Method for realizing optical cable fiber core health prediction by using B-OTDR technology

Abstract

The invention discloses a method for realizing optical cable fiber core health prediction by using a B-OTDR technology, which comprises the following steps: a multi-physical quantity fusion detection module synchronously obtains Brillouin scattering and Raman scattering signals by using a double-pulse modulation technology, separates two kinds of scattered light through frequency division multiplexing, and constructs a strain-temperature decoupling model; and the dynamic self-adaptive sampling module automatically switches sampling resolution according to the position of the optical fiber, adopts high-resolution continuous scanning on a historical fault high-incidence area, adopts low-resolution sparse sampling on a normal area, and automatically triggers local encryption sampling when an abnormal event is detected. The double pulses are respectively calculated by adopting a double-pulse modulation technology, so that the prediction accuracy is improved; dynamic self-adaptive sampling is adopted, segmented sampling is carried out on a high fault area and a normal area, meanwhile, local sampling is carried out on an abnormal section, the invalid data volume is reduced, and meanwhile the monitoring precision of a key area is ensured.

Description

Technical Field

[0001] This invention relates to the field of optical fiber core technology, specifically to a method for predicting the health of optical fiber cores using B-OTDR technology. Background Technology

[0002] In recent years, a number of online monitoring devices for optical cables have been deployed. Conventional online monitoring devices utilize ordinary OTDRs based on Rayleigh reflection to monitor changes in line attenuation and perform comprehensive degradation trend analysis of the optical cable based on long-term statistical attenuation curves. Another method uses fiber Bragg grating encoding, replacing patch cords and inserting coded gratings into the optical cable splice closure to reduce positioning errors. While fiber Bragg grating is suitable for new optical cables, upgrading existing cables requires opening the existing splice closures, posing significant engineering risks. This can easily lead to fiber core breakage, and resealing the splice closure may result in incomplete sealing. This method involves high investment, high risk, and extremely low cost-effectiveness.

[0003] Existing B-OTDR technology can only measure a single parameter, such as strain or temperature, but cannot monitor multiple parameters simultaneously. The data processing method is not efficient enough, resulting in prediction delays. Existing systems perform poorly in high-noise environments, and data drift is a serious problem during long-term monitoring, which can easily lead to false alarms. In addition, existing sampling generally uses fixed resolution sampling, which leads to data redundancy and wasted computing resources. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for predicting the health of optical fiber cores using B-OTDR technology, comprising:

[0005] The multi-physical quantity fusion detection module uses dual-pulse modulation technology to simultaneously acquire Brillouin scattering and Raman scattering signals, and separates the two types of scattered light through frequency division multiplexing to construct a strain-temperature decoupling model;

[0006] The dynamic adaptive sampling module automatically switches the sampling resolution according to the fiber location. It uses high-resolution continuous scanning for areas with high historical fault incidence and low-resolution sparse sampling for normal areas. When an abnormal event is detected, it automatically triggers local encrypted sampling.

[0007] The hybrid analysis module, which includes edge computing nodes and cloud prediction models, establishes a quantitative assessment model for fiber core health and defines a comprehensive fiber core health index.

[0008] The self-calibration module uses a dynamic baseline calibration mechanism based on historical data to periodically update the initial Brillouin frequency shift value.

[0009] The deep learning module uses deep learning neural networks to extract detection data from multiple time periods for horizontal and vertical comparisons and stores dynamic data.

[0010] The environmental noise suppression module employs chirped pulse modulation technology to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum, thereby improving the signal-to-noise ratio. It also utilizes a joint denoising algorithm combining wavelet transform and Kalman filtering to suppress signal fluctuations caused by external vibrations.

[0011] Preferably, the dynamic adaptive sampling module identifies high-fault-risk areas through a convolutional neural network and triggers local encrypted sampling.

[0012] Preferably, the edge computing nodes in the hybrid analysis module use compressed sensing technology to transmit feature parameters to the cloud.

[0013] Preferably, the strain-temperature decoupling model formula is:

[0014] V B =V BO +C VT ΔT+C Vε Δε

[0015] P B =P BO +C PT ΔT+C Pε Δε

[0016] In the formula, V B P represents the Brillouin frequency shift. B V represents the power of Brillouin scattering. BO P represents the Brillouin frequency shift at reference temperature and stress. BO The power of Brillouin scattering at the reference temperature is represented by T, where T represents the temperature value, ε represents the stress value, and C represents the stress value. VT C Vε C represents the temperature coefficient and stress coefficient of the Brillouin frequency shift, respectively. PT C Pε These represent the temperature coefficient and stress coefficient of Brillouin power, respectively.

[0017] Preferably, the formula for the fiber core comprehensive health index (FHI) is:

[0018] FHI = α·Δν_B / ν_B+β·(dP / dz)+γ·τ

[0019] Where: Δν_B is the Brillouin frequency shift change; dP / dz is the optical power attenuation gradient; τ is the polarization state fluctuation variance; α, β, γ are weighting coefficients, which are dynamically optimized through machine learning.

