Optical cable online monitoring method and system based on time series big data analysis

Abstract

The present invention relates to the field of optical cable online monitoring technology, and is an optical cable online monitoring method and system based on time series big data analysis, and the specific method includes: obtaining a historical cutting time series data segment, and preprocessing the historical cutting time series data segment; establishing an optical cable time series data model, and performing coupling analysis on real-time optical fiber fingerprint data packets through the optical cable time series data model; according to the coupling analysis results, performing interference superposition analysis on interference points of the same optical cable operation status category on each optical cable; calculating and obtaining the risk probability value of each operation status characterization parameter, and generating an optical cable monitoring report. The present invention solves the problems of strong manual dependence, poor real-time performance and low efficiency in the traditional optical cable monitoring method in the prior art.

Description

Technical Field

[0001] The present invention relates to the technical field of optical cable online monitoring, and relates to an optical cable online monitoring method and system based on time series big data analysis. Background Art

[0002] As the core infrastructure of modern communication networks, the stability and reliability of optical cable operation are directly related to the service quality of communication networks. However, as the scale and complexity of optical cable networks continue to expand, traditional optical cable monitoring methods have gradually exposed many problems, including:

[0003] First of all, the traditional optical cable monitoring method mainly relies on manual inspections and regular testing. This method is not only inefficient, but also cannot achieve real-time monitoring of the operating status of the optical cable. The periodicity and lag of manual inspections lead to delays in fault discovery and processing, which often increases the risk of network interruption and affects the stability and service quality of the communication network.

[0004] Secondly, the single parameter monitoring method used in the prior art is difficult to fully reflect the actual operating status of the optical cable, resulting in insufficient accuracy and comprehensiveness of fault detection.

[0005] In addition, the existing technology lacks an effective time series data analysis method when processing optical cable operation status data, resulting in insufficient prediction and anomaly detection capabilities for optical cable status changes. Especially when processing large-scale optical cable data, there are problems of low data processing efficiency and high consumption of computing resources, making it difficult to meet the real-time monitoring needs of large-scale optical cable networks.

[0006] In summary, the existing optical cable monitoring methods have significant deficiencies in real-time, comprehensiveness and data analysis capabilities. There is an urgent need for a comprehensive optical cable online monitoring method to achieve efficient, accurate and real-time monitoring of the optical cable operating status and improve the operation and maintenance efficiency and reliability of the communication network. Summary of the invention

[0007] The technical problem to be solved by the present invention is to address the problems of strong manual dependence, poor real-time performance and low efficiency in the traditional optical cable monitoring method in the prior art, and propose an optical cable online monitoring method and system based on time series big data analysis.

[0008] In order to achieve the above object, the technical solution of the optical cable online monitoring method based on time series big data analysis of the present invention includes the following steps:

[0009] S1: continuously obtain the real-time fiber fingerprint data packets of the real-time operation status of all optical cables in the monitoring area and the historical fiber fingerprint data packets of the historical operation status, cut the historical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments;

[0010] S2: Establishing an optical cable timing data model, and performing coupling analysis on the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model;

[0011] S3: Based on the coupling analysis results, interference superposition analysis is performed on the interference points of the same cable operation status category on each optical cable;

[0012] S4: extracting the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculating and obtaining the risk probability value of each operation status characterization parameter;

[0013] S5: Determine the risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualize the risk characterization parameters, and generate an optical cable monitoring report.

[0014] Specifically, S1 includes the following specific steps:

[0015] S11: continuously record the real-time fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the sensor network, and store them in the local server of the optical cable monitoring platform, wherein the total number of optical cables in the monitoring area is G, and g is the index of the optical cable;

[0016] The historical fiber fingerprint data packet of the historical operation status includes: a historical fiber fingerprint data master packet and a historical fiber fingerprint data sub-packet; the historical fiber fingerprint data master packet includes: optical signal data of the optical cable at different time nodes; the historical fiber fingerprint data sub-packet includes: R types of operation status characterization parameters of the optical cable at different time nodes;

[0017] S12: Preset a unit time step, extract the historical fiber fingerprint data packet in the local server of the optical cable monitoring platform, cut the historical fiber fingerprint data packet according to the unit time step, and obtain a historical cutting time series data segment, wherein the unit time step is 10 seconds; the historical cutting time series data segment includes: a historical cutting time series data mother segment and a historical cutting time series data sub-segment;

[0018] S13: At the same time, the mother segment of the historical cutting time series data is scattered;

[0019] S14: Time domain performance of optical signal data in the mother segment of statistical historical cutting time series data , where t is the time axis consisting of historical unit time steps.

