Intelligent optical fiber detection system and method

Abstract

The invention discloses a wavelength-selective optical fiber fault detection method and system, and belongs to the technical field of optical communication monitoring. According to the method, a multi-wavelength OTDR (Optical Time Domain Reflectometer) technology is adopted, OTDR echo signals with different wavelengths are analyzed through an AI module, and the wavelength most suitable for detecting a specific type of optical fiber fault is intelligently selected, so that the detection precision and efficiency are improved. Historical data can be analyzed by using a machine learning algorithm, a wavelength use strategy is optimized, and a comprehensive optical fiber state report is generated in combination with multi-wavelength measurement data. The intelligent level of optical fiber fault detection can be remarkably improved, and the method is particularly suitable for maintenance and management of a large-scale optical fiber network.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to an intelligent optical fiber detection system and method. Background Art

[0002] In modern communication networks, optical fibers undertake important data transmission tasks. However, during the use of optical fibers, faults may occur due to reasons such as aging, external force damage, and excessive bending. There are many problems with traditional optical fiber fault detection methods: The optical time domain reflectometer (OTDR) needs to interrupt services for testing, and the positioning of fault locations in complex network environments is not accurate enough; The method of manually checking optical fiber faults section by section is not only inefficient but also increases labor costs. Therefore, there is an urgent need for a detection technology that can quickly and accurately locate the fault location of optical fibers without affecting services. Summary of the Invention

[0003] In view of the problems existing in the prior art, the present invention provides an intelligent optical fiber detection system, which includes a light source, a circulator, a detection module, a first optical fiber to be measured and its corresponding first optical filter, a second optical fiber to be measured and its corresponding second optical filter; The light source is used to emit lasers of different wavelengths; The circulator guides the laser into the optical fiber to be measured; The detection module is used to record the received laser time information; The first optical filter is arranged on the transmission path of the first optical fiber to be measured; The second optical filter is arranged on the transmission path of the second optical fiber to be measured. The functions of the first optical filter and the second optical filter are to reflect lasers of a preset wavelength and allow lasers of other wavelengths to pass through.

[0004] Preferably, the system further includes an AI module, and the AI module is used to automatically select the most suitable wavelength for detecting specific types of optical fiber faults by analyzing OTDR echo signals of different wavelengths.

[0005] Preferably, the preset wavelength reflected by the first optical filter is different from the preset wavelength reflected by the second optical filter.

[0006] Preferably, the preset wavelength reflected by the first optical filter and the preset wavelength reflected by the second optical filter are spaced 20 - 100 nm apart.

[0007] Preferably, the first optical fiber to be measured and the second optical fiber to be measured are connected in series.

[0008] The present invention also provides a detection method using the above system, and the method includes the following steps:

[0009] S1: The light source emits two wavelengths of lasers, laser λ1 and laser λ2;

[0010] S2: The lasers λ1 and λ2 enter the first fiber under test and the second fiber under test via the circulator.

[0011] S3: In the first fiber under test, when the lasers λ1 and λ2 reach the first optical filter, the laser λ1 is reflected back and finally received by the detection module, and the arrival time t1 is recorded; the laser λ2 continues to move forward until it encounters an obstacle point and returns, and is finally received by the detection module, and the arrival time t2 is recorded; and the specific location of the fault is obtained through the fault point formula L = v*(t1 - t2) / 2, where v is the speed of light propagation in the optical fiber.

[0012] S4: In the second fiber under test, when the lasers λ1 and λ2 reach the second optical filter, the laser λ2 is reflected back and finally received by the detection module, and the arrival time t2 is recorded; the laser λ1 continues to move forward until it encounters an obstacle point and returns, and is finally received by the detection module, and the arrival time t1 is recorded; and the specific location of the fault is obtained through the fault point formula L = v*(t2 - t1) / 2, where v is the speed of light propagation in the optical fiber.

[0013] Wherein, before step S1, an AI module is also adopted to automatically select the wavelength most suitable for detecting fiber faults of a specific type by analyzing OTDR echo signals of different wavelengths.

[0014] Preferably, the preset wavelength reflected by the first optical filter is different from the preset wavelength reflected by the second optical filter.

[0015] Preferably, the preset wavelength reflected by the first optical filter and the preset wavelength reflected by the second optical filter are spaced 20 - 100 nm apart.