[0020] Preferably, it includes the following steps;

[0021] S1. Acquire various initial data for fiber optic measurements, including Brillouin scattering and Raman scattering signals obtained under two-pulse signals at different resolutions;

[0022] S2. The collected initial data are processed in the hybrid analysis module, and the environmental noise suppression module is introduced to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum. The combined denoising algorithm of wavelet transform and Kalman filtering is used to suppress signal fluctuations caused by external vibrations, optimize the data, and obtain the health index through the fiber core comprehensive health index formula.

[0023] S3. Output health index and assess and predict fiber optic health.

[0024] S4. Store the optimized data, compare it with historical data, and optimize the weighting coefficients.

[0025] The beneficial effects of this invention are reflected in:

[0026] 1. By employing dual-pulse modulation technology, the two pulses are calculated separately, thereby improving the accuracy of prediction;

[0027] 2. Dynamic adaptive sampling is adopted, and high-fault areas and normal areas are sampled in segments. At the same time, local sampling is added to abnormal segments, which not only reduces the amount of invalid data, but also ensures the monitoring accuracy of key areas.

[0028] 3. Employing edge computing and cloud prediction, a lightweight AI model is deployed at the optical cable monitoring node to extract Brillouin gain spectrum feature parameters in real time. The processed data is then uploaded to the cloud, where the fiber core health is assessed and predicted, achieving the dual capabilities of millisecond-level local anomaly alarm and hour-level global lifespan prediction.

[0029] 4. By using chirped pulse modulation technology to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum, the signal-to-noise ratio is improved and the false alarm rate is reduced.

[0030] 5. Add self-calibration function, based on historical data dynamic baseline calibration mechanism, regularly update the initial Brillouin frequency shift value, and improve the accuracy of prediction. Attached Figure Description

[0031] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0032] Figure 1 A flowchart illustrating a method for predicting the health of optical fiber cores using B-OTDR technology. Detailed Implementation

[0033] The embodiments of the technical solution of the present invention will now be described in detail with reference to the accompanying drawings. These embodiments are merely illustrative of the technical solution of the present invention and are therefore intended to limit the scope of protection of the present invention.

[0034] It should be noted that, unless otherwise stated, the technical or scientific terms used in this application should have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.

[0035] like Figure 1 As shown, the method for predicting the health of optical fiber cores using B-OTDR technology includes:

[0036] The multi-physical quantity fusion detection module uses dual-pulse modulation technology to simultaneously acquire Brillouin scattering and Raman scattering signals. It separates the two types of scattered light through frequency division multiplexing and constructs a strain-temperature decoupling model. The formula for the strain-temperature decoupling model is as follows:

[0037] V B =V BO +C VT ΔT+C Vε Δε

[0038] P B =P BO +C PT ΔT+C Pε Δε

[0039] In the formula, V B P represents the Brillouin frequency shift. B V represents the power of Brillouin scattering. BO P represents the Brillouin frequency shift at reference temperature and stress. BO The power of Brillouin scattering at the reference temperature is represented by T, where T represents the temperature value, ε represents the stress value, and C represents the stress value. VT C Vε C represents the temperature coefficient and stress coefficient of the Brillouin frequency shift, respectively. PT C Pε These represent the temperature coefficient and stress coefficient of Brillouin power, respectively.

[0040] The dynamic adaptive sampling module identifies high-fault-risk areas through a convolutional neural network and triggers local encrypted sampling. It automatically switches the sampling resolution according to the fiber location. High-resolution continuous scanning is used for areas with a high historical fault incidence rate, with a sampling interval of 1 meter. Low-resolution sparse sampling is used for normal areas, with a sampling interval of 10 meters. When an abnormal event is detected, local encrypted sampling is automatically triggered, with a sampling interval of 0.5 meters.

[0041] The hybrid analysis module, comprising edge computing nodes and a cloud-based prediction model, establishes a quantitative assessment model for fiber core health and defines a comprehensive fiber core health index. The edge computing nodes in the hybrid analysis module use compressed sensing technology to transmit feature parameters to the cloud. Data is prioritized through these edge computing nodes, and then the processed data is sent to the cloud, achieving both millisecond-level local anomaly alerts and hourly-level global lifetime prediction capabilities, reducing latency. The formula for the comprehensive fiber core health index (FHI) is:

[0042] FHI = α·Δν_B / ν_B+β·(dP / dz)+γ·τ

[0043] Where: Δν_B is the Brillouin frequency shift change; dP / dz is the optical power attenuation gradient; τ is the polarization state fluctuation variance; α, β, γ are weighting coefficients, which are dynamically optimized through machine learning.

[0044] The self-calibration module uses a dynamic baseline calibration mechanism based on historical data to periodically update the initial Brillouin frequency shift value.

[0045] The deep learning module uses deep learning neural networks to extract detection data from multiple time periods for horizontal and vertical comparisons and stores dynamic data. It can continuously update the weight coefficients in the fiber core comprehensive health index formula, thereby improving the accuracy of prediction.

[0046] The environmental noise suppression module employs chirped pulse modulation technology to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum, thereby improving the signal-to-noise ratio. It also utilizes a joint denoising algorithm combining wavelet transform and Kalman filtering to suppress signal fluctuations caused by external vibrations.