[0020] Specifically, S1 also includes the following specific steps:

[0021] S15: Convert the optical signal data in the mother segment of the historical cutting time series data into the frequency domain through Fourier transform, and count the frequency domain performance of the optical signal data in the mother segment of the historical cutting time series data , where f is the frequency axis composed of historical frequency components corresponding to the historical unit time step;

[0022] S16: performing wavelet transformation and denoising on the mother segment of the historical cutting time series data, wherein the wavelet transformation includes:

[0023] ;

[0024] in, is the wavelet coefficient of the optical signal data recorded by the i-th sensor on the g-th optical cable at the scale j;

[0025] is the wavelet function, j is the scale, and k represents the position;

[0026] The denoising process comprises: ;

[0027] for Wavelet coefficients after denoising;

[0028] Indicates the noise threshold when the scale is j;

[0029] S17: According to steps S14-S16, the multivariate signal features of the mother segment of the historical cutting time series data are extracted respectively, and the multivariate signal features include: time domain features , frequency domain characteristics and time-frequency domain features ;

[0030] The extraction strategy of the multivariate signal features specifically includes:

[0031] ;

[0032] x is the multivariate signal feature index, ;

[0033] When x=1, The time domain representation of the optical signal data in the mother segment of the historical cutting time series data , Time domain features ;

[0034] When x=2, The frequency domain representation of the optical signal data in the mother segment of the historical cutting time series data , is the frequency domain feature ;

[0035] When x=3, is the wavelet coefficient of the denoised optical signal data in the mother segment of the historical cutting time series data , is the time-frequency domain feature ;

[0036] When the index of the multivariate signal feature is x, The mean of

[0037] When the index of the multivariate signal feature is x, The standard deviation of

[0038] When the index of the multivariate signal feature is x, The maximum and minimum amplitudes of

[0039] When the index of the multivariate signal feature is x, skewness;

[0040] When the index of the multivariate signal feature is x, The kurtosis of .

[0041] Specifically, S2 includes:

[0042] S21: dividing the historical fiber fingerprint data mother package and the corresponding historical cutting time series data mother segment into a 9:1 ratio to form a training set and a test set;

[0043] S22: Establishing an optical cable time series data model, and inputting the historical optical fiber fingerprint data mother package, the historical cutting time series data mother section and the multivariate signal features in the training set into the input layer of the optical cable time series data model;

[0044] S23: extracting context information of the historical optical fiber fingerprint data mother package through a recurrent neural network, and outputting a context feature vector of the historical optical fiber fingerprint data mother package;

[0045] S24: extracting local features of the mother segment of the historical cutting time series data through a convolutional neural network, and outputting an initial feature vector of the mother segment of the historical cutting time series data;

[0046] S25: extract three high-dimensional feature vectors corresponding to the multivariate signal features through the fully connected layer;

[0047] S26: The initial feature vector of the mother segment of the historical cutting time series data is fused through the attention mechanism to extract high-dimensional fusion features;

[0048] S27: Perform nonlinear transformation on the input high-dimensional fusion features through the fully connected layer to extract high-level abstract representation of the features, and use the last fully connected layer to classify and map each mother section of the historical cutting time series data into Y optical cable operation status categories, wherein the activation function of the last fully connected layer is a softmax activation function, and the output of the last fully connected layer is Y probability values;

[0049] S28: Train the optical cable timing data model and use the test set to test the classification accuracy of the optical cable timing data model. When the accurately classified historical cutting timing data mother section reaches 90%, the training of the optical cable timing data model is completed.

