Intelligent repairing equipment for optical fiber end face defects

Abstract

The invention discloses an intelligent repairing device for optical fiber end face defects, and relates to the technical field of optical fiber end face repairing, the intelligent repairing device for the optical fiber end face defects fundamentally improves the quality and efficiency of optical fiber end face defect repairing through the design of intelligence, automation and multi-technology fusion, and improves the repairing efficiency of the optical fiber end face defects. The method further promotes the crossing of the optical fiber repair technology from experience driving to data driving and from manual operation to intelligent equipment, and provides powerful support for efficient maintenance and performance guarantee of optical fiber devices in the field of optical fiber communication.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical fiber end face repair, in particular to an optical fiber end face defect intelligent repair device. BACKGROUND

[0002] The existing optical fiber end face defect intelligent repair device still has the following disadvantages in actual use:

[0003] In the current optical fiber end face repair field, the existing technology has obvious limitations in many key technical links, making it difficult to achieve an efficient, precise and intelligent repair process. Traditional technology lacks integrated high-precision detection means, making it difficult to collect micro images and data of the optical fiber end face in real time and comprehensively, and unable to provide accurate defect information support for subsequent repair. And the existing technology relies on manual judgment or simple rule setting, lacks automatic identification and analysis capability based on advanced algorithms such as deep learning, and is difficult to accurately identify the location and severity of different types of defects such as scratches, cracks and pollution. It is also impossible to automatically generate an optimal repair strategy that fits the situation, resulting in a significant reduction in the relevance and effectiveness of the repair process.

[0004] The existing device has a single function and is difficult to cope with complex and diverse defect situations. The positioning accuracy of traditional devices is insufficient, making it difficult to achieve stable clamping of the optical fiber and micron-level precise alignment of the repair tool and the defect area, directly affecting the consistency and reliability of the repair effect.

[0005] At the same time, the existing technology generally lacks a perfect closed-loop feedback system, making it impossible to verify the repair effect through secondary detection after repair, difficult to ensure repair quality, and prone to the situation of unqualified products flowing out. It is also not conducive to the improvement of operation efficiency and the timely handling of abnormal situations, as it cannot provide convenient manual intervention or parameter adjustment methods for operators. These technical deficiencies make it difficult for the existing optical fiber end face repair device to meet the high-precision, high-efficiency and intelligent repair requirements. SUMMARY

[0006] The purpose of the present application is to provide an optical fiber end face defect intelligent repair device to solve the above problems.

[0007] In order to achieve the above purpose, the present application provides the following technical scheme: an optical fiber end face defect intelligent repair device, comprising a defect detection module, an intelligent decision module, a repair execution module, a motion positioning module and a closed-loop feedback module;

[0008] The defect detection module adopts a multi-modal detection structure composed of a microscopic imaging assembly and a spectral analysis assembly for collecting defect images and spectral data of the optical fiber end face;

[0009] The intelligent decision module is connected with the defect detection module, and performs feature extraction and classification processing on the defect image and the spectrum data to generate repair parameters corresponding to the defect type.

[0010] The repair execution module includes switchable grinding units, laser repair units and cleaning units, and is connected with the intelligent decision module to perform repair operations according to the repair parameters.

[0011] The motion positioning module is used to adjust the relative position of the optical fiber and the repair execution module.

[0012] The closed-loop feedback module is connected with the defect detection module and the intelligent decision module respectively, and can feed back the detection results after repair to the intelligent decision module to adjust the repair parameters.

[0013] Further, the microscopic imaging assembly of the defect detection module includes a high-resolution CCD camera, a metallographic objective lens and a ring light source, the magnification of the metallographic objective lens is 50-200 times, and the wavelength range of the ring light source is 400-700 nm.

[0014] Further, the defect detection module further includes an image preprocessing assembly for noise reduction, contrast enhancement and edge sharpening processing of the collected defect image.

[0015] Further, the intelligent decision module includes a feature extraction submodule and a strategy generation submodule; the feature extraction submodule can extract the geometric parameters of the defect from the defect image and extract the material composition information of the defect from the spectrum data; the strategy generation submodule matches the pre-set repair process database based on the geometric parameters and the material composition information to generate the repair parameters.

