Optical fiber coupling module abnormality detection method, device, medium and equipment

Abstract

The present application provides a method, device, medium and equipment for detecting abnormalities in a fiber coupling module, which belongs to the field of optical device detection technology. The fiber coupling module includes a multi-channel laser coupling unit, each laser coupling unit includes a plurality of groups of laser units, and the method includes: obtaining the optical power value of each group of laser units; generating an optical power sequence of each laser coupling unit based on the optical power value; performing an abnormal analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes that the laser coupling unit is normal or the laser coupling unit is abnormal. The present application can greatly improve the accuracy and efficiency of identifying abnormal laser coupling units.

Description

Technical Field

[0001] The present application relates to the technical field of optical device detection, and in particular to a method, device, medium and equipment for detecting abnormality of an optical fiber coupling module. Background Art

[0002] In the optical fiber communication system, the optical fiber coupling module is a key component responsible for coupling multiple laser beams into the optical fiber. Its performance directly affects the stability and reliability of the entire communication system. The optical fiber coupling module usually contains multiple optical components, such as COS, collimator, lens, reflector, etc. The precise matching and working state of these components are crucial to the quality of signal transmission.

[0003] Before the fiber coupling module leaves the factory, it needs to be tested for abnormalities to prevent abnormalities in the optical components of the fiber coupler. Currently, the abnormality detection of the optical components in the fiber coupling module mainly relies on manual measurement and experience judgment.

[0004] Specifically, operators usually use equipment such as optical power meters to manually measure the output optical power of each optical component, and then compare it with the preset standard value to determine whether there is an abnormality. However, this method has obvious limitations. First, the efficiency of manual measurement is low, especially in the case of large-scale production, manual detection is difficult to meet the needs of rapid detection. Secondly, since manual measurement is greatly affected by the operator's experience and skills, it is easy to misjudge or miss the detection. In addition, manual measurement is difficult to accurately locate the fault point. Once a problem is found, it is difficult to quickly find which specific component has a problem, which affects the efficiency of maintenance and adjustment of the fiber coupling module.

[0005] Therefore, it is particularly important to propose a method that can automatically and accurately identify and locate abnormal parts of fiber coupling modules. Summary of the invention

[0006] Based on this, the purpose of the present application is to provide a method, device, medium and equipment for detecting abnormality of a fiber coupling module, so as to improve the accuracy and efficiency of abnormality detection of the fiber coupling module.

[0007] In a first aspect of the present application, a method for detecting abnormality of an optical fiber coupling module is provided, wherein the optical fiber coupling module includes at least one laser coupling unit, each laser coupling unit includes multiple groups of laser units, and the method includes:

[0008] Obtain the optical power value of each group of laser units;

[0009] Generate an optical power sequence of each laser coupling unit based on the optical power value;

[0010] An abnormality analysis is performed on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes that the laser coupling unit is normal or the laser coupling unit is abnormal.

[0011] Optionally, performing abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes:

[0012] Analyzing the difference characteristics between any two optical power values ​​in the optical power sequence;

[0013] An abnormality analysis is performed on the laser coupling unit based on the difference feature to obtain the first analysis result.

[0014] Optionally, performing abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes:

[0015] Comparing the optical power curve generated by the optical power sequence with a preset standard curve to obtain a comparison result;

[0016] An abnormality analysis is performed on the laser coupling unit based on the comparison result to obtain the first analysis result.

[0017] Optionally, performing an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes: inputting the optical power sequence into a preset first detection model to perform an abnormality analysis on the laser coupling unit to obtain a first analysis result.

[0018] Optionally, before inputting the optical power sequence into a preset first detection model to perform abnormal analysis on the laser coupling unit, the method further includes: training the first detection model; and training the first detection model includes:

[0019] Acquire an initial sample set, wherein the initial sample set includes an optical power sequence consisting of numerically quantized optical power values ​​under normal and abnormal conditions;

[0020] Preprocessing the initial sample set to form a standard training set and a standard validation set for model training;

[0021] Iteratively training the first initial detection model using the standard training set, and adjusting the model parameters during each iterative training process to obtain the first initial detection model after completing the iterative training;

[0022] The first initial detection model after iterative training is verified by the standard verification set to obtain a model verification result. When the model verification result indicates that the first initial detection model after iterative training can accurately distinguish the optical power sequences under normal and abnormal conditions, the first initial detection model after iterative training is used as the first detection model.

[0023] Optionally, when the laser coupling unit is abnormal, the first analysis result further includes a unit number of the abnormal laser coupling unit;

[0024] When the first analysis result is that the laser coupling unit is abnormal, the method further includes:

[0025] Define the abnormal optical power sequence of the laser coupling unit as the target optical power sequence;

[0026] Inputting the first analysis result and the target optical power sequence into a preset second detection model to determine a second analysis result of the abnormality of the laser coupling unit;

[0027] The second analysis result includes the unit number of the abnormal laser unit and the abnormal component of the abnormal laser unit; the abnormal component includes one or more of a COS, a fast axis collimator, a slow axis collimator, a reflector and a focusing lens.

[0028] Optionally, obtaining the optical power value emitted by each group of laser units includes:

[0029] Measuring the optical power signal of each group of laser units;

[0030] The optical power signal is numerically quantified based on a measurement error rate to obtain an optical power value of the laser unit.

[0031] Optionally, before obtaining the optical power value emitted by each group of laser units, the method further includes:

[0032] Acquiring an abnormal state of the optical fiber coupling module;

[0033] When the optical fiber coupling module is abnormal, the optical power value of each group of laser units is obtained.

