Optical fiber fault positioning system and positioning method based on distributed optical fiber sensing technology

Abstract

The invention relates to the technical field of optical cable fault processing, in particular to an optical fiber fault positioning system and method based on a distributed optical fiber sensing technology, and the system comprises an optical subsystem, a circuit subsystem, a software subsystem, a mechanical subsystem and an auxiliary subsystem, the circuit subsystem is provided with an edge intelligent processing unit, the auxiliary subsystem has a self-calibration function, and the software subsystem comprises a digital twinning and fault prediction module. According to the positioning method, an RBS baseline curve is obtained by accessing an optical cable to be measured, external disturbance is applied to a pre-estimated fault point, and after photoelectric conversion and signal conditioning, the range of the fault point is accurately positioned by adopting an optical path difference principle, time-frequency conjoint analysis and a self-adaptive step approximation algorithm; and identifying a disturbance type by combining deep learning, and predicting a fault risk by an LSTM model. The positioning precision is superior to + / -1m, fault point geographical mapping and predictive maintenance can be realized, the problems of inaccurate positioning and high maintenance cost of the existing system are solved, and the optical cable repair efficiency and the network stability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical cable fault processing, and in particular to an optical fiber fault positioning system and method based on distributed optical fiber sensing technology. BACKGROUND

[0002] In the geographical positioning of optical cable faults, accurately positioning the optical cable fault is the most important work in the processing of optical cable faults. However, due to the fact that the actual laying of the optical cable does not match the theoretical geographical position (in the actual laying process, the optical cable cannot be laid completely according to the designed route, including bending or bending due to terrain, so that the specific geographical laying position is uncertain, especially in the later stage due to external factors, so that the specific geographical position cannot be determined), when the optical cable fails later, only the distance of the optical cable fault point from the origin (this origin is artificially set at any position of the actual optical cable) can be known, making it more difficult to determine the geographical position of the optical cable fault point, and bringing great difficulty to the real-time repair of the optical cable.

[0003] The existing positioning system of the optical cable fault point includes two systems: one is an optical cable actual length display system, and the other is an optical cable actual geographical position display system. The optical cable actual length display system adopts a signal device connected to the monitoring fiber core in the joint box, and according to the signal device and the monitoring fiber core, the actual length of the optical cable is displayed in sections, and the actual distance between each section is displayed. Each signal device serves as a node in the optical cable actual length system; the actual geographical position display system of the optical cable is provided with a signal device on each joint box of the optical cable line, and the signal device, the signal scanning device and the monitoring fiber core in the optical cable line constitute an optical cable fault point positioning system, and the GIS system is used to accurately display the fault point on the geographical position. The first system has the disadvantage that it can only display the length of the optical cable between the fault point and the measuring point, and cannot locate the actual position of the fault point; the other system has the disadvantage that the positioning system is provided with a signal device on each joint box of the optical cable line, which is prone to failure and has high maintenance cost. SUMMARY

[0004] In order to make up for the above shortcomings, the present application provides an optical fiber fault positioning system and method based on distributed optical fiber sensing technology, which can effectively and accurately locate the optical cable fault point, greatly reduce the difficulty of real-time repair of the optical cable, and solve the problem that the fault measurement distance of the optical time domain reflectometer does not correspond to the actual fault point position.

[0005] In order to achieve the above-mentioned target, the technical scheme adopted by the present application is as follows:

[0006] An optical fiber fault positioning system based on distributed optical fiber sensing technology, comprising an optical subsystem, a circuit subsystem, a software subsystem, a mechanical subsystem and an auxiliary subsystem.

[0007] The optical subsystem includes a pulse light synthesis and amplification module and a photoelectric conversion module; the pulse synthesis and amplification module is used to generate and emit an optical detection signal, and the photoelectric conversion module is used to realize mutual conversion between an optical signal and an electrical signal.

[0008] The circuit subsystem includes a system control module, a signal acquisition module and a signal conditioning module; the system control module is used to realize basic system operation of the device, the signal acquisition module is used to acquire an optical signal, and the signal conditioning module is used to realize modulation and demodulation of the optical signal.