[0016] Further, the geometric parameters include the length, width, depth and area of the defect, and the material composition information is used to distinguish between contaminating defects and structural defects.

[0017] Further, the grinding unit of the repair execution module includes a precision rotary motor, a grinding disc and a pressure adjusting assembly, the diameter of the grinding disc is 10-50 mm, and the pressure adjusting range of the pressure adjusting assembly is 0.1-5 N.

[0018] Further, the laser repair unit of the repair execution module adopts a nanosecond pulse laser, the output wavelength of which is 1064 nm, the pulse width is 10-100 ns, and the energy adjusting range is 1-50 μJ.

[0019] Further, the motion positioning module includes a three-dimensional translation stage and an optical fiber clamp, the displacement resolution of the three-dimensional translation stage is 0.1 microns, and the optical fiber clamp is adapted to optical fibers with diameters of 0.125 mm-0.9 mm.

[0020] Further, the closed loop feedback module is provided with a determination threshold, when the repaired defect parameter is lower than the determination threshold, it is determined that the repair is qualified, when the repaired defect parameter is higher than the determination threshold, the intelligent decision module generates the repair parameter again and controls the repair execution module to execute the repair operation again.

[0021] Further, a man-machine interaction module is further included, which is connected with the intelligent decision module, can display the defect detection result, the repair parameter and the repair progress, and supports manual modification of the repair parameter.

[0022] Compared with the prior art, the optical fiber end face defect intelligent repair equipment provided by the application has the following beneficial effects:

[0023] The optical fiber end face defect intelligent repair equipment, from the automation degree of the repair process, the equipment builds a complete closed loop system of "defect detection-intelligent decision-repair execution-closed loop verification", and can complete the whole process operation from defect identification to repair qualification without manual intervention; this design changes the status of relying on manual experience judgment and manual parameter adjustment in the traditional repair method, greatly reduces the requirement for the skill level of the operator, avoids the error that may be introduced by manual operation, and ensures the stability and consistency of the repair process.

[0024] In terms of accuracy of defect identification and classification, the equipment combines multi-modal detection technology of microscopic imaging and spectral analysis; the microscopic imaging can clearly capture the geometric morphological features of the defect, and the spectral analysis can deeply identify the material composition attribute of the defect, and the combination of the two realizes the accurate differentiation and comprehensive characterization of various microscopic defects (such as scratches, cracks, pollution, etc.) of the optical fiber end face; such multi-dimensional defect cognition ability lays a solid foundation for the subsequent targeted formulation of repair strategies, and effectively avoids the problem of improper repair scheme caused by defect misjudgment.

[0025] The adaptability and flexibility of the repair process is another outstanding advantage of the equipment; the equipment integrates multiple repair units such as laser repair, grinding, cleaning, etc., and realizes adaptive switching and dynamic matching of parameters of different processes through the intelligent decision module; for different types and degrees of defects, the system can automatically select the optimal repair path and process parameters, which not only ensures the effective repair of structural defects, but also properly handles pollution defects, while avoiding the limitations of a single repair process in complex defect processing, significantly improving the coverage ability of the equipment for diversified defect scenarios.

[0026] From the convenience of operation and the expansion of application scenarios, the device is equipped with an intuitive human-computer interaction interface, and the operator can master the defect information, repair progress and parameter state in real time, and can also perform manual intervention when necessary, balancing automation and controllability; in addition, the precise motion positioning system and modular design of the device enable it to adapt to various specifications of optical fiber types, and have application potential in different environments such as laboratories, production workshops and even field operations, while ensuring repair accuracy, breaking the dependence of traditional repair equipment on specific operating environments.

[0027] In summary, through the design of intelligence, automation and multi-technology integration, not only the quality and efficiency of optical fiber end face defect repair are fundamentally improved, but also the optical fiber repair technology is promoted from experience-driven to data-driven and from manual operation to intelligent equipment, providing strong support for efficient maintenance and performance guarantee of optical fiber devices in the field of optical fiber communication. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0029] Figure 1 It is a schematic diagram of the overall process of the present application. DETAILED DESCRIPTION

[0030] In order to enable those skilled in the art to better understand the technical solutions of the present application, the present application will be further described in detail below with reference to the drawings.