[0034] In a second aspect of the present application, a device for detecting abnormality of an optical fiber coupling module is provided, wherein the optical fiber coupling module includes at least one laser coupling unit, each laser coupling unit includes multiple groups of laser units, and the device includes:

[0035] An optical power value extraction unit, used to obtain the optical power value of each group of laser units;

[0036] An optical power sequence processing unit, used for generating an optical power sequence of each laser coupling unit based on the optical power value;

[0037] The abnormality analysis unit is used to perform abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes that the laser coupling unit is normal or the laser coupling unit is abnormal.

[0038] In a third aspect of the present application, a computer-readable storage medium is provided, on which executable instructions are stored. When the executable instructions are executed by a processor, the processor executes the method described in any embodiment of the present application.

[0039] In a fourth aspect of the present application, an electronic device is provided, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the method described in any one of the embodiments of the present application.

[0040] The optical fiber coupling module abnormality detection method, device, medium and equipment in the present application can quickly identify whether each laser coupling unit in the optical fiber coupling module has an abnormality by measuring the optical power value of each group of laser units, generating an optical power sequence of the corresponding laser coupling unit, and performing abnormality analysis based on the optical power sequence. Compared with manual detection, the present application can greatly improve the accuracy and efficiency of identifying abnormal laser coupling units. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope of the present application.

[0042] Figure 1 A schematic diagram of a process of detecting abnormality of an optical fiber coupling module in one embodiment;

[0043] Figure 2 is a schematic structural diagram of an optical fiber coupling module in one embodiment;

[0044] Figure 3 is a schematic diagram of an image formed based on an optical power sequence in the first embodiment;

[0045] Figure 4 is a schematic diagram of an image formed based on an optical power sequence in the second embodiment;

[0046] Figure 5 is a schematic diagram of an image formed based on an optical power sequence in the third embodiment;

[0047] Figure 6 is a schematic diagram of an image formed based on an optical power sequence in the fourth embodiment;

[0048] Figure 7 is a schematic diagram of an image formed based on an optical power sequence in the fifth embodiment;

[0049] Figure 8 is a schematic diagram of an image formed based on an optical power sequence in the sixth embodiment;

[0050] Fig. 9 is a schematic diagram of the structure of an integrating sphere in one embodiment;

[0051] Fig.10 is a schematic diagram of a first detection model in an embodiment;

[0052] Fig.11 A schematic diagram of the structure of an abnormality detection device for an optical fiber coupling module in one embodiment;

[0053] Fig.12 FIG. 4 is a structural block diagram of an electronic device in an embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] All terms (including technical and scientific terms) used in this application have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used here should be interpreted as having a meaning consistent with the context of this specification, and should not be interpreted in an idealized or overly rigid manner.

[0056] For example, the terms "first", "second", etc. used in this application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, without departing from the scope of this application, a first analysis result may be referred to as a second analysis result, and similarly, a second analysis result may be referred to as a first analysis result. Both the first analysis result and the second analysis result are analysis results, but they are not the same analysis result.

[0057] For another example, the terms “include”, “comprising”, etc. used in this application indicate the existence of features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0058] In one embodiment, a method for detecting abnormality of an optical fiber coupling module is provided. Figure 1 As shown, the method includes:

[0059] Step 110, obtaining the optical power value emitted by each group of laser units.

[0060] In the present application, the optical power value is the signal value emitted after applying an excitation signal to the laser unit. By applying an excitation signal to the COS of the laser unit, the COS emits a laser, and the laser is coupled into the optical fiber through the laser fast axis collimator, slow axis collimator, reflector and focusing lens in sequence. The optical power value is the signal value collected after the laser is coupled into the optical fiber. The fiber coupling module includes a multi-channel laser coupling unit, and each laser coupling unit includes multiple groups of laser units. Each group of laser units includes COS (Chip on Substrate, laser chip), FAC (Fast Axis Collimator, fast axis collimator), SAC (Slow Axis Collimator, slow axis collimator), reflector (Mirror), FOC focusing lens group and other components. The coupling process of each group of laser units is: the laser emitted by COS is first collimated by the fast axis collimator (FAC), then collimated by the slow axis collimator (SAC), reflected by the reflector (mirror) to the focusing lens group (FOC) for collimation, and enters the optical fiber. Among them, after the laser light emitted by each group of laser units undergoes the above-mentioned collimation and reflection process, before entering the optical fiber, the laser light emitted by each group of laser units will be collimated and converged by a focusing lens again and enter the optical fiber.

[0061] Combination Figure 2 As shown, it shows a schematic structural diagram of a laser coupling unit in a fiber coupling module. Figure 2 The number of groups of laser units included in the single-channel laser coupling unit shown in the figure is 5. It can be understood that the number of groups of laser units included in each channel of the laser coupling unit in the optical fiber coupling module can also be any other number of groups, such as the number of groups of laser units included in each channel of the laser coupling unit can be any number of groups such as 10 groups, 12 groups, 15 groups, 20 groups, etc. For example, the optical fiber coupling module includes 6 channels of laser coupling units, and each channel of the laser coupling unit includes 21 groups of laser units.

[0062] Take one group of laser units as an example, it includes: COS1, fast axis collimator 2, slow axis collimator 3, reflector 4, ceramic heat sink 5, anti-reflection film 6, first focusing lens 7 and second focusing lens 8. Among them, multiple groups of laser units in each laser coupling unit can share the same anti-reflection film 6, first focusing lens 7 and second focusing lens 8. COS1, fast axis collimator 2, slow axis collimator 3 and reflector 4 in each group of laser units are arranged in sequence to collimate / reflect the laser emitted by COS1.