[0009] The software subsystem includes a parameter control module, a data processing module and a data storage module; the parameter control module is used to realize adjustment and setting of parameters such as precision, range and sensitivity, the data processing module is used to realize processing of an electrical signal and a series of algorithm processing, and the data storage module is used to store data and images.

[0010] The mechanical subsystem includes a device integration module and an interface interconnection device; the device integration module is used to integrate various functional modules, and the interface interconnection device is used to realize interconnection between the device and a fiber core of an optical cable.

[0011] The auxiliary subsystem includes a system power supply device, a temperature control module and a man-machine interaction interface; the system power supply device is used to realize power supply of the device, the temperature control module is used to acquire temperature information of the device and adjust and control the temperature of the device, and the man-machine interaction interface is convenient for personnel operation.

[0012] The optical subsystem further integrates a polarization controller and a polarization diversity receiver, and synchronously acquires phase information, intensity information and polarization state information of an RBS signal, thereby realizing joint demodulation of multiple physical quantities.

[0013] The auxiliary subsystem further includes a built-in reference optical fiber ring or a virtual disturbance generator, which is used to periodically inject a standard disturbance signal to perform self-calibration and self-diagnosis on system sensitivity, positioning accuracy and hardware state.

[0014] The circuit subsystem integrates an edge intelligent processing unit, and the edge intelligent processing unit is configured with a neural network accelerator, which is used to complete disturbance event detection, feature extraction and preliminary positioning in real time locally, thereby reducing dependence on cloud computing.

[0015] A positioning method of a fiber fault positioning system based on a distributed optical fiber sensing technology, including the following steps:

[0016] S1: connecting a fiber core of a to-be-tested fault optical cable to a system interface, emitting a detection light pulse through a pulse light synthesis and amplification module, and acquiring an RBS baseline curve in a stable state.

[0017] S2: Apply external disturbance at the position W0 near the estimated fault point; the external disturbance includes one or more of artificial knocking, ultrasonic excitation, thermal pulse or electromagnetic excitation; the system pre-establishes an RBS response fingerprint library of different disturbance types, and intelligently identifies and verifies the disturbance type and position through a deep learning model; the type matching formula of the deep learning model is:

[0018]

[0019] T is the identified disturbance type, m is the total number of disturbance types in the fingerprint library, F is the feature vector of the current disturbance, is the standard feature vector of the kth type of disturbance in the fingerprint library, and ‖·‖2 is the Euclidean distance.

[0020] S3: Capture the RBS coherent fading waveform changes caused by the disturbance through the signal acquisition module, and calculate the phase disturbance position by the data processing module after photoelectric conversion and signal conditioning. The data processing module calculates the phase disturbance position based on the principle of optical path difference, and the core formula is:

[0021]

[0022] Where x is the phase disturbance position, c is the speed of light in vacuum, is the time difference of the probe light pulse to and fro the disturbance point, and n is the refractive index of the optical cable core.

[0023] The data processing module uses a time-frequency joint analysis method, including cross-correlation operation on the two disturbance signals, and combining wavelet packet decomposition to accurately locate the disturbance center position, with a positioning accuracy better than ±1 meter; the cross-correlation operation is as follows:

[0024]

[0025] Where, is the cross-correlation coefficient, x(t) is the baseline RBS signal, y(t) is the RBS signal after disturbance, and τ is the time delay; the positioning correction formula combined with wavelet packet decomposition is:

[0026]

[0027] Where, is the corrected positioning position, Ent(j, k) is the energy entropy of the jth layer and the kth wavelet packet node, is the maximum energy entropy of the layer, and δ is the correction coefficient.

[0028] S4: If a valid perturbation response is detected at W0, the next perturbation is applied at the far end W1=W0+ΔL; otherwise, the perturbation is applied at the near end W1=W0-ΔL, where ΔL is an adaptive step size. The adaptive step size ΔL is dynamically adjusted according to the current signal-to-noise ratio and perturbation response confidence, and the initial step size is 50-100 meters, which is gradually reduced to 0.5-2 meters during the approximation process; the adjustment formula is:

[0029]

[0030] wherein, is the initial step size; k1, k2 are adjustment coefficients, k1=0.1-0.3, k2=0.2-0.4; the step size is accelerated to reduce when the signal-to-noise ratio S / N≥10; the step size is stable at 0.5-2 meters when the confidence C≥0.9; the lower limit of the step size is constrained to ≥0.5 meters.