[0031] Please refer to Figure 1 An optical fiber end face defect intelligent repair device, comprising a defect detection module, an intelligent decision module, a repair execution module, a motion positioning module and a closed-loop feedback module;

[0032] The defect detection module adopts a multi-modal detection structure composed of a microscopic imaging assembly and a spectral analysis assembly, for collecting defect images and spectral data of the optical fiber end face;

[0033] The intelligent decision module is connected with the defect detection module, and performs feature extraction and classification processing on the defect images and spectral data to generate repair parameters corresponding to the defect types;

[0034] The repair execution module includes switchable grinding units, laser repair units and cleaning units, and is connected with the intelligent decision module for performing repair operations according to the repair parameters;

[0035] The motion positioning module is used to adjust the relative position of the optical fiber and the repair execution module;

[0036] The closed-loop feedback module is connected with the defect detection module and the intelligent decision module respectively, and can feed back the detection result after repair to the intelligent decision module to adjust the repair parameters.

[0037] The microscopic imaging assembly of the defect detection module comprises a high-resolution CCD camera, a metallographic objective lens and a ring light source, the magnification of the metallographic objective lens is 50-200 times, and the wavelength range of the ring light source is 400-700 nm.

[0038] The defect detection module further comprises an image preprocessing assembly for noise reduction, contrast enhancement and edge sharpening processing of the collected defect image.

[0039] The spectral analysis unit: grating spectrometer, detection wave band 300-1100 nm, resolution 0.5 nm, fiber probe, diameter 1 mm, coaxial integration with the microscopic imaging unit, synchronous detection in the same field of view through a half-transmission half-reflection mirror.

[0040] The image preprocessing unit: embedded FPGA chip, processing speed 1 GB / s, integrated median filter, adaptive histogram equalization, contrast enhancement 200%, Sobel edge detection function, threshold 0-255 adjustable.

[0041] The intelligent decision module comprises a feature extraction submodule and a strategy generation submodule; the feature extraction submodule can extract the geometric parameters of the defect from the defect image and extract the material composition information of the defect from the spectral data; the strategy generation submodule matches the preset repair process database based on the geometric parameters and the material composition information to generate the repair parameters.

[0042] The geometric parameters include the length, width, depth and area of the defect, and the material composition information is used to distinguish between the contaminating defects and the structural defects.

[0043] Industrial-grade CPU + 16 GB memory, equipped with Linux real-time operating system;

[0044] Feature database: store the feature templates (including geometric parameter range, spectral characteristic value) of 20 types of defects (such as linear scratches, network cracks, oil pollution, etc.).

[0045] Decision-making process: adopt the combination logic of "rule tree + parameter mapping table", specifically:

[0046] Extract the geometric parameters (length L, width W, depth H) of the defect through image gray gradient calculation: H=k×(defect area gray value-normal area gray value), k is the calibration coefficient 0.02 μm / grey level);

[0047] Extract the spectral characteristic value (such as the reflectivity of oil stain at 550 nm R≤30%, normal quartz optical fiber R≈92%);

[0048] The matching feature database determines the defect type, and calls the corresponding repair parameter mapping table, as shown in Table 1.

[0049] Defect type Geometric parameter range Spectral feature Repair parameter Light scratch L≤10 μm, W≤2 μm R≈90% (no abnormality) Laser energy 5 μJ, spot diameter 8 μm, scanning 1 time Moderate scratch 10 μm < L ≤ 30 μm R≈85% (slightly low) Laser pretreatment (10 μJ + grinding (pressure 1 N) Oil pollution No fixed geometric shape R≤30% (550 nm) Isopropyl alcohol injection (0.5 μL) + air blowing (0.3 MPa)

[0050] Table 1

[0051] The grinding unit of the repair execution module includes a precision rotary motor, a grinding disc, and a pressure adjusting assembly. The diameter of the grinding disc is 10-50 mm, and the pressure adjusting range of the pressure adjusting assembly is 0.1-5 N.

[0052] The laser repair unit of the repair execution module adopts a nanosecond pulse laser, whose output wavelength is 1064 nm, pulse width is 10-100 ns, and energy adjusting range is 1-50 μJ.