[0063] The excitation signal applied to the fiber coupling unit / laser coupling unit / laser unit may be a current signal or a voltage signal. Specifically, the excitation signal may be a rated current signal adapted to the fiber coupling module. When the excitation signal is applied to COS1, the laser emitted by COS1 passes through the fast axis collimator 2, the slow axis collimator 3, the reflector 4, the anti-reflection sheet 6, the first focusing lens 7 and the second focusing lens 8 in sequence, and the laser emitted from the second focusing lens 8 finally enters the optical fiber 9, on which a ceramic ferrule 10 may be mounted.

[0064] The laser emitted by the laser unit can be measured by a related optical power measuring device to obtain the optical power value of the laser unit. The optical power value can be the actual optical power value emitted by the laser unit, or it can be the optical power value after processing and quantification. The value can be a pre-processed value to avoid interference with subsequent detection due to missing values ​​in the optical power sequence that has not been pre-processed or the values ​​in the optical power sequence are non-standardized values, resulting in inaccurate abnormal analysis results. Optionally, the optical power measuring device can be an optical power meter such as a handheld optical power meter, a desktop optical power meter, or an integrating sphere.

[0065] In this embodiment, the excitation signal can be applied to each group of laser units independently in turn to obtain the optical power value of each group of laser units respectively, and finally the optical power value measurement of all laser units in each laser coupling unit is completed.

[0066] Step 120: Generate an optical power sequence for each laser coupling unit based on the optical power value.

[0067] In this embodiment, the optical power sequence is a set generated by arranging the optical power values ​​of all laser units in each laser coupling unit according to a preset arrangement rule. The arrangement rule can be determined according to the position sequence of the laser units in the laser coupling unit, for example, the laser units of each group can be arranged in sequence according to the position in the corresponding laser coupling unit, so that the optical power value of the laser unit at the i-th position is at the i-th position in the optical power sequence.

[0068] Specifically, each optical power value in the optical power sequence is sorted according to the position of the laser coupling unit in each group of laser units under the corresponding number of paths. The optical power sequence X of the i-th laser coupling unit in the optical fiber coupling module i It can be expressed as X i ={x i1 ,x i2 ,x i3 ……x in}.x ijrepresents the optical power value of the jth group of laser units in the i-th laser coupling unit. For example, the jth group of laser units can be as follows Figure 2 The laser coupling units shown are in the jth group of laser units from left to right.

[0069] For a fiber coupling module in a normal state, in the same laser coupling unit, the optical power values ​​corresponding to the laser units at different positions (different groups) are not necessarily the same; in laser coupling units of different paths, the optical power values ​​corresponding to the laser units at the same position are not necessarily the same.

[0070] The optical power values ​​of the optical power sequence show a trend of increasing first and then decreasing. The closer the laser units are to the two sides of the laser coupling unit, the smaller the optical power value is, and the closer the laser units are to the middle position, the larger the optical power value is. The curve generated by the optical power values ​​of the laser coupling unit presents a parabola-like shape as a whole. For example, for a laser coupling unit containing 21 groups of laser units, a group of optical power sequences formed is: X={375,383,388,394,398,402,406,409,411,412,413,412,411,410,406,403,400,397,390,383,377}, in which the optical power values ​​have been quantized.

[0071] Step 130 , performing abnormal analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result.

[0072] In this embodiment, the first analysis result includes whether the laser coupling unit is normal or abnormal. For the generated optical power sequence, it is possible to determine whether the corresponding laser coupling unit is abnormal according to the numerical value of each optical power value in the optical power sequence and the numerical change trend, and output the first analysis result. The first analysis result includes information that the laser coupling unit is normal or abnormal. When the laser coupling unit is abnormal, the first analysis result further includes the unit number of the abnormal laser unit, and the unit number is used to uniquely identify the laser unit in the corresponding laser coupling unit.

[0073] Among them, when the numerical value of one or more optical power values ​​is obviously larger / smaller than the standard numerical value presented by the normal laser unit at the corresponding position, and / or the numerical change trend between the various optical power values ​​is inconsistent with the standard numerical change trend presented by the normal laser coupling unit, it can be determined that the laser coupling unit is abnormal.

[0074] Furthermore, for the optical fiber coupling module determined to have an abnormal laser coupling unit, based on the difference between the optical power value of the laser unit and the standard value, or the difference between the optical power sequence and the standard trend, the specific optical power value with the abnormality is located, and it is determined that the laser unit corresponding to the optical power value has an abnormality, and the unit number of the located abnormal laser unit is written into the first analysis result.

[0075] Specifically, a standard sequence corresponding to the coupler number of the laser coupling unit to be detected can be pre-set, and the standard sequence is the standard value presented by each group of laser units in the laser coupling unit under the corresponding number of paths in the optical fiber coupling module. The optical power sequence is compared with the standard sequence, and the difference and / or difference between the two are detected. When the difference exceeds the preset difference threshold, and / or the difference exceeds the preset difference threshold, it is determined that the laser coupling unit is abnormal. Specifically, a corresponding difference threshold can be set for the standard value at each position in the standard sequence, and the difference at the corresponding position of the two is compared with the corresponding difference threshold. The difference thresholds corresponding to different laser units are not necessarily the same.

[0076] The optical fiber coupling module abnormality detection method in this application measures the optical power value of each channel, generates the optical power sequence of the corresponding laser coupling unit, and performs abnormality analysis based on the optical power sequence, which can quickly identify whether each laser coupling unit in the optical fiber coupling module is abnormal. Compared with manual detection, this application can greatly improve the accuracy and efficiency of abnormal laser coupling unit identification.

[0077] In one embodiment, step 110 includes: measuring the optical power signal of each group of laser units; and numerically quantifying the optical power signal based on the measurement error rate to obtain the optical power value of the laser unit.