[0031] S5: When the two adjacent perturbation positions W i and W i+1 only one of which detects a valid response, it is determined that the optical cable fault point is located between W i and W i+1 .

[0032] Further, the software subsystem further comprises a digital twin mapping module, which integrates a geographic information system and an optical cable laying path model; when the positioning result falls into a high-risk area such as a joint box, a road crossing section, or a historical fault point, the system automatically preferentially guides the perturbation test and issues a warning prompt.

[0033] Further, the system continuously collects RBS baseline data and constructs a time series of optical fiber health status, models the micro-bending accumulation, stress fatigue, or environmental aging trend through a time series prediction model, realizes fault risk prediction, and supports predictive maintenance decision-making; the time series prediction model uses LSTM, and the core prediction formula is as follows:

[0034]

[0035] wherein, is the optical fiber health status value after τ time, ranging from 0 to 1, and the closer to 0, the higher the fault risk; is a Sigmoid activation function; is an output layer weight, is a cell state vector, is an output layer bias term.

[0036] The present application has the following beneficial effects:

[0037] High-precision positioning:

[0038] Through the principle of optical path difference, time-frequency joint analysis (including cross-correlation operation and wavelet packet decomposition), and adaptive step approximation algorithm, the fault point positioning accuracy is better than ±1 meter, and the step can be dynamically adjusted, the fault range can be quickly narrowed, the positioning efficiency and accuracy are significantly improved, and the problem of inaccurate positioning of the existing system is solved.

[0039] Intelligent fault identification and predictive maintenance:

[0040] The integrated deep learning model and LSTM time series prediction model can intelligently identify disturbance types and predict the health status of optical fibers based on historical data, realize early warning of fault risk, support predictive maintenance decision-making, reduce the probability of sudden failure, and improve network stability.

[0041] High reliability:

[0042] The auxiliary subsystem is built-in with a reference optical fiber ring or a virtual disturbance generator, which can periodically inject standard disturbance signals to realize self-calibration and self-diagnosis of system sensitivity, positioning accuracy and hardware state, ensure the reliability and stability of long-term operation of the system, and reduce maintenance cost.

[0043] Local processing and geographic information integration:

[0044] The circuit subsystem integrates an edge intelligent processing unit equipped with a neural network accelerator, which can complete disturbance event detection and feature extraction in real time locally, reducing dependence on cloud computing; the software subsystem combines a digital twin module with a geographic information system to realize geographic mapping of fault points and automatic early warning of high-risk areas, guiding on-site repair and improving optical cable maintenance efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 A system structure diagram of an optical fiber fault positioning system based on distributed optical fiber sensing technology is provided.

[0046] Figure 2 A flowchart of a positioning method of an optical fiber fault positioning system based on distributed optical fiber sensing technology is provided. DETAILED DESCRIPTION

[0047] The present application will be further described below in conjunction with specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0048] 1. Implementation scenario

[0049] A 10 km communication cable (model GYTA-48B1, core refractive index n = 1.468) in a certain urban area was used as the test object. The cable passes through 3 road crossing sections and 5 joint boxes. The historical fault points are concentrated in the 4.8-5.2 km interval, and it is necessary to quickly locate the sudden failure and evaluate the health status.

[0050] 2. System parameter setting

[0051] Optical subsystem: probe light pulse wavelength 1550 nm, polarization controller adjustment range 0-360°, polarization diversity receiver sampling rate 100 MHz;

[0052] Circuit subsystem: edge intelligent processing unit with neural network accelerator, signal acquisition module bandwidth 200 MHz, signal conditioning module gain 40 dB;

[0053] Software subsystem: RBS response fingerprint library contains 2 types of standard feature vectors, artificial knocking and ultrasonic excitation, LSTM model training batch 1000 times, digital twin module integrates local GIS map and optical cable laying path model;

[0054] Auxiliary subsystem: reference optical fiber ring periodically injects standard disturbance signal, temperature control range 0-45℃, human-computer interaction interface real-time display of positioning results and health status.