[0053] The motion positioning module includes a three-dimensional translation stage and an optical fiber clamp. The displacement resolution of the three-dimensional translation stage is 0.1 microns, and the optical fiber clamp is adapted to optical fibers with diameters of 0.125 mm-0.9 mm.

[0054] The closed-loop feedback module is provided with a judgment threshold. When the repaired defect parameter is lower than the judgment threshold, it is determined that the repair is qualified. When the repaired defect parameter is higher than the judgment threshold, the intelligent decision-making module generates repair parameters again and controls the repair execution module to perform repair operation again.

[0055] The preset qualified threshold is: scratch L≤5 μm, W≤1 μm, H≤0.5 μm; pollution defect R≥90%; no crack.

[0056] Feedback logic: when the detection value after repair does not meet the standard, secondary repair is automatically triggered (the parameters are adjusted by 10%-20% based on the original parameters), and the maximum number of cycles is 3 (to avoid over-repair).

[0057] It also includes a human-computer interaction module connected with the intelligent decision-making module, which can display defect detection results, repair parameters and repair progress, and support manual modification of repair parameters.

[0058] Optical fiber clamping

[0059] The single-mode optical fiber to be repaired (diameter 0.125 mm, end face with scratch of length 20 μm and width 3 μm) is fixed on the optical fiber clamp, the optical fiber type parameters are set through the human-computer interaction module, the device automatically adjusts the clamping force to 1 N, and the three-dimensional translation stage drives the optical fiber to move to the initial detection position.

[0060] Defect detection

[0061] Microscopic imaging component starts: the annular light source light intensity is adjusted to 1500 lux, the CCD camera collects the fiber end surface image, the image preprocessing component first carries out 3*3 median filtering denoising, then enhances the contrast through adaptive histogram equalization, and finally uses the Sobel operator to extract the defect edge to obtain a clear scratch profile image.

[0062] Spectral analysis component starts: the defect area is scanned (integration time 50 ms) to obtain the reflectance spectrum curve of the 300-1100 nm waveband, which is compared with the spectrum curve of the standard clean fiber end surface to determine that it is a non-polluting structural defect.

[0063] Intelligent decision

[0064] The feature extraction sub-module calculates the scratch geometric parameters from the preprocessed image: length 20 pm, width 3 pm, depth 1.2 pm (calculated by image gray value gradient), and the spectral analysis confirms that there is no pollutant composition.

[0065] The strategy generation submodule matches the above parameters with the repair process database, and calls the scratch repair scheme: first laser pretreatment (energy 10 pJ, spot diameter 10 pm, scanning speed 50 pm / s), then fine grinding (pressure 1.5 N, rotation speed 1500 rpm, time 3 s), and finally cleaning treatment.

[0066] Repair execution

[0067] The motion positioning module drives the fiber end surface to align with the laser repair unit, and the laser scans the scratch area according to the preset path to remove the surface protrusions; then it switches to the grinding unit, and the grinding disc rotates under the control of the pressure to grind the defect area; finally, the cleaning unit sprays compressed air and a small amount of isopropyl alcohol to remove the grinding debris.

[0068] Closed-loop feedback verification

[0069] After the repair is completed, the defect detection module detects the fiber end surface again, and obtains the scratch length 3 pm, width 0.8 pm, and depth 0.3 pm, which are all below the determination threshold. The closed-loop feedback module determines that the repair is qualified, and the device displays "repair completed" through the human-computer interaction module.

[0070] Example one: repair of "moderate scratch + light oil stain" composite defect

[0071] Preconditions: the fiber to be repaired is a single-mode fiber (diameter 0.125 mm), the end surface has a scratch of L=25 pm and W=3 pm, and is covered with light oil stain.

[0072] Step 1: fiber clamping and initialization

[0073] The operator puts the optical fiber into the clamp, clicks the "start" button on the touch screen, and the device automatically executes:

[0074] a. The clamp torque motor starts, and the clamping force gradually increases from 0 to 1 N (feedback stable through force sensor);

[0075] b. The three-dimensional translation stage moves the optical fiber to the "initial detection position" (X=25 mm, Y=25 mm, Z=10 mm, preset coordinates);

[0076] c. The ring light source is turned on (intensity 1500 lux), and the CCD camera and spectrometer are preheated (3 seconds).