[0016] Preferably, K-Means clustering is adopted in the AI module to classify OTDR echo data points, distinguish normal signals, attenuation signals, mutation signals (breaks), etc., and judge which wavelengths are the most sensitive to the current fault by clustering and analyzing OTDR data of different wavelengths.

[0017] Preferably, the first fiber under test and the second fiber under test are connected in series.

[0018] Preferably, in steps S3 and S4, a dispersion compensation algorithm can be adopted for correction.

[0019] Compared with the prior art solutions, the present invention has at least the following beneficial effects:

[0020] 1) The invention utilizes a light source to emit lasers of multiple different wavelengths. When each wavelength of laser propagates in an optical fiber and encounters an optical filter, it will be reflected according to its preset wavelength characteristics. The detection module records the first time when different wavelengths of lasers are received (the time when the laser directly emitted from the light source arrives) and the second time of each reflected laser. By calculating the time difference between the first time and each second time and combining with the propagation speed of light in the optical fiber, the distance from the fault point to the light source can be accurately calculated. This method avoids the problem of inaccurate positioning caused by the increasing error between the optical cable length and the ground projection length in the traditional OTDR technology. It has obvious advantages especially in long-distance optical fiber detection, can achieve high-precision fault location, and solves the problem of insufficient accuracy in optical fiber fault location in the prior art.

[0021] 2) The invention organically integrates components such as a light source, a circulator, a detection module, an optical fiber to be measured, and an optical filter to form a complete intelligent detection system. Each component works together through specific connection and signal transmission methods. For example, the circulator can not only accurately introduce the laser emitted by the light source into the optical fiber to be measured, but also guide the laser reflected by the optical filter to the detection module to ensure the complete transmission of the signal; the cooperation between the detection module and the light source can accurately record time information. This integrated design enables the system to automatically complete the whole process from laser emission to fault location without manual intervention, greatly improving the detection efficiency, solving the problems of scattered equipment, complex operation, and low efficiency in existing optical fiber detection, and realizing the automation and intelligence of optical fiber detection.

[0022] 3) Adding intelligent optimized wavelength selection in multi-wavelength detection of optical fibers can greatly improve the detection accuracy: traditional OTDR relies on manual wavelength selection, which may lead to blind spots in fault detection, while AI can ensure the optimal detection scheme by adaptively selecting wavelengths. In addition, AI automatically analyzes the echo data, reduces human errors, and improves the efficiency of optical fiber maintenance. Finally, this embodiment can be applied to different optical fiber scenarios to improve adaptability: it is applicable to both data centers and urban optical fiber networks (micro-bending faults), and also to long-distance trunk optical fibers (long-distance attenuation monitoring). BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic structural diagram of the intelligent optical fiber detection system of the present invention;

[0024] Reference numerals: 1, light source; 2, circulator; 3, detection module; 41, first optical fiber fault point; 42, second optical fiber fault point; 43, third optical fiber fault point; 51, first semi-transmissive and semi-reflective mirror; 52, second semi-transmissive and semi-reflective mirror; 53, third semi-transmissive and semi-reflective mirror; 61, first filter; 62, second filter; 63, third filter.

[0025] The present invention will be further described in detail below. However, the following examples are merely simple examples of the present invention and do not represent or limit the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims. Specific embodiments

[0026] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings and through specific embodiments.

[0027] As Figure 1 shown, the intelligent optical fiber detection system of the present invention includes a light source 1, a circulator 2, a detection module 3, at least one optical fiber to be tested, and at least one optical filter. The light source inputs lasers of different wavelengths into the first end of the circulator and the first input end of the detection module; the circulator inputs the laser into the optical fiber to be tested; the optical fiber to be tested transmits the laser input by the circulator to the side of the optical filter; the optical filter reflects the laser of the preset wavelength back to the optical fiber to be tested and transmits the laser other than the preset wavelength; the optical fiber to be tested transmits the laser reflected by the optical filter to the side of the circulator; the circulator inputs the reflected laser into the detection module; the detection module records the first time when the laser input by the light source is received and the second times when each reflected laser is received, and determines the faulty optical fiber according to the first time and each second time.

[0028] Embodiment 1

[0029] For a single optical fiber to be tested, the light source emits two different wavelengths of lasers, namely λ1 and λ2. The optical filter is preset to reflect the wavelength λ1. When the laser enters the optical fiber to be tested through the circulator and reaches the optical filter, the laser with the wavelength λ1 is reflected back to the optical fiber to be tested, while the laser with the wavelength λ2 continues to transmit. The detection module records the time t1 when the reflected laser with the wavelength λ1 is received and the time t2 when the laser with the wavelength λ2 is received. According to the propagation speed v of light in the optical fiber and the time difference Δt = t2 - t1, the distance L from the fault point to the light source can be calculated as L = v×Δt / 2. In this way, the fault position in a single optical fiber can be quickly determined.

