intensity seismic and geological disasters. According to different intensity seismic and geological disasters, distributed sensing optical fibers cooperate with the ground to compress or stretch longitudinally and obtain the data of longitudinal land subsidence and deformation. The correction coefficient is introduced to modify the monitoring data of confining pressure-sensing optical fiber and complete the precise monitoring of vertical land subsidence and deformation in seismic and geological hazards. The experimental results show that this method can monitor the vertical ground subsidence and deformation of seismic and geological hazards under different conditions, and the monitoring efficiency and cost are superior to GPS and inertial monitoring methods, and the practical application value is high.

A Dual-Wavelength Scheme for Brillouin Temperature Sensing in Optically Heated Co2+-Doped Fibers

In this paper, we propose the use of a dual wavelength scheme for active distributed temperature sensing based on Co2+-doped fibers. A Brillouin Optical Frequency-Domain Analysis (BOFDA) set-up, operating at 850 nm, is used to detect the temperature changes along Co2+-doped optical fibers with various absorption coefficients, while optical heating is performed through a 1550-nm optical pump. A simple model describing optical heating along the Co2+-doped fiber is introduced and experimentally validated. As a proof of concept, we demonstrate that the sensor is capable to distinguish a 1-m portion of fiber immersed into water, from a section of equal length surrounded by air.

Simulation of Distributed Current Sensor Based on Optical Fiber Deformation

Due to the development of automation and control systems, methods and devices for measuring of electric current large values are of great interest. The aim of the work was to develop a schematic diagram of a distributed current strength sensor based on the Brillouin optical frequency domain analysis; to create a mathematical model of the sensor to demonstrate its work and to calculate the basic parameters of the sensor. To provide the measurement optical fiber with conductive coating is used. Between the current bus, where current is measured, and conductive coating the Ampere force arises. Strain occurs in optical fiber due to this force. Stimulated Brillouin scattering has the strain dependent characteristic frequency. Shift of the characteristic frequency allows to measure current in the bus. To measure the characteristic frequency and the location of its shift Brillouin optical frequency domain analysis is used.The mathematical model of sensor operation based on tree-wave model of stimulated Brillouin scattering is demonstrated. This model allows calculating intensity of optical signal in the fiber in dependence of characteristic frequency shift. Brillouin optical frequency domain analysis uses inverse Fourier transform to obtain pulse response.A schematic diagram of a distributed current sensor based on the method of Brillouin optical frequency domain analysis is presented. An a priori estimate of parameters of the measuring system was carried out on the basis of the mathematical model of stimulated Brillouing scattering in an optical fiber. The spatial resolution of the sensor when determining the length and location of fiber sections was 0.06 m. The resolution of the sensor was 0.22 kA, the maximum value of the current strength was 25 kA. Dependence of the sensor operation at different powers of the laser used was investigated. The refractive index change influence on the result of measurements was estimated.

Monitoring of vertical deformation response to water draining–recharging conditions using BOFDA-based distributed optical fiber sensors

Land subsidence is a geological disaster, which has a variety of inducements. This paper presents the results on the investigation of the soil deformation in land subsidence with a combination of laboratory and field experiments using the distributed fiber optics sensing (DFOS) technology. The laboratory experiment was mainly carried out to study the soil deformation in draining–recharging cycles based on Brillouin optical frequency domain analysis (BOFDA). In the field experiment, the BOFDA technology and some grating sensors were applied in a borehole in the Yangtze River Delta region, to monitor the pore pressure in certain specific strata and the change in strata parameter of the full borehole with the data that monitored 2 years. The result of the laboratory experiment demonstrated that the soil structures changed in the draining–recharging cycles. The soil gradually tends toward elastic deformation. The field experiment results show that the main deformation occurs at the aquitards adjacent to the pumping aquifer. The compression amount in the vertical distribution is not uniform which is related to the distance from the pumping aquifer and the soil properties. The compressive deformation of each layer is closely related to the change in groundwater level. The result proposes that the BOFDA technology is suitable for monitoring land subsidence caused by seasonal water-level fluctuations and dewatering projects. It is also of great significance to investigate land subsidence mechanisms.

A 100-km BOFDA Assisted by First-Order Bi-Directional Raman Amplification.

We present, to our knowledge for the first time, a 100-km Brillouin Optical Frequency-Domain Analysis (BOFDA) employing a 200-km fiber loop. Compared to our previous publication, enhanced sensor length, sensor accuracy and spatial resolution are presented. The performance improvements are achieved by applying distributed Raman amplification (DRA) and a digital high-pass filter. We report on temperature measurements over sensing distances of 75 km and 100 km both with a 12.5-m spatial resolution. Temperature changes of 5 C have been measured along 75 km sensing fiber. A temperature change of 30 C has been detected at 99.5 km.

63 km BOFDA for Temperature and Strain Monitoring.

We demonstrate (and are the first to do so) 63 km Brillouin Optical Frequency-Domain Analysis (BOFDA) for temperature and strain monitoring using a 100 km fiber loop. The use of BOFDA for long-range applications can be considered a novel approach, as previous investigations focused on the utilization of Brillouin Optical Time-Domain Reflectometry and Analysis (BOTDR and BOTDA, respectively). At 51.7 km, a 100 m hotspot (37 C) was detected without using distributed Raman amplification or image processing.

High-Pass Filtering for Accurate Reconstruction of the Brillouin Frequency Shift Profile From Brillouin Optical Frequency Domain Analysis Data

In this paper, we propose and demonstrate a new method to reconstruct the Brillouin frequency shift profile in high spatial resolution Brillouin optical frequency domain analysis (BOFDA) sensors. The method aims to compensate the distorting terms affecting BOFDA measurements, which originate from the modulation impressed on the acoustic wave involved in the scattering phenomenon. We show that these terms can be easily removed by applying a high-pass filter to the data acquired in the frequency-domain, at the cost of a degradation of the signal-to-noise ratio. A numerical analysis, as well as experimental tests carried out at 8-mm spatial resolution, validate the proposed technique.