[0077] Step 2: Defect detection and analysis

[0078] Image acquisition and preprocessing:

[0079] a. The CCD camera takes the end face image (exposure time 50 ms), and the original image is processed by FPGA: 3x3 median filter to remove noise → adaptive histogram equalization to enhance scratch edge → Sobel operator to extract defect profile;

[0080] b. The system automatically measures the scratch geometric parameters: L=25 μm, W=3 μm, H=1.2 μm (calculated by gray gradient: H=0.02x(180-120)=1.2 μm).

[0081] Spectral analysis:

[0082] a. The spectrometer scans the defect area (integration time 50 ms), and obtains the 300-1100 nm reflectance spectrum, where R=25% at 550 nm (lower than the standard value 92%), which is determined as "oil pollution";

[0083] b. The detection data is packaged and sent to the intelligent decision module (data volume about 2 MB, transmission time <1 second).

[0084] Step 3: Intelligent decision and parameter generation

[0085] Decision module matches feature database:

[0086] a. Scratch parameters (L=25 μm) match "moderate scratch" template, and oil (R=25%) match "mild pollution" template;

[0087] b. Call composite repair strategy: clean first (remove oil to avoid pollution spread during grinding) → laser pretreatment (melt the scratch protrusion) → fine grinding last (reduce surface roughness);

[0088] c. Generate specific parameters (based on Table 1 mapping + dynamic correction):

[0089] Cleaning: Isopropyl alcohol 0.5 μL, air blow 0.3 MPa, 1 second;

[0090] Laser: Energy 12 μJ (base value 10 μJ + 20% correction), spot 10 μm, scanning speed 50 μm / s, path is reciprocating 2 times along the scratch direction;

[0091] Grinding: Pressure 1.2 N (base value 1 N + 20% correction), rotation speed 1500 rpm, time 3 seconds.

[0092] Step 4: Repair execution

[0093] Cleaning operation:

[0094] a. The three-dimensional translation table drives the optical fiber to move to the "cleaning position" (X = 25 mm, Y = 35 mm), and the cleaning unit nozzle is aligned with the defect area (distance 5 mm);

[0095] b. The spray valve releases 0.5 μL of isopropyl alcohol (for 50 ms), and after an interval of 1 second, the air pump is started (0.3 MPa, for 1 second), and the residual liquid is removed.

[0096] Laser repair:

[0097] a. The rotating shaft system switches the laser head to the working position (positioning deviation ≤0.3 μm), and the three-dimensional translation table fine-tunes the position of the optical fiber, so that the center of the laser spot is aligned with the starting point of the scratch;

[0098] b. The laser starts according to the preset parameters: 12 μJ energy, 10 μm spot, scanning at a speed of 50 μm / s along the scratch direction (reciprocating 2 times), and real-time monitoring of laser power stability (fluctuation ≤±5%).

[0099] Grinding repair:

[0100] a. The rotating shaft system switches to the grinding unit, and the grinding disc rotation speed rises to 1500 rpm (stabilization time 1 second);

[0101] b. The Z-axis slowly descends, so that the grinding disc contacts the end face of the optical fiber, and the pressure sensor feedback pressure reaches 1.2 N after 3 seconds, during which the X / Y axis is fine-tuned according to the "spiral trajectory" (range ±5 μm), to ensure uniform grinding.

[0102] Step 5: Closed-loop feedback verification

[0103] The three-dimensional translation table moves the optical fiber back to the detection position, and the detection process of step 2 is repeated:

[0104] a. Image detection: scratch L = 4 μm, W = 0.8 μm, H = 0.3 μm (all below the qualified threshold);

[0105] b. Spectral detection: R = 93% at 550 nm (above the qualified threshold of 90%);

[0106] c. Closed-loop module determines "repair qualified", generates detection report (including parameter comparison chart before and after repair).

[0107] Step 6: End and unload

[0108] Device automatic reset: repair head returns to the initial position, translation table returns to zero, light source is turned off;

[0109] Touch screen displays "repair complete", clamp is automatically released (clamping force is reduced to 0), and the operator removes the optical fiber.