[0078] The measurement error rate is the error rate of the optical power measurement device itself, which is used to reflect the measurement accuracy of the optical power measurement device. For example, if the optical power measurement device is an integrating sphere, the optical power signal emitted by each group of laser units is measured by the integrating sphere; the optical power signal is numerically quantified based on the error rate of the integrating sphere to form an optical power value. The optical power signal can be specifically the original optical power value actually measured.

[0079] like Fig. 9 The figure shows a schematic diagram of the structure of an integrating sphere in an embodiment, which includes an external integrating sphere 91 , an internal diffuse reflection surface 92 , an adjustable aperture 93 , a first receiver 94 , a second receiver 95 , and an adjustable optical distance base 96 .

[0080] During the measurement process, the light emitted by the laser unit is introduced into the external integrating sphere 91 through the optical fiber and the adjustable aperture 93. The light is continuously reflected and scattered on the internal diffuse reflection surface 92 of the sphere, and the light is gradually evenly distributed in the sphere. The first receiver 94 and the second receiver 95 are respectively placed at different positions of the integrating sphere to measure the optical power in the uniform light field and obtain the above-mentioned optical power signal.

[0081] The adjustable aperture 93 controls the amount of light entering the integrating sphere by adjusting its size, thereby adapting to light sources of different intensities; the adjustable optical distance base 96 allows the user to adjust the distance between the receiver and the integrating sphere according to actual needs to ensure measurement accuracy.

[0082] The error rate of the integrating sphere is used to reflect the measurement accuracy of the integrating sphere. For example, the error rate of the integrating sphere is ±1%, and the maximum power it can measure is 10W. Based on the error rate and the maximum measured power, it can be obtained that the error of the integrating sphere is ±100mW. The measured original optical power value is quantified by the error, and the value obtained after quantification is used as the optical power value. The quantized value can be an integer multiple of the error of the integrating sphere. The quantization method can be specifically to divide the original optical power value by the error of the integrating sphere, round the quotient obtained by the division, and the obtained integer value is used as the optical power value. The rounding method can be any suitable rounding method such as rounding, rounding up, rounding down, etc.

[0083] For example, if the laser unit measured by the integrating sphere is 4330mW, then the quantization is performed based on the error ±100mW, and the quantized value is 44 power units, and the value 44 can be used as the optical power value. By numerically quantizing the optical power signal based on the error rate of the integrating sphere, the convenience of calculation can be improved, and the error in the actual measurement result caused by the error rate of the integrating sphere can be avoided, which will affect the subsequent judgment of abnormality of the optical fiber coupling module.

[0084] For example, for the above optical power sequence X={375,383,388,394,398,402,406,409,411,412,413,412,411,410,406,403,400,397,390,383,377}, the optical power signal (in mW) corresponding to each optical power data measured by the integrating sphere is X'={37 The error rate of the integrating sphere is ±0.1%, and the maximum power it can measure is 10W. Based on the error rate and the maximum measured power, the error of the integrating sphere is ±10mW. By adopting the rounding-up quantization method, the above optical power sequence X can be obtained.

[0085] By using the error rate of the integrating sphere to quantify the optical power signal and form the optical power value, the deviation of subsequent abnormal analysis caused by the error of the integrating sphere can be avoided, and the accuracy of the analysis results can be improved. In addition, by performing numerical quantization, the measured optical power signal can be simplified, the complexity of the optical power value can be reduced, and the efficiency of subsequent abnormal analysis can be improved.

[0086] In one embodiment, before step 110, the method further includes: obtaining an abnormal state of the optical fiber coupling module; when the optical fiber coupling module is abnormal, obtaining an optical power value of each group of laser units.

[0087] In this embodiment, before step 110, it is possible to first detect whether the optical fiber coupling unit has an abnormality, and only perform abnormality analysis on the abnormal optical fiber coupling module.

[0088] Specifically, obtaining the abnormal state of the optical fiber coupling module includes: measuring the total power of the optical fiber coupling module, and when the total power does not match the standard power, determining that the optical fiber coupling module is abnormal. When there is an abnormality, step 110, step 120 and step 130 are performed in sequence.

[0089] By applying rated excitation signals to all laser units in the fiber coupling module at the same time, the light emitted by all laser units is coupled into the optical fiber and emitted out, and the optical power emitted from the optical fiber is measured by the corresponding optical power measurement equipment, which is the total power.

[0090] The electronic device is pre-configured with the standard power of the fiber coupling module. For the total power measured, it can be compared with the standard power to calculate the deviation from the standard power. When the deviation is lower than the preset deviation threshold, it means that the total power does not match the standard power, and the fiber coupling module is judged to be abnormal. When the deviation between the total power and the standard power is less than the deviation threshold, it means that the total power matches the standard power, indicating that the fiber coupling module is normal.

[0091] In this embodiment, the total power can be measured to identify whether the fiber coupling module is normal based on the total power and the standard power. When the deviation between the total power and the standard power meets the preset conditions, it means that the fiber coupling module is normal, and there is no need to perform steps 110, 120, and 130, which reduces the workload of abnormal identification of the fiber coupling module. When it is identified that the total power does not match the standard power, it means that the fiber coupling module is abnormal. At this time, it is impossible to locate which laser coupling unit has an abnormality, so steps 110, 120, and 130 can be performed sequentially to further identify which group or groups of laser units in which one or more laser coupling units have an abnormality.

[0092] For example, for a fiber coupling module composed of 6 laser coupling units, each of which has 12 groups of laser units, the total power formed by the 6×12 groups of laser units can be measured first to analyze whether the total power is abnormal. If there is no abnormality, it can be directly determined that the 6×12 groups of laser units are normal. If there is an abnormality in the total power, the above steps 110 to 130 can be executed to obtain the optical power values ​​of each group of laser units under each laser coupling unit, and perform abnormal analysis based on the optical power sequence formed.