[0050] Specifically, S2 also includes the following specific steps:

[0051] S29: extracting a real-time optical fiber fingerprint data packet of the real-time operating status of all optical cables in the monitoring area, and cutting the real-time optical fiber fingerprint data packet according to the unit time step to obtain a real-time cutting time series data segment, and breaking up the real-time cutting time series data mother segment, wherein the real-time cutting time series data segment includes: a real-time cutting time series data mother segment and a real-time cutting time series data sub-segment;

[0052] S210: Inputting the real-time cutting time series data mother section into the trained optical cable time series data model, and outputting the optical cable operation status category to which the real-time cutting time series data mother section belongs.

[0053] Specifically, S3 includes the following specific steps:

[0054] S31: matching all the scattered real-time cutting time sequence data segments to the corresponding optical cables according to the index of the optical cable on the real-time cutting time sequence data segment;

[0055] S32: extracting all the cable operation status categories on each section of each optical cable to form an optical cable operation status category set, wherein the optical cable operation status category set of the g-th optical cable is , , is the set of interference points on the g-th optical cable whose operating status category is the y-th category;

[0056] S33: Get The straight-line distances between all interference points and the optical fiber transmitting end are calculated, and the straight-line distances between each interference point and the optical fiber transmitting end are arranged in ascending order;

[0057] S34: Interference point collection The first interfering subpoint As the reference point, The other adjacent interference points in the interference stacking analysis are combined to perform interference stacking analysis, and the interference stacking analysis specifically includes:

[0058] ;

[0059] in, represents the comprehensive interference superposition coefficient between the u-th and u+1-th interference points, represents the feature weight of the x-th multivariate signal feature;

[0060] It represents the single interference superposition coefficient of the u-th and u+1-th interference points on the x-th multivariate signal feature;

[0061] ;

[0062] is the covariance of the u-th and u+1-th interference points on the x-th multivariate signal feature, is the standard deviation of the u-th and u+1-th interference points on the x-th multivariate signal feature.

[0063] Specifically, S3 also includes the following specific steps:

[0064] S35: Evaluate the set of interference points The comprehensive interference superposition coefficient of each adjacent interference point combination is calculated, and the average level of the comprehensive interference superposition coefficient is calculated, and each adjacent interference point combination whose comprehensive interference superposition coefficient is greater than the average level is selected, and at the same time, the optical cable operation status category of an interference point farther from the optical fiber transmitting end in the adjacent interference point combination is reduced from the yth category to the y-1th category;

[0065] S36: Traverse the cable operation status category set of the g-th optical cable , conduct interference superposition analysis;

[0066] S37: After the traversal is completed, the optical cable operation status categories of all interference points on the g-th optical cable are output.

[0067] Specifically, S4 includes the following steps:

[0068] S41: extracting the real-time cutting time series data sub-segment of the interference point corresponding to the optical cable operation status category of special concern on each optical cable, wherein the confirmation strategy of the optical cable operation status category of special concern on each optical cable includes: arranging the number of occurrences of each optical cable operation status category in the historical optical fiber fingerprint data packet on each optical cable in descending order, and updating the sequence of the number of occurrences of each optical cable operation status category in real time, intercepting the top three optical cable operation status categories with the highest number of occurrences on each optical cable as the optical cable operation status categories of special concern, and the number of interference points corresponding to each optical cable operation status category of special concern is respectively ;

[0069] S42: on each optical cable, calculating and obtaining risk probability values ​​of R types of operation status characterization parameters of the optical cable operation status category of particular concern, to form a risk probability value set, , where a is the index of the cable operation status category of special concern, Represents the risk probability value in the R-type operating status characterization parameters of the optical cable operating status category of special concern a.

[0070] Specifically, in S42, the calculation strategy of the risk probability value is:

[0071] S421: Preset curve deviation thresholds corresponding to R types of operating state characterization parameters;

[0072] S422: Optical cable The unit time step of the interference point is taken as the horizontal axis, and the rth operating state characterization parameter is taken as the vertical axis. The actual geometric characteristic curve of the rth operating state characterization parameter is drawn, and the At an interference point, the actual geometric characteristic curve of the rth operating state characterization parameter is compared with the standard geometric characteristic curve of the rth operating state characterization parameter in the factory state of the optical cable, and the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is obtained. When the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is less than the rth curve deviation threshold, the rth operating state characterization parameter of the interference point is marked as risk;

[0073] S423: Calculate the sub-risk probability value of the rth operating state characterization parameter , the sub-risk probability value The calculation strategy is: the number of interference points marked as risk and the total number of interference points on this optical cable The ratio of

[0074] S424: Repeat steps S422-S423 to obtain the sub-risk probability values ​​of the R types of operating status characterization parameters, and select the maximum sub-risk probability value as the risk probability value in the R type operating status characterization parameters of the a-th type of optical cable operating status category of special concern .