[0030] The light source is one of the key components of the system and is responsible for emitting lasers of different wavelengths. To meet different detection requirements, a tunable laser can be used as the light source, which can adjust the wavelength of the emitted laser as needed. In practical applications, the wavelength selection of the light source should be determined according to the characteristics of the optical fiber and the reflection wavelength range of the optical filter.

[0031] The function of the circulator is to introduce the laser emitted by the light source into the optical fiber to be tested and guide the laser reflected by the optical filter to the detection module. The performance of the circulator directly affects the transmission efficiency of the laser and the accuracy of detection. Therefore, the circulator should have the characteristics of low insertion loss and high isolation to ensure the signal quality of the laser during transmission.

[0032] The detection module is used to record the received laser time information. It includes a high-precision photodetector and a time measurement unit. The photodetector can convert the received laser signal into an electrical signal, and the time measurement unit is used to accurately measure the time when different wavelength lasers are received. The accuracy of the detection module directly determines the accuracy of fault location. In practical applications, the detection module should have the characteristics of high sensitivity and high time resolution.

[0033] The function of the optical filter is to reflect the laser with a preset wavelength while allowing lasers of other wavelengths to pass through. The performance of the optical filter is crucial for the accuracy of fault location. It should have high reflectivity and high selectivity to ensure that only the laser with the preset wavelength is reflected while lasers of other wavelengths can be transmitted normally. The reflection wavelength of the optical filter can be customized according to actual application requirements.

[0034] The optical fiber to be measured is the object of detection, and its quality directly affects the performance of the communication network. During the detection process, the optical fiber to be measured should maintain a good connection state to avoid signal transmission problems caused by poor connection. At the same time, the length and characteristics of the optical fiber to be measured will also affect the detection results. Therefore, in practical applications, corresponding parameter settings need to be made according to the specific situation of the optical fiber.

[0035] This method is actually an improvement of optical time domain reflectometry (OTDR). It uses the propagation characteristics of light with different wavelengths in the optical fiber to determine the location of the fault point. It has the following advantages: (1) It is not necessary to measure section by section along the optical fiber, and the location of the fault point can be directly calculated through the time difference. (2) As long as the time resolution of the light source and the detection module is high enough, high-precision positioning can be achieved. (3) OTDR devices can usually measure optical fibers in a range of dozens of kilometers, and this method can also be extended to similar application scenarios.

[0036] Embodiment 2

[0037] In an actual communication network, there are often multiple optical fibers. The system of the present invention can detect multiple optical fibers simultaneously. Suppose there are multiple optical fibers to be measured, each optical fiber is connected to an optical filter, and the preset reflection wavelengths of each optical filter are different. The light source sequentially emits lasers of different wavelengths, and the detection module respectively records the reception time of each wavelength laser and the reception time of the reflected laser. By comparing the time differences corresponding to each optical fiber, the location of the fault point in each optical fiber can be calculated respectively. This method greatly improves the efficiency of optical fiber detection and is especially suitable for the maintenance of large-scale optical fiber networks.

[0038] Since the reflection wavelengths of each optical fiber are different, the system can quickly poll multiple wavelengths to achieve simultaneous detection of multiple optical fibers, avoiding the inefficient mode of testing each fiber one by one. In data centers, metropolitan area networks, or backbone optical fiber networks, where there are numerous optical fibers, this method can detect the health status of multiple optical fiber channels in a short time, improving the operation and maintenance efficiency. Combined with an automatic optical fiber switching device (such as an optical switch), a fully automated optical fiber monitoring system can be constructed to achieve remote fault location and alarm.

[0039] Select an appropriate wavelength interval to avoid signal confusion caused by overlapping filter bandwidths. The speed of light in an optical fiber is approximately 2 * 10 8 m / s, and the refractive index of light in the optical fiber n ≈ 1.5.

[0040] The number of optical fibers to be measured is 3. Each optical fiber is connected to a different optical filter. The preset reflection wavelengths of the optical filters are specifically: for the first optical fiber, λ1 = 1550 nm; for the second optical fiber, λ2 = 1570 nm; for the third optical fiber: λ3 = 1590 nm; the time resolution of the detection module: 10 ps. Based on the resolution of the existing detection module, a reflection wavelength interval of 20 - 100 nm is more appropriate.