[0047] The method for predicting the health of optical fiber cores using B-OTDR technology includes the following steps;

[0048] S1. Acquire various initial data for fiber optic measurements, including Brillouin scattering and Raman scattering signals obtained under two-pulse signals at different resolutions;

[0049] S2. The collected initial data are processed in the hybrid analysis module, and the environmental noise suppression module is introduced to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum. The combined denoising algorithm of wavelet transform and Kalman filtering is used to suppress signal fluctuations caused by external vibrations, optimize the data, and obtain the health index through the fiber core comprehensive health index formula.

[0050] S3. Output health index and assess and predict fiber optic health.

[0051] S4. Store the optimized data, compare it with historical data, and optimize the weighting coefficients.

[0052] Table 1 below compares the method of this invention with that of traditional B-OTD.

[0053] Table 1

[0054] index Traditional B-OTDR This invention Increase Strain measurement error ±50με ±15με 70%↓ Fault location accuracy ±5m ±15με 94%↑ Data processing delay 5-10s <1s 80%↓

[0055] The table above clearly shows that, compared with the traditional B-OTDR, there is a significant reduction in strain measurement error and data processing delay, and a significant increase in fault location accuracy.

[0056] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for predicting the health of optical fiber cores using B-OTDR technology, characterized by: include: The multi-physical quantity fusion detection module uses dual-pulse modulation technology to simultaneously acquire Brillouin scattering and Raman scattering signals, and separates the two types of scattered light through frequency division multiplexing to construct a strain-temperature decoupling model; The dynamic adaptive sampling module automatically switches the sampling resolution according to the fiber location. It uses high-resolution continuous scanning for areas with high historical fault incidence and low-resolution sparse sampling for normal areas. When an abnormal event is detected, it automatically triggers local encrypted sampling. The hybrid analysis module, which includes edge computing nodes and cloud prediction models, establishes a quantitative assessment model for fiber core health and defines a comprehensive fiber core health index. The self-calibration module uses a dynamic baseline calibration mechanism based on historical data to periodically update the initial Brillouin frequency shift value. The deep learning module uses deep learning neural networks to extract detection data from multiple time periods for horizontal and vertical comparisons and stores dynamic data. The environmental noise suppression module employs chirped pulse modulation technology to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum, thereby improving the signal-to-noise ratio. It also utilizes a joint denoising algorithm combining wavelet transform and Kalman filtering to suppress signal fluctuations caused by external vibrations. The risk assessment module evaluates the comprehensive analysis of the data and determines the risk level.

2. The method for predicting the health of optical fiber cores using B-OTDR technology as described in claim 1, characterized in that: The dynamic adaptive sampling module identifies high-fault-risk areas through a convolutional neural network and triggers local encrypted sampling.

3. The method for predicting the health of optical fiber cores using B-OTDR technology as described in claim 2, characterized in that: The edge computing nodes in the hybrid analysis module use compressed sensing technology to transmit feature parameters to the cloud.

4. The method for predicting the health of optical fiber cores using B-OTDR technology as described in claim 3, characterized in that: The strain-temperature decoupling model formula is as follows: V B =V BO +C VT ΔT+C Vε Δε P B =P BO +C PT ΔT+C Pε Δε In the formula, V B P represents the Brillouin frequency shift. B V represents the power of Brillouin scattering. BO P represents the Brillouin frequency shift at reference temperature and stress. BO The power of Brillouin scattering at the reference temperature is represented by T, where T represents the temperature value, ε represents the stress value, and C represents the stress value. VT C Vε C represents the temperature coefficient and stress coefficient of the Brillouin frequency shift, respectively. PT C Pε These represent the temperature coefficient and stress coefficient of Brillouin power, respectively.

5. The method for predicting the health of optical fiber cores using B-OTDR technology as described in claim 1, characterized in that: The formula for the Fiber Core Comprehensive Health Index (FHI) is as follows: FHI = α·Δν_B / ν_B+β·(dP / dz)+γ·τ Where: Δν_B is the Brillouin frequency shift change; dP / dz is the optical power attenuation gradient; τ is the polarization state fluctuation variance; α, β, γ are weighting coefficients, which are dynamically optimized through machine learning.

6. The method for predicting the health of optical fiber cores using B-OTDR technology as described in claim 1, characterized in that: Includes the following steps; S1. Acquire various initial data for fiber optic measurements, including Brillouin scattering and Raman scattering signals obtained under two-pulse signals at different resolutions; S2. The collected initial data are processed in the hybrid analysis module, and the environmental noise suppression module is introduced to match the time-frequency characteristics of the probe light pulse with the Brillouin gain spectrum. The combined denoising algorithm of wavelet transform and Kalman filtering is used to suppress signal fluctuations caused by external vibrations, optimize the data, and obtain the health index through the fiber core comprehensive health index formula. S3. Output health index and assess and predict fiber optic health. S4. Store the optimized data, compare it with historical data, and optimize the weighting coefficients.

Images