[0075] Specifically, in S422, the calculation strategy of the curve deviation is:

[0076] ;

[0077] in, For the The curve deviation of the geometric characteristic curve of the rth operating state characterization parameter at the interference point;

[0078] It represents the data volume of the parameters representing the rth operating state at the time point T in the unit time step;

[0079] It represents the mean value of the data volume of the rth operating state characterization parameter when the optical cable leaves the factory.

[0080] Specifically, S5 includes the following specific steps:

[0081] S51: Extract the risk probability value of the optical cable operation status category of special concern a The corresponding operating status characterization parameter is selected as the risk characterization parameter of the optical cable operating status category of special concern a;

[0082] S52: Visualize the risk characterization parameters of all optical cable operation status categories of special concern, and generate an optical cable monitoring report, wherein the optical cable monitoring report also includes: the optical cable position of the interference point corresponding to the maximum curve deviation in the curve deviation set composed of the curve deviation of each risk characterization parameter.

[0083] In addition, the optical cable online monitoring system based on time series big data analysis of the present invention includes the following modules:

[0084] Fiber fingerprint data packet division module, model building module, interference superposition analysis module, risk probability value calculation module and optical cable monitoring report generation module;

[0085] The optical fiber fingerprint data packet division module is used to continuously obtain the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area and the historical optical fiber fingerprint data packets of the historical operating status, cut the historical optical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments;

[0086] The model building module is used to establish an optical cable timing data model, and to perform coupling analysis on real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model;

[0087] The interference superposition analysis module performs interference superposition analysis on interference points of the same optical cable operation status category on each optical cable according to the coupling analysis result;

[0088] The risk probability value calculation module is used to extract the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculate the risk probability value of each operation status characterization parameter;

[0089] The optical cable monitoring report generating module determines the risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualizes the risk characterization parameters, and generates an optical cable monitoring report.

[0090] Compared with the prior art, the technical effects of the present invention are as follows:

[0091] 1. The present invention combines time series big data analysis and machine learning models to achieve accurate monitoring of the real-time operating status of optical cables and rapid fault response, significantly improving the stability and reliability of optical cable networks.

[0092] 2. The present invention adopts a multi-dimensional feature fusion analysis method to comprehensively evaluate the operating status of the optical cable, greatly enhancing the comprehensiveness and depth of fault diagnosis and reducing the false alarm and missed alarm rates.

[0093] 3. The present invention introduces a fault classification method based on risk probability value to achieve quantitative assessment and hierarchical management of optical cable fault risks, optimize maintenance strategies and resource allocation, and significantly reduce operation and maintenance costs.

[0094] 4. The present invention visualizes the risk characterization parameters to achieve an intuitive display of the fault location and risk level, greatly facilitating the troubleshooting and management of operation and maintenance personnel and improving the overall operation and maintenance efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0096] Figure 1 It is a flow chart of the optical cable online monitoring method based on time series big data analysis of the present invention;

[0097] Figure 2 It is a structural schematic diagram of the optical cable online monitoring system based on time series big data analysis of the present invention. DETAILED DESCRIPTION

[0098] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0099] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0100] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0101] Embodiment 1

[0102] like Figure 1 As shown, the optical cable online monitoring method based on time series big data analysis of an embodiment of the present invention is as follows: Figure 1 As shown, the specific steps are as follows:

[0103] S1: continuously obtain the real-time fiber fingerprint data packets of the real-time operation status of all optical cables in the monitoring area and the historical fiber fingerprint data packets of the historical operation status, cut the historical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments;

[0104] S1 includes the following specific steps:

[0105] S11: continuously record the real-time fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the sensor network, and store them in the local server of the optical cable monitoring platform, wherein the total number of optical cables in the monitoring area is G, and g is the index of the optical cable;