[0093] In one embodiment, step 130 includes: analyzing a difference feature between any two optical power values ​​in the optical power sequence, and performing an abnormality analysis on the laser coupling unit based on the difference feature to obtain a first analysis result.

[0094] Among them, the difference feature represents the difference feature between two optical power values ​​in the same group of optical power sequences. Any two optical power values ​​can be the optical power values ​​of two adjacent groups of laser units, or can be the optical power values ​​of two non-adjacent groups of laser units. The difference feature includes a time difference feature and / or a mathematical statistical feature. The electronic device can also set the standard features corresponding to each laser coupling unit. In the optical fiber coupling module, the standard features corresponding to laser coupling units of different numbers are not necessarily the same. When the optical fiber coupling module contains N laser coupling units, N groups of standard features are set for the N laser coupling units, that is, corresponding to each group of standard features, at least one coupler number of the corresponding laser coupling unit is established to generate a corresponding relationship. For the laser coupling unit to be detected, the corresponding standard feature is obtained based on the corresponding relationship. The coupler number is used to uniquely identify the laser coupling unit in a fiber coupling module.

[0095] The extracted difference features such as the timing difference features and / or mathematical statistical features are compared with the standard features to detect whether the difference features and the standard features meet the consistency. If the consistency is met, the corresponding laser coupling unit is determined to be normal, otherwise the laser coupling unit is determined to be abnormal. The consistency can be measured based on the difference between the difference features and the standard features. For the laser coupling units determined to be abnormal, the specific values ​​that do not meet the consistency in the timing difference features and / or mathematical statistical features can be further compared, and the corresponding abnormal laser unit can be located based on the values.

[0096] The time series difference represents the difference characteristics between two adjacent optical power values ​​in the optical power sequence, which can be the difference between adjacent optical power values, and the differential sequence formed based on the difference. The optical power values ​​can capture the slight changes between the optical power values ​​in the optical power sequence, and calculate the difference between the differential sequence and the preset standard differential sequence, and determine whether there is an abnormality based on the difference. For example, when the difference exceeds the corresponding threshold, it is determined that the laser coupling unit is abnormal.

[0097] Mathematical statistical features represent the features presented by all the optical power values ​​in the optical power sequence. They include one or more statistical values ​​such as the average value, standard deviation, variance, maximum value, minimum value, kurtosis, skewness, etc. corresponding to each optical power value in the optical power sequence calculated based on a series of mathematical statistical rules. These mathematical statistical features can reflect the amplitude, frequency and stability of optical power changes. Similarly, the difference between the calculated mathematical statistical features and the preset standard features can be used to determine whether an abnormality exists based on the difference. For example, when the difference exceeds the corresponding threshold, it is determined that the laser coupling unit is abnormal.

[0098] In one embodiment, trend analysis is performed on the obtained timing differential characteristics and / or mathematical statistical characteristics, including analyzing the long-term trend and periodic changes of the optical power sequence based on one or more methods such as linear regression and moving average, to identify whether the fiber coupling module / laser coupling unit / laser unit has an abnormal trend.

[0099] Among them, in the fiber coupling module, the standard features presented by the standard sequences corresponding to different laser coupling units are not necessarily the same. The electronic device pre-establishes the standard features corresponding to each laser coupling unit in the fiber coupling module, and the standard features can be features extracted from the optical power sequence of the identified normal laser coupling unit.

[0100] For example, a normal laser coupling unit of 21 groups of laser units in a certain channel has an optical power sequence of: Xa={375,383,388,394,398,402,406,409,411,412,413,412,411,410,406,403,400,397,390,383,377}, or Xa is the above standard sequence.

[0101] For the five optical fiber coupling modules, the optical power sequences formed by the laser coupling units in the same path as Xa are recorded as Xb, Xc, Xd, Xe, and Xf respectively. Among them, Xb={375,383,388,394,398,402,406,409,411,412,413,412,411,410,407,405,403,400,396,391,387}, Xc={387,391,396,400,403,405,407,410,411,412,413,412,411,409,406,402,398,394,388,383,375}, Xd={332,339,344,348,352,356,359,362,364,3 65,365,365,364,363,359,357,354,352,346,339,334}, ,403,400,397,390,383,377},

[0102] By performing differential feature analysis, it can be identified that the optical power values ​​of the 16th to 21st groups of paths in Xb are relatively high, and it is determined that there is an abnormality; it can be identified that the optical power values ​​of the 1st to 6th groups of paths in Xc are relatively high, and it is determined that there is an abnormality; it can be identified that the optical power values ​​of the 1st to 21st groups of paths in Xd are generally low, and it is determined that there is an abnormality; it can be identified that the optical power value of the 12th group of paths in Xe is relatively low, and it is determined that there is an abnormality; it can be identified that the optical power value of the 12th group of paths in Xf is relatively low, and it is determined that there is an abnormality.

[0103] In this embodiment, by performing abnormality identification based on the difference features extracted from the optical power sequence, the accuracy of abnormality identification can be improved.

[0104] In one embodiment, step 130 includes: comparing an optical power curve generated by the optical power sequence with a preset standard curve to obtain a comparison result, and performing an abnormality analysis on the laser coupling unit based on the comparison result to obtain a first analysis result.

[0105] In this embodiment, the standard curve may be a curve generated based on the above standard sequence. The standard curve is a curve presented by the optical power sequence corresponding to a normal laser coupling unit. Laser coupling units of different numbers correspond to different standard curves.

[0106] Based on the laser coupling unit to be identified, the corresponding standard curve is determined based on the number of paths of the laser coupling unit, the optical power curve generated by the optical power sequence to be identified is compared with the standard curve, the difference between the two is analyzed, and whether the laser coupling unit is normal is determined based on the difference. When it is determined that there is an abnormality, the laser unit with the abnormality is further located based on the difference.