[0055] 3. Positioning implementation steps

[0056] S1: Connect the core of the test cable to the mechanical subsystem interface interconnection device, emit probe light pulses through the pulse light synthesis and amplification module, continuously collect 30 seconds of stable RBS baseline curve and store it in the data processing module.

[0057] S2: According to the historical fault data, set the estimated fault point W0 = 5.0 km, and apply artificial knocking disturbance; the system calls the fingerprint library and identifies the disturbance type as artificial knocking (T = 1) through the Euclidean distance matching formula, with a matching confidence of 0.92.

[0058] S3: The signal acquisition module captures the RBS coherent fading waveform changes, after photoelectric conversion, the data processing module calculates the initial position according to the core formula:

[0059]

[0060] c = 3 × 10 8 m / s, Δt = 46.8 μs), x = 4986 m; through cross-correlation operation τ = 0.2 μs, combined with wavelet packet decomposition:

[0061]

[0062] j=5 layers, k=12 nodes, Ent(j,k)=0.32 =0.85, δ=0.3m, corrected position =4985.8m.

[0063] S4: Since a valid response was detected at W0, the next perturbation is applied according to W1=W0+ΔL; the initial step size ΔL0=80m, the current signal-to-noise ratio S / N=12, the confidence level C=0.85, the adjustment coefficients k1=0.2, k2=0.3, and the step size is calculated by substituting into the step size formula:

[0064]

[0065] We get ΔL≈22.3m; subsequently, as S / N increases to 15 and C=0.91, the step size is gradually reduced to 1.2m.

[0066] S5: When the disturbance position is W i A valid response was detected at =4980m, W i+1 =4981.2m no response, the fault point is determined to be between 4980-4981.2m; the digital twin module maps this range to the GIS map, showing that the fault point is located in a road crossing section (high-risk area), and the system automatically issues an early warning.

[0067] 4. Fault prediction application

[0068] The system continuously collects RBS baseline data, constructs a one-month fiber health status time series, and models it using an LSTM model; it takes parameters such as microbending accumulation and ambient temperature as inputs and outputs the health status value after τ=30 days.

[0069]

[0070] have to =0.35, indicating a high risk of failure within 5km of the optical cable, and recommending targeted maintenance.

[0071] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A fiber optic fault location system based on distributed fiber optic sensing technology, comprising five parts: an optical subsystem, a circuit subsystem, a software subsystem, a mechanical subsystem, and an auxiliary subsystem, characterized in that: The optical subsystem includes a pulsed light synthesis and amplification module and a photoelectric conversion module; the circuit subsystem includes a system control module, a signal acquisition module, and a signal conditioning module; the software subsystem includes a parameter control module, a data processing module, and a data storage module; the mechanical subsystem includes a device integration module and an interface interconnection device; and the auxiliary subsystem includes a system power supply device, a temperature control module, and a human-machine interface.

2. The fiber optic fault location system based on distributed fiber optic sensing technology according to claim 1, characterized in that: The optical subsystem further integrates a polarization controller and a polarization diversity receiver to simultaneously acquire the phase information, intensity information and polarization state information of the RBS signal, thereby achieving joint demodulation of multi-dimensional physical quantities.

3. The fiber optic fault location system based on distributed fiber optic sensing technology according to claim 1, characterized in that: The auxiliary subsystem also includes a built-in reference fiber optic ring or virtual disturbance generator for periodically injecting standard disturbance signals to perform self-calibration and self-diagnosis of system sensitivity, positioning accuracy and hardware status.

4. The fiber optic fault location system based on distributed fiber optic sensing technology according to claim 1, characterized in that: The circuit subsystem integrates an edge intelligent processing unit, which is equipped with a neural network accelerator to perform disturbance event detection, feature extraction and preliminary localization locally in real time, reducing reliance on cloud computing.