[0041] In the first optical fiber, the arrival time of the reflected light (t1) reflected by the first filter 61: 20.000 ns, the arrival time of the reflected light (t2) from the fault point 41: 20.600 ns, time difference: Δt1 = t2 - t1 = 20.600 - 20.000 = 0.600 ns;

[0042] In the second optical fiber, the arrival time of the reflected light (t1) reflected by the second filter 62: 19.500 ns, the arrival time of the reflected light (t2) from the fault point 42: 20.300 ns, time difference: Δt1 = t2 - t1 = 20.300 - 19.500 = 0.800 ns;

[0043] In the third optical fiber, the arrival time of the reflected light (t1) reflected by the third filter 63: 21.000 ns, the arrival time of the reflected light (t2) from the fault point 43: 21.400 ns, time difference: Δt1 = t2 - t1 = 21.400 - 21.000 = 0.400 ns.

[0044] Table 1 Data table for detecting the location of optical fiber fault points reflected by different wavelengths

[0045]

[0046] To measure such time intervals, a photodetector needs to have extremely high time resolution. It usually includes: a photodiode (PD). Common models: avalanche photodiode (APD); response time: dozens of picoseconds (ps). For example, an InGaAs APD (indium gallium arsenide avalanche photodiode) can reach a response speed of the order of 10 ps and is suitable for communication wavelengths of 1.3 μm and 1.55 μm. In addition, the photodetector itself only converts the optical signal into an electrical signal to measure the nanosecond-picosecond level time difference. For example, for a high-speed oscilloscope, its bandwidth requirement: > 10 GHz (ideally above 40 GHz). Different wavelengths may produce different dispersion effects in long-distance optical fibers, affecting the measurement accuracy, and a dispersion compensation algorithm can be used for correction.

[0047] Embodiment 3

[0048] In the above Embodiments 1 and 2, testers usually need to manually select the test wavelength and rely on experience to analyze the OTDR echo signal to judge the type and location of fiber faults. However, due to the various types of fiber faults, such as microbend loss, breakage, fusion loss, etc., a single-wavelength OTDR may have detection blind spots in some scenarios. Therefore, this embodiment proposes an OTDR fault detection system with adaptive wavelength selection optimized by artificial intelligence (AI) to improve the detection accuracy and efficiency.

[0049] The OTDR system of this Embodiment 3 includes: a multi-wavelength laser light source module (supporting 1310 nm, 1550 nm, 1625 nm, etc.); an optical signal detection module (using a high-precision APD or SPAD as a detector); a data processing and AI analysis unit (deep learning model) and a control and decision-making unit (dynamically selecting the optimal wavelength according to the analysis result).

[0050] Working steps:

[0051] 1. Initial test: The OTDR sequentially emits laser pulses of different wavelengths such as 1310 nm, 1550 nm, 1625 nm, etc. and records the echo signal.

[0052] 2. AI analysis: Use a machine learning model to analyze the OTDR echo data of different wavelengths to judge the possible types of faults in the optical fiber. Here, the AI pre-analysis is to apply a mathematical model to the field for analyzing the OTDR echo data of different wavelengths. For example: classical machine learning (ML) methods. This classical machine learning method is suitable for small-scale data analysis and can quickly identify common optical fiber fault patterns.

[0053] For example, K-Means clustering is adopted. Its principle is to classify the OTDR echo data points. For example, different wavelengths of OTDR data are classified by cluster analysis to distinguish normal signals, attenuation signals, mutation signals (breaks), etc., and to determine which wavelengths are the most sensitive to the current fault.

[0054] After detecting OTDR signals of multiple wavelengths, K-Means can find that the attenuation of 1310nm is extremely large in the case of micro-bend faults, and automatically preferentially select 1310nm for the next fine scan. If micro-bend loss is identified, 1310nm is preferentially used for detailed testing. If a long-distance attenuation problem is identified, 1550nm or 1625nm is preferentially used for long-distance detection. If a possible fiber break or severe loss is identified, multi-wavelength joint analysis is used to improve the fault precise positioning ability. Final fault diagnosis and reporting: Using the above method combined with OTDR data of multiple wavelengths, accurately calculate the fault point location and possible fault types.