[0107] For example, if Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 As shown, Figure 3 The standard curve formed by the above-mentioned Xa; Figure 4~Figure 8 are the optical power curves formed by the above Xb~Xf. Figure 4~Figure 8 By comparing the curve in with the standard curve, it can be determined that the laser coupling unit corresponding to Xb~Xf is abnormal, and the abnormal laser unit can be further located.

[0108] By forming an optical power curve for abnormal analysis, the abnormal laser unit can be quickly and intuitively located.

[0109] In one embodiment, step 130 further includes: inputting the optical power sequence into a preset first detection model to perform abnormality analysis on the laser coupling unit to obtain a first analysis result.

[0110] In this embodiment, the first detection model is a model for abnormal identification of the laser coupling unit, and the model can be any suitable neural network model such as a multi-layer perceptron (MLP), a convolutional neural network (CNN), a recurrent neural network (RNN), etc. Specifically, the first detection model can be a preset decision tree model, such as a CART (Classification and Regression Tree) model.

[0111] Optionally, the electronic device pre-prepares a positive sample set formed according to a normal optical power sequence of a laser coupling unit, and a negative sample set formed according to an optical power sequence of a laser coupling unit with an abnormality. Based on the positive sample set and the negative sample, a training set for model training and a validation set for validating the trained model are jointly constructed. Both the validation set and the training set contain an appropriate number of positive samples and negative samples. Each positive sample and negative sample is annotated with a corresponding abnormal result, the abnormal result of the positive sample indicates that the corresponding optical power sequence is normal, the abnormal result of the negative sample indicates that the corresponding optical power sequence is abnormal, and further includes an optical power value with an abnormality. The first initial detection model is trained according to the validation set and the training set to form a trained first initial detection model, and the trained first initial detection model is used as the above-mentioned first detection model.

[0112] In one embodiment, before inputting the optical power sequence into the preset first detection model to perform abnormal analysis on the laser coupling unit, the above-mentioned detection method also includes: training the first detection model; the training process includes: obtaining an initial sample set, the initial sample set containing an optical power sequence consisting of optical power values ​​formed by numerical quantization under normal and abnormal conditions; preprocessing the initial sample set to form a standard training set and a standard verification set for model training; iteratively training the first initial detection model using the standard training set, and adjusting the model parameters during each iterative training process to obtain the first initial detection model after completing the iterative training; verifying the first initial detection model after completing the iterative training through the standard verification set to obtain a model verification result, and when the model verification result indicates that the first initial detection model after completing the iterative training can accurately distinguish the optical power sequences under normal and abnormal conditions, the first initial detection model after completing the iterative training is used as the first detection model.

[0113] In this embodiment, the optical power values ​​in the initial sample set are optical power values ​​measured historically. Specifically, it can be the historical optical power signal measured by the above-mentioned integrating sphere, and the optical power value formed after the historical optical power signal is numerically quantized based on the error rate of the integrating sphere. By performing model training based on the optical power values ​​formed by numerical quantization, the efficiency of model training can be improved.

[0114] For the initial sample set, before training, data preprocessing is also performed on the initial sample set. The preprocessing process includes one or more of data cleaning (removing noise and outliers), feature extraction (extracting features that are meaningful for anomaly detection) and standardization (ensuring that the data has the same scale).

[0115] The preprocessed initial sample set is divided into a standard training set and a standard validation set. Both the standard training set and the standard validation set contain an appropriate number of positive samples and negative samples. The first initial detection model to be trained is iteratively trained using the standard training set, and the parameters in the model are adjusted during each iterative training process. When the iterative stop condition is met (for example, the number of iterative training reaches a preset number threshold, or the model prediction result tends to be stable, etc.), the iterative training is completed, and the first initial detection model after the iterative training is obtained.

[0116] For the first initial detection model that has completed iterative training, the standard validation set is input into the model for anomaly detection, and the anomaly detection results of each sample in the standard validation set are output. The anomaly detection accuracy of the model is calculated by combining the anomaly results marked in the sample and the anomaly detection results. When the accuracy reaches the preset accuracy threshold, it means that the first initial detection model after iterative training can accurately distinguish the optical power sequences under normal and abnormal conditions. At this time, the first initial detection model after iterative training can be used as the first detection model. When the accuracy does not reach the accuracy threshold, the above-mentioned iterative training process is returned to re-iterate the first initial detection model.

[0117] The above-mentioned number threshold and accuracy threshold may be any appropriate values ​​set according to actual conditions.

[0118] Take the CART decision tree model as an example: the model parameters such as the depth of the tree, the number of leaf nodes, and the splitting criteria are set in the decision tree model. In the iterative process, the model uses the Gini index as the splitting criterion, recursively splits the data in the standard training set, and gradually builds a decision tree. Specifically, combined with Fig.10As shown in the figure, in each iteration, all features are traversed to find the feature that can best distinguish the normal state from the abnormal state as the split node; according to the value of the split feature, the data set is divided into two left and right subsets, and the subsets are recursively processed in the same way until the stopping condition is met (such as reaching the maximum tree depth, the number of samples contained in the node is less than the threshold, etc.).

[0119] For the trained model, CART pruning is performed on the validation set to avoid overfitting of the model. Specifically, a method such as CCP pruning (Cost Complexity Pruning) can be used for pruning, and the pruned decision tree model is used as the first detection model.

[0120] By constructing a first detection model that meets the requirements, inputting the optical power sequence into the model for detection, and outputting a first analysis result of whether the laser coupling unit corresponding to the optical power sequence is abnormal, the accuracy of abnormal identification of the laser unit can be improved.