5. A location method for an optical fiber fault location system based on distributed optical fiber sensing technology, characterized in that: Includes the following steps: S1: Connect the fiber core of the optical cable under test to the system interface, and emit probe light pulses through the pulse light synthesis and amplification module to obtain the RBS baseline curve under stable conditions; S2: Apply an external disturbance at location W0 near the predicted fault point; S3: The signal acquisition module captures the RBS coherent fading waveform changes caused by the disturbance. After photoelectric conversion and signal conditioning, the data processing module calculates the position of the phase disturbance. S4: If a valid perturbation response is detected at W0, then apply the next perturbation at the far end W1=W0+ΔL; otherwise, apply the perturbation at the near end W1=W0–ΔL, where ΔL is the adaptive step size. S5: When two adjacent disturbance positions W i With W i+1 If only one valid response is detected at any location, the optical cable fault point is determined to be located at W. i With W i+1 between.

6. The location method of an optical fiber fault location system based on distributed optical fiber sensing technology according to claim 5, characterized in that: In S2, the external disturbance includes one or more of the following: manual knocking, ultrasonic excitation, thermal pulse, or electromagnetic excitation; the system pre-establishes an RBS response fingerprint database for different disturbance types, and uses a deep learning model to intelligently identify and verify the disturbance type and location; the type matching formula of the deep learning model is: ， T represents the identified perturbation type, m represents the total number of perturbation types in the fingerprint database, and F represents the feature vector of the current perturbation. Let be the standard feature vector of the k-th perturbation in the fingerprint database, and ‖·‖2 be the Euclidean distance.

7. The location method of an optical fiber fault location system based on distributed optical fiber sensing technology according to claim 5, characterized in that: In S3, the data processing module calculates the phase perturbation position based on the optical path difference principle. The core formula is: ， Where x is the position of the phase perturbation, and c is the speed of light in vacuum. To detect the time difference between the round-trip disturbance point of the optical pulse, n is the refractive index of the optical fiber core; The data processing module employs a time-frequency joint analysis method, including cross-correlation calculations on the two disturbance signals and wavelet packet decomposition to accurately locate the disturbance center with an accuracy better than ±1 meter; the cross-correlation calculation is as follows: ， in, Let x(t) be the cross-correlation coefficient, y(t) be the baseline RBS signal, y(t) be the perturbed RBS signal, and τ be the time delay; combined with the positioning correction formula from wavelet packet decomposition: ， in, To determine the corrected location, Ent(j,k) represents the energy entropy of the k-th wavelet packet node in the j-th layer. δ represents the maximum energy entropy of this layer, and δ is the correction coefficient.

8. The location method of an optical fiber fault location system based on distributed optical fiber sensing technology according to claim 5, characterized in that: In S4, the adaptive step size ΔL is dynamically adjusted based on the current signal-to-noise ratio and the confidence level of the disturbance response. The initial step size is 50–100 meters, which is gradually reduced to 0.5–2 meters during the approximation process. The adjustment formula is: ， in, The initial step size is defined as follows: k1 and k2 are adjustment coefficients, where k1 = 0.1-0.3 and k2 = 0.2-0.

4. The step size decreases rapidly when the signal-to-noise ratio (S / N) ≥ 10; the step size stabilizes between 0.5 and 2 meters when the confidence level (C) ≥ 0.9; the lower limit constraint for the step size is... ≥0.5 meters.

9. The location method of an optical fiber fault location system based on distributed optical fiber sensing technology according to claim 5, characterized in that: The software subsystem further includes a digital twin mapping module, which integrates a geographic information system with an optical cable laying path model. When the positioning result falls into a high-risk area such as a junction box, a road crossing section, or a historical fault point, the system automatically prioritizes disturbance testing and issues an early warning.

10. The location method of an optical fiber fault location system based on distributed optical fiber sensing technology according to claim 5, characterized in that: The system continuously collects RBS baseline data and constructs a time series of fiber health status. It then models microbending accumulation, stress fatigue, or environmental aging trends using a time-series prediction model to predict fault risks and support predictive maintenance decisions. The time-series prediction model employs LSTM, and its core prediction formula is as follows: ， in, This is the fiber health status value after time τ, ranging from 0 to 1. The closer it is to 0, the higher the risk of failure. Use the Sigmoid activation function; For output layer weights, Let be the cell state vector. This is the output layer bias term.

Images