[0106] The historical fiber fingerprint data packet of the historical operation status includes: a historical fiber fingerprint data master packet and a historical fiber fingerprint data sub-packet; the historical fiber fingerprint data master packet includes: optical signal data of the optical cable at different time nodes; the historical fiber fingerprint data sub-packet includes: R types of operation status characterization parameters of the optical cable at different time nodes;

[0107] Exemplarily, in this embodiment, the operating status characterization parameters of the optical cable include: optical cable temperature value, optical cable vibration amplitude value, and optical cable mechanical stress value;

[0108] S12: Preset a unit time step, extract the historical fiber fingerprint data packet in the local server of the optical cable monitoring platform, cut the historical fiber fingerprint data packet according to the unit time step, and obtain a historical cutting time series data segment, wherein the unit time step is 10 seconds; the historical cutting time series data segment includes: a historical cutting time series data mother segment and a historical cutting time series data sub-segment;

[0109] S13: At the same time, the mother segment of the historical cutting time series data is scattered;

[0110] S14: Time domain performance of optical signal data in the mother segment of statistical historical cutting time series data , where t is the time axis consisting of historical unit time steps.

[0111] S1 also includes the following specific steps:

[0112] S15: Convert the optical signal data in the mother segment of the historical cutting time series data into the frequency domain through Fourier transform, and count the frequency domain performance of the optical signal data in the mother segment of the historical cutting time series data , where f is the frequency axis composed of historical frequency components corresponding to the historical unit time step;

[0113] S16: performing wavelet transformation and denoising on the mother segment of the historical cutting time series data, wherein the wavelet transformation includes:

[0114] ;

[0115] in, is the wavelet coefficient of the optical signal data recorded by the i-th sensor on the g-th optical cable at the scale j;

[0116] is the wavelet function, j is the scale, and k represents the position;

[0117] The denoising process comprises: ;

[0118] for Wavelet coefficients after denoising;

[0119] Indicates the noise threshold when the scale is j;

[0120] Exemplarily, in this embodiment, a method for confirming a noise threshold is provided, which specifically includes: ,in, represents the local standard deviation of the wavelet coefficients at the jth scale, is the number of wavelet coefficients at the jth scale;

[0121] S17: According to steps S14-S16, the multivariate signal features of the mother segment of the historical cutting time series data are extracted respectively, and the multivariate signal features include: time domain features , frequency domain characteristics and time-frequency domain features ;

[0122] The extraction strategy of the multivariate signal features specifically includes:

[0123] ;

[0124] x is the multivariate signal feature index, ;

[0125] When x=1, The time domain representation of the optical signal data in the mother segment of the historical cutting time series data , Time domain features ;

[0126] When x=2, The frequency domain representation of the optical signal data in the mother segment of the historical cutting time series data , is the frequency domain feature ;

[0127] When x=3, is the wavelet coefficient of the denoised optical signal data in the mother segment of the historical cutting time series data , is the time-frequency domain feature ;

[0128] When the index of the multivariate signal feature is x, The mean of

[0129] When the index of the multivariate signal feature is x, The standard deviation of

[0130] When the index of the multivariate signal feature is x, The maximum and minimum amplitudes of

[0131] When the index of the multivariate signal feature is x, skewness;

[0132] When the index of the multivariate signal feature is x, The kurtosis of .

[0133] S2: Establishing an optical cable timing data model, and performing coupling analysis on the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model;

[0134] S2 includes:

[0135] S21: dividing the historical fiber fingerprint data mother package and the corresponding historical cutting time series data mother segment into a 9:1 ratio to form a training set and a test set;

[0136] S22: Establishing an optical cable time series data model, and inputting the historical optical fiber fingerprint data mother package, the historical cutting time series data mother section and the multivariate signal features in the training set into the input layer of the optical cable time series data model;

[0137] S23: extracting context information of the historical optical fiber fingerprint data mother package through a recurrent neural network, and outputting a context feature vector of the historical optical fiber fingerprint data mother package;

[0138] S24: extracting local features of the mother segment of the historical cutting time series data through a convolutional neural network, and outputting an initial feature vector of the mother segment of the historical cutting time series data;

[0139] S25: extract three high-dimensional feature vectors corresponding to the multivariate signal features through the fully connected layer;