[0110] Through the above steps, the full-automatic repair of the composite defect is realized, and the comparison of key indicators is as follows:

[0111] Before repair: surface roughness Ra = 1.5 μm, insertion loss 0.8 dB, and defect recognition accuracy 100%;

[0112] After repair: surface roughness Ra = 0.08 μm, insertion loss 0.1 dB, and repair qualified rate 100% (single repair);

[0113] Efficiency: total time consumption is about 53 seconds (including clamping), which is increased by 66% compared with the traditional manual repair (about 180 seconds), and manual judgment of defect type or intervention in the repair process is not required.

[0114] The above only describes certain exemplary embodiments of the present application in a descriptive manner, and it is needless to say that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present application. Therefore, the above drawings and descriptions are illustrative in nature and should not be understood as limiting the scope of protection of the claims of the present application.

Claims

1. An optical fiber end face defect intelligent repairing device, characterized in that, The module comprises a defect detection module, an intelligent decision module, a repair execution module, a motion positioning module and a closed-loop feedback module. The defect detection module adopts a multi-modal detection structure composed of a microscopic imaging assembly and a spectral analysis assembly to collect defect images and spectral data of the fiber end face. The intelligent decision module is connected with the defect detection module to perform feature extraction and classification processing on the defect images and spectral data, and generate repair parameters corresponding to the defect type. The repair execution module comprises switchable grinding units, laser repair units and cleaning units, and is connected with the intelligent decision module to perform repair operations according to the repair parameters. The motion positioning module is used to adjust the relative position of the fiber and the repair execution module. The closed-loop feedback module is connected with the defect detection module and the intelligent decision module to feed back the detection results after repair to the intelligent decision module to adjust the repair parameters.

2. The optical fiber end face defect intelligent repairing device according to claim 1, characterized in that, The microscopic imaging assembly of the defect detection module comprises a high-resolution CCD camera, a metallographic objective lens and a ring light source, the magnification of the metallographic objective lens is 50-200 times, and the wavelength range of the ring light source is 400-700 nm.

3. The intelligent optical fiber end face defect repairing device according to claim 1, characterized in that, The defect detection module further comprises an image preprocessing assembly for noise reduction, contrast enhancement and edge sharpening processing of the collected defect images.

4. The intelligent optical fiber end face defect repairing device according to claim 1, characterized in that, The intelligent decision module comprises a feature extraction submodule and a strategy generation submodule; the feature extraction submodule can extract the geometric parameters of the defect from the defect image and the material composition information of the defect from the spectral data; The strategy generation submodule matches the pre-set repair process database based on the geometric parameters and the material composition information to generate repair parameters.

5. The intelligent optical fiber end face defect repairing device according to claim 4, characterized in that, The geometric parameters include the length, width, depth and area of the defect, and the material composition information is used to distinguish between contaminating defects and structural defects.

6. The intelligent optical fiber end face defect repairing device according to claim 1, wherein, The grinding unit of the repair execution module comprises a precision rotary motor, a grinding disc and a pressure adjusting assembly, the diameter of the grinding disc is 10-50 mm, and the pressure adjusting range of the pressure adjusting assembly is 0.1-5 N.

7. The intelligent optical fiber end face defect repairing device according to claim 1, wherein, The laser repair unit of the repair execution module adopts a nanosecond pulse laser with an output wavelength of 1064 nm, a pulse width of 10-100 ns and an energy adjusting range of 1-50 μJ.

8. The intelligent optical fiber end face defect repairing device according to claim 1, wherein, The motion positioning module comprises a three-dimensional translation stage and a fiber clamp, the displacement resolution of the three-dimensional translation stage is 0.1 microns, and the fiber clamp is adapted to fibers with diameters of 0.125 mm-0.9 mm.

9. The intelligent optical fiber end face defect repairing device according to claim 1, wherein, The closed-loop feedback module is provided with a judgment threshold, when the defect parameters after repair are lower than the judgment threshold, it is determined that the repair is qualified; when the defect parameters after repair are higher than the judgment threshold, the intelligent decision module generates repair parameters again and controls the repair execution module to perform repair operations again.

10. The intelligent optical fiber end face defect repairing device according to claim 1, characterized in that, It further comprises a human-computer interaction module connected with the intelligent decision module, which can display the defect detection results, repair parameters and repair progress, and support manual modification of repair parameters.

Images