[0121] In one embodiment, when the laser coupling unit is abnormal, the first analysis result also includes the unit number of the abnormal laser unit. When the first analysis result is that the laser coupling unit is abnormal, the method further includes: defining the optical power sequence of the abnormal laser coupling unit as the target optical power sequence; inputting the first analysis result and the target optical power sequence into a preset second detection model to determine the second analysis result of the abnormal laser coupling unit.

[0122] The second analysis result includes the unit number of the abnormal laser unit and the abnormal component of the abnormal laser unit, and the abnormal component includes one or more of a COS, a fast axis collimator, a slow axis collimator, a reflector and a focusing lens.

[0123] In this embodiment, even if there are abnormalities in components of the same type of laser units, if the laser units are in different laser coupling units, the abnormal characteristics they exhibit may not be the same. For example, in a fiber coupling module, if there is an abnormality in the fast axis collimator in the first group of laser units in the first laser coupling unit, and there is an abnormality in the fast axis collimator in the first group of laser units in the second laser coupling unit, the abnormal characteristics presented in the target optical power sequence of the two are different.

[0124] The second detection model is a model for identifying abnormalities of specific components in the laser unit. The model can also be any suitable machine learning model, such as a clustering model, a random forest, a gradient boosting tree (GBDT), a support vector machine (SVM) or a deep learning model such as a convolutional neural network SNN.

[0125] Taking the second detection model as a random forest model as an example, the electronic device can pre-establish a random forest model according to the above-mentioned process of building a decision tree model. Among them, the samples in the validation set and sample set used by the random forest model are also optical power sequences, and each sample data is further marked with the coupler number, the unit number of the abnormal laser unit, and the abnormal component in the abnormal laser unit. For example, sample data 1 marks the COS abnormality of laser unit a, sample data 2 marks the fast axis collimator abnormality of laser unit b, sample data 3 marks the slow axis collimator abnormality of laser unit c and laser unit d, etc. The sample data 1, 2, and 3 are all the first laser coupling units in the same model of fiber coupling module.

[0126] By calling the random forest model to learn and train the validation set and the sample set, abnormal features corresponding to different second analysis results can be identified.

[0127] For the first analysis result with an abnormality identified in step 130, the coupler number of the laser coupling unit, the unit number of the laser unit with the abnormality, and the target optical power sequence can be input into the trained random forest model. By running the random forest model, the specific abnormal component can be further located to obtain the second analysis result.

[0128] In this embodiment, by introducing the second detection model, the target optical power sequence is further analyzed based on the first analysis result, so that the abnormal component in the abnormal laser unit can be located, thereby improving the precision of abnormality identification.

[0129] In one embodiment, Fig.11 As shown, a device for detecting abnormality of an optical fiber coupling module is provided, the device comprising:

[0130] The optical power value extraction unit 1110 is used to obtain the optical power value of each group of laser units.

[0131] The optical power sequence processing unit 1120 is used to generate an optical power sequence of each laser coupling unit based on the optical power value.

[0132] The abnormality analysis unit 1130 is used to perform abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes that the laser coupling unit is normal or the laser coupling unit is abnormal.

[0133] In one embodiment, the abnormality analysis unit 1130 is further used to analyze the difference characteristics between any two optical power values ​​in the optical power sequence; based on the difference characteristics, an abnormality analysis is performed on the laser coupling unit to obtain a first analysis result.

[0134] In one embodiment, the abnormality analysis unit 1130 is further used to compare the optical power curve generated by the optical power sequence with a preset standard curve to obtain a comparison result; and perform abnormality analysis on the laser coupling unit based on the comparison result to obtain a first analysis result.

[0135] In one embodiment, the abnormality analysis unit 1130 is further used to input the optical power sequence into a preset first detection model to perform abnormality analysis on the laser coupling unit to obtain a first analysis result.

[0136] In one embodiment, the above-mentioned device also includes: a model training module, which is used to train the first detection model. The training process includes: obtaining an initial sample set, the initial sample set containing an optical power sequence composed of optical power values ​​formed by numerical quantization under normal and abnormal conditions; preprocessing the initial sample set to form a standard training set and a standard verification set for model training; iteratively training the first initial detection model using the standard training set, and adjusting the model parameters during each iterative training process to obtain the first initial detection model after the iterative training is completed; verifying the first initial detection model after the iterative training through the standard verification set to obtain a model verification result, and when the model verification result indicates that the first initial detection model after the iterative training can accurately distinguish the optical power sequences under normal and abnormal conditions, the first initial detection model after the iterative training is used as the first detection model.

[0137] In one embodiment, when the laser coupling unit is abnormal, the first analysis result further includes a unit number of the abnormal laser coupling unit.

[0138] The abnormality analysis unit 1130 is also used to define the optical power sequence of the abnormal laser coupling unit as the target optical power sequence when the first analysis result is that the laser coupling unit is abnormal; input the first analysis result and the target optical power sequence into the preset second detection model to determine the second analysis result of the abnormal laser coupling unit. The second analysis result includes the unit number of the abnormal laser unit and the abnormal component of the abnormal laser unit; the abnormal component includes one or more of COS, fast axis collimator, slow axis collimator, reflector and focusing lens.

[0139] In one embodiment, the optical power value extraction unit 1110 is further used to measure the optical power signal of each group of laser units; and to numerically quantify the optical power signal based on the measurement error rate to obtain the optical power value of the laser unit.

[0140] In one embodiment, the optical power value extraction unit 1110 is also used to obtain the abnormal state of the optical fiber coupling module; when the optical fiber coupling module is abnormal, the optical power value of each group of laser units is obtained.

[0141] In one embodiment, a computer-readable storage medium is provided, on which executable instructions are stored. When the instructions are executed by a processor, the processor executes the steps in the above-mentioned method embodiments.