[0140] S26: The initial feature vector of the mother segment of the historical cutting time series data is fused through the attention mechanism to extract high-dimensional fusion features;

[0141] S27: Perform nonlinear transformation on the input high-dimensional fusion features through the fully connected layer to extract high-level abstract representation of the features, and use the last fully connected layer to classify and map each mother section of the historical cutting time series data into Y optical cable operation status categories, wherein the activation function of the last fully connected layer is a softmax activation function, and the output of the last fully connected layer is Y probability values;

[0142] S28: Train the optical cable timing data model and use the test set to test the classification accuracy of the optical cable timing data model. When the accurately classified historical cutting timing data mother section reaches 90%, the training of the optical cable timing data model is completed.

[0143] S29: extracting a real-time optical fiber fingerprint data packet of the real-time operating status of all optical cables in the monitoring area, and cutting the real-time optical fiber fingerprint data packet according to the unit time step to obtain a real-time cutting time series data segment, and breaking up the real-time cutting time series data mother segment, wherein the real-time cutting time series data segment includes: a real-time cutting time series data mother segment and a real-time cutting time series data sub-segment;

[0144] S210: Inputting the real-time cutting time series data mother section into the trained optical cable time series data model, and outputting the optical cable operation status category to which the real-time cutting time series data mother section belongs.

[0145] S3: Based on the coupling analysis results, interference superposition analysis is performed on the interference points of the same cable operation status category on each optical cable;

[0146] S3 includes the following specific steps:

[0147] S31: matching all the scattered real-time cutting time sequence data segments to the corresponding optical cables according to the index of the optical cable on the real-time cutting time sequence data segment;

[0148] S32: extracting all the cable operation status categories on each section of each optical cable to form an optical cable operation status category set, wherein the optical cable operation status category set of the g-th optical cable is , , is the set of interference points on the g-th optical cable whose operating status category is the y-th category;

[0149] S33: Get The straight-line distances between all interference points and the optical fiber transmitting end are calculated, and the straight-line distances between each interference point and the optical fiber transmitting end are arranged in ascending order;

[0150] S34: Interference point collection The first interfering subpoint As the reference point, The other adjacent interference points in the interference stacking analysis are combined to perform interference stacking analysis, and the interference stacking analysis specifically includes:

[0151] ;

[0152] in, represents the comprehensive interference superposition coefficient between the u-th and u+1-th interference points, represents the feature weight of the x-th multivariate signal feature;

[0153] It represents the single interference superposition coefficient of the u-th and u+1-th interference points on the x-th multivariate signal feature;

[0154] ;

[0155] is the covariance of the u-th and u+1-th interference points on the x-th multivariate signal feature, is the standard deviation of the u-th and u+1-th interference points on the x-th multivariate signal feature.

[0156] S3 also includes the following specific steps:

[0157] S35: Evaluate the set of interference points The comprehensive interference superposition coefficient of each adjacent interference point combination is calculated, and the average level of the comprehensive interference superposition coefficient is calculated, and each adjacent interference point combination whose comprehensive interference superposition coefficient is greater than the average level is selected, and at the same time, the optical cable operation status category of an interference point farther from the optical fiber transmitting end in the adjacent interference point combination is reduced from the yth category to the y-1th category;

[0158] S36: Traverse the cable operation status category set of the g-th optical cable , conduct interference superposition analysis;

[0159] S37: After the traversal is completed, the optical cable operation status categories of all interference points on the g-th optical cable are output.

[0160] S4: extracting the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculating and obtaining the risk probability value of each operation status characterization parameter;

[0161] S4 includes the following steps:

[0162] S41: extracting the real-time cutting time series data sub-segment of the interference point corresponding to the optical cable operation status category of special concern on each optical cable, wherein the confirmation strategy of the optical cable operation status category of special concern on each optical cable includes: arranging the number of occurrences of each optical cable operation status category in the historical optical fiber fingerprint data packet on each optical cable in descending order, and updating the sequence of the number of occurrences of each optical cable operation status category in real time, intercepting the top three optical cable operation status categories with the highest number of occurrences on each optical cable as the optical cable operation status categories of special concern, and the number of interference points corresponding to each optical cable operation status category of special concern is respectively ;

[