[0055] For example, in the AI recognition of micro-bend faults: Optimization of micro-bend fault detection In an actual optical fiber communication network, the optical fiber may be subject to micro-bend losses due to external pressure (such as tying wires, bending). Such faults are usually more sensitive to the 1310nm wavelength and have less impact on 1550nm and 1625nm. Test process: It is found that the attenuation of the 1310nm OTDR signal is greater than that of 1550nm, and it is speculated that there may be micro-bend losses. The system automatically selects the 1310nm wavelength for further scanning to improve the detection accuracy. By learning the echo characteristics, judge the severity of the micro-bend loss and provide repair suggestions.

[0056] In addition to the above application scenarios, in the optimization of long-distance optical fiber attenuation monitoring, for example, for long-distance trunk optical fibers, the overall attenuation of the optical fiber is a key issue. Usually, the 1550nm or 1625nm wavelength has less attenuation during long-distance propagation and is more suitable for long-distance monitoring. Test process: The model finds that the intensity of the 1550nm OTDR echo signal is higher than that of 1310nm, indicating that 1550nm is more suitable for long-distance monitoring of this line.

[0057] The above use of AI to find the 1550nm echo can also be applicable to long-distance line monitoring, including:

[0058] (1) Data acquisition:

[0059] The OTDR sequentially sends optical pulses of 1310nm, 1550nm, and 1625nm wavelengths and records the echo signals of each wavelength.

[0060] The key data among them include: Backscatter Level (BS), Attenuation (ATT), Signal-to-Noise Ratio (SNR), Event Points (EP) (such as joints, bends, break points).

[0061] (2) AI calculates the attenuation of each wavelength:

[0062] AI uses regression analysis + machine learning to calculate the signal attenuation of each wavelength at different distances:

[0063]

[0064] Among them: ATT λ : The attenuation value of a certain wavelength (dB / km); P input : Input optical power; P output : Return optical power L: Fiber length (km).

[0065] AI compares the average attenuation values of each wavelength:

[0066] If ATT(1310nm) > ATT(1550nm), it means that the loss of 1550nm is lower and it is more suitable for long-distance monitoring.

[0067] If ATT(1550nm) ≈ ATT(1625nm) and SNR(1550nm) > SNR(1625nm), then 1550nm is more suitable for long-distance monitoring than 1625nm.

[0068] (3) AI calculates the signal strength ratio:

[0069] AI calculates the ratio of the backscatter signal strength of 1550nm and 1310nm:

[0070]

[0071] If R > 1 (that is, the OTDR backscatter signal strength of 1550nm is higher than that of 1310nm), further calculate the applicability:

[0072] R >> 1R (such as above 1.5): 1550nm is applicable to long-distance monitoring and it is recommended as the main wavelength.

[0073] 1.0 < R < 1.5: The applicability of 1550nm and 1310nm is similar, and it is recommended to make a decision in combination with other parameters.

[0074] R ≤ 1: 1550nm is affected by abnormal attenuation and it is not recommended as the long-distance monitoring wavelength.

[0075] (4) The AI combines the signal-to-noise ratio (SNR) to confirm the final wavelength:

[0076] The AI calculates the signal-to-noise ratio for each wavelength:

[0077]

[0078] If SNR(1550nm) > SNR(1310nm), it indicates that 1550nm is more stable on this line and is suitable for long-distance monitoring.

[0079] If the SNR of 1550nm is lower than that of 1310nm, there may be non-linear effects and bending losses, and manual confirmation is required.

[0080] The decision-making logic of the AI to select 1550nm as the long-distance monitoring wavelength: The AI uses a decision tree model to automatically select the optimal wavelength:

[0081]

[0082] If the above conditions are met, the AI recommends 1550nm as the long-distance monitoring wavelength.

[0083] The AI automatically adjusts the OTDR measurement strategy:

[0084] After the AI finds that 1550nm is suitable for long distances, it automatically optimizes the OTDR measurement parameters: reduces the test frequency of 1310nm to save power consumption. Improves the test resolution of 1550nm for refined long-distance monitoring. When monitoring fiber aging, combines with 1625nm for auxiliary analysis.

[0085] The above AI automatically identifies the most suitable OTDR wavelength for long-distance monitoring through attenuation calculation, signal strength comparison, and SNR analysis. Adopts 1550nm as the main wavelength to optimize the OTDR monitoring scheme and improve long-distance measurement accuracy. The AI dynamically adjusts the measurement strategy to improve the fiber monitoring efficiency and reduce manual judgment errors. In this way, the AI can automatically determine whether 1550nm is the most suitable for long-distance monitoring and optimize the OTDR signal processing strategy.