[0142] In one embodiment, an electronic device is also provided, including one or more processors; a memory, in which one or more programs are stored, wherein when the one or more programs are executed by one or more processors, the one or more processors execute the steps in the above-mentioned method embodiments.

[0143] In one embodiment, an electronic device is provided, which may be the CRT measuring instrument described above. Fig.12 As shown, the electronic device 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1202 or the program loaded from the storage part 1208 to the random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the electronic device 1200 are also stored. The CPU 1201, the ROM 1202, and the RAM 1203 are connected to each other through a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0144] The following components are connected to the I / O interface 1205: an input section 1206 including a keyboard, a mouse, etc.; an output section 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN card, a modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the I / O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1210 as needed, so that a computer program read therefrom is installed into the storage section 1208 as needed.

[0145] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, including a computer-readable medium carrying instructions, in such an embodiment, the instructions can be downloaded and installed from a network through a communication part 1209, and / or installed from a removable medium 1211. When the instructions are executed by a central processing unit (CPU) 1201, the various method steps described in the present invention are executed.

[0146] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements 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 application.

[0147] In addition, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present application and form different embodiments. The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present application, and should not be regarded as an admission or in any form to imply that the information constitutes prior art already known to those skilled in the art.

Claims

1. A method for detecting abnormality of an optical fiber coupling module, characterized in that: The optical fiber coupling module includes at least one laser coupling unit, each laser coupling unit includes multiple groups of laser units, and the method includes: Obtain the optical power value of each group of laser units; Generate an optical power sequence of each laser coupling unit based on the optical power value; Performing an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes that the laser coupling unit is normal or the laser coupling unit is abnormal; When the laser coupling unit is abnormal, the first analysis result further includes a unit number of the abnormal laser coupling unit; When the first analysis result is that the laser coupling unit is abnormal, define the optical power sequence of the laser coupling unit abnormality as a target optical power sequence, input the first analysis result and the target optical power sequence into a preset second detection model, and determine a second analysis result of the laser coupling unit abnormality; The second analysis result includes the unit number of the abnormal laser unit and the abnormal component of the abnormal laser unit; the abnormal component includes one or more of a COS, a fast axis collimator, a slow axis collimator, a reflector and a focusing lens.

2. The method according to claim 1, characterized in that: Performing an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes: Analyzing the difference characteristics between any two optical power values ​​in the optical power sequence; An abnormality analysis is performed on the laser coupling unit based on the difference feature to obtain the first analysis result.

3. The method according to claim 1, characterized in that Performing an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes: Comparing the optical power curve generated by the optical power sequence with a preset standard curve to obtain a comparison result; An abnormality analysis is performed on the laser coupling unit based on the comparison result to obtain the first analysis result.

4. The method according to claim 1, characterized in that: Performing an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result includes: The optical power sequence is input into a preset first detection model to perform abnormality analysis on the laser coupling unit to obtain a first analysis result.

5. The method according to claim 4, characterized in that Before inputting the optical power sequence into a preset first detection model to perform abnormal analysis on the laser coupling unit, the method further includes: training the first detection model; Training the first detection model includes: Acquire an initial sample set, wherein the initial sample set includes an optical power sequence consisting of numerically quantized optical power values ​​under normal and abnormal conditions; Preprocessing the initial sample set to form a standard training set and a standard validation set for model training; Iteratively training the first initial detection model using the standard training set, and adjusting the model parameters during each iterative training process to obtain the first initial detection model after completing the iterative training; The first initial detection model after iterative training is verified by the standard verification set to obtain a model verification result. When the model verification result indicates that the first initial detection model after iterative training can accurately distinguish the optical power sequences under normal and abnormal conditions, the first initial detection model after iterative training is used as the first detection model.

6. The method according to claim 1, characterized in that The obtaining of the optical power value emitted by each group of laser units comprises: Measuring the optical power signal of each group of laser units; The optical power signal is numerically quantified based on a measurement error rate to obtain an optical power value of the laser unit.

7. The method according to any one of claims 1 to 6, characterized in that Before obtaining the optical power value emitted by each group of laser units, the method further includes: Acquiring an abnormal state of the optical fiber coupling module; When the optical fiber coupling module is abnormal, the optical power value of each group of laser units is obtained.

8. A device for detecting abnormality of an optical fiber coupling module, characterized in that: The optical fiber coupling module includes at least one laser coupling unit, each laser coupling unit includes multiple groups of laser units, and the device includes: An optical power value extraction unit, used to obtain the optical power value of each group of laser units; An optical power sequence processing unit, used for generating an optical power sequence of each laser coupling unit based on the optical power value; an abnormality analysis unit, configured to perform an abnormality analysis on the laser coupling unit based on the optical power sequence to obtain a first analysis result; the first analysis result includes whether the laser coupling unit is normal or the laser coupling unit is abnormal, and when the laser coupling unit is abnormal, the first analysis result also includes a unit number of the abnormal laser coupling unit; The abnormality analysis unit is also used to define the optical power sequence of the abnormal laser coupling unit as a target optical power sequence when the first analysis result is that the laser coupling unit is abnormal, input the first analysis result and the target optical power sequence into a preset second detection model, and determine the second analysis result of the abnormal laser coupling unit, wherein the second analysis result includes the unit number of the abnormal laser unit and the abnormal component of the abnormal laser unit, and the abnormal component includes one or more of a COS, a fast-axis collimator, a slow-axis collimator, a reflector, and a focusing lens.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores executable instructions, and when the executable instructions are executed by a processor, the processor is enabled to perform the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs, which, when executed by the one or more processors, causes the one or more processors to perform the method according to any one of claims 1 to 7.

Images