[0086] The system automatically uses a wavelength of 1550nm (or 1625nm) for precise scanning to improve the ranging ability.

[0087] The AI combines data of different wavelengths to analyze the fiber aging status and give early warnings of potential problems.

[0088] The intelligent optical fiber detection system and method of the present invention effectively solve the problems existing in the existing optical fiber fault detection through refined invention points and innovative technical solutions. It can quickly and accurately locate the optical fiber fault position without affecting communication services, reduce labor costs, and improve the maintenance efficiency of optical cables. The present invention has broad application prospects and is of great significance for ensuring the stable operation of communication networks.

[0089] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solutions of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0090] In addition, it should be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without conflict. To avoid unnecessary repetition, the present invention will not separately describe various possible combination methods.

[0091] Furthermore, any combination can be made between different embodiments of the present invention as long as it does not violate the idea of the present invention, and it should also be regarded as the content disclosed by the present invention.

Claims

1. An intelligent optical fiber detection system, characterized in that: The system includes a light source, a circulator, a detection module, a first optical fiber to be tested and a corresponding first optical filter, a second optical fiber to be tested and a corresponding second optical filter; The light source is used to emit lasers of different wavelengths; The circulator guides the laser into the optical fiber to be tested; The detection module is used to record the received laser time information; The first optical filter is arranged on the transmission path of the first optical fiber to be tested; the second optical filter is arranged on the transmission path of the second optical fiber to be tested The first optical filter and the second optical filter function to reflect laser light of a preset wavelength while allowing laser light of other wavelengths to pass through.

2. The system according to claim 1 further comprises an AI module, wherein the AI ​​module is used to automatically select the wavelength most suitable for detecting a specific type of optical fiber fault by analyzing OTDR echo signals of different wavelengths.

3. The system according to claim 1, characterized in that: The predetermined wavelength reflected by the first optical filter is different from the predetermined wavelength reflected by the second optical filter.

4. The system according to claim 2, characterized in that: The preset wavelength reflected by the first optical filter and the preset wavelength reflected by the second optical filter are spaced 20-100 nm apart.

5. The system according to claim 2, characterized in that: The first optical fiber to be tested and the second optical fiber to be tested are connected in series.

6. A detection method using the system of any one of claims 1 to 5, characterized in that: The method comprises the following steps: S1: The light source emits two wavelengths of laser light λ1 and laser light λ2; S2: the laser λ1 and the laser λ2 enter the first optical fiber to be tested and the second optical fiber to be tested through the circulator; S3: In the first optical fiber to be tested, when the laser λ1 and the laser λ2 reach the first optical filter, the laser λ1 is reflected back and is finally received by the detection module, and the arrival time t1 is recorded; the laser λ2 continues to move forward until it encounters an obstacle point and returns, and is finally received by the detection module, and the arrival time t2 is recorded; and the specific location of the fault is obtained through the fault point formula L=v*(t1-t2) / 2, where v is the speed of light propagation in the optical fiber; S4: In the second optical fiber to be tested, when the laser λ1 and the laser λ2 reach the second optical filter, the laser λ2 is reflected back and is finally received by the detection module, and the arrival time t2 is recorded; the laser λ1 continues to move forward until it encounters an obstacle point and returns, and is finally received by the detection module, and the arrival time t1 is recorded; and the specific location of the fault is obtained through the fault point formula L=v*(t2-t1) / 2, where v is the speed of light propagation in the optical fiber; Before step S1, an AI module is used to automatically select the wavelength most suitable for detecting a specific type of optical fiber fault by analyzing OTDR echo signals of different wavelengths.

7. The method according to claim 6, characterized in that: The predetermined wavelength reflected by the first optical filter is different from the predetermined wavelength reflected by the second optical filter.

8. The method according to claim 7, characterized in that: The AI ​​module adopts K-Means clustering to classify OTDR echo data points, distinguish normal signals, attenuated signals, mutation signals (breaks), etc., and analyzes OTDR data of different wavelengths through clustering to determine which wavelengths are most sensitive to the current fault.

9. The method according to claim 7, characterized in that: The first optical fiber to be tested and the second optical fiber to be tested are connected in series.

10. The method according to claim 6, characterized in that: In step S3 and step S4, a dispersion compensation algorithm may be used for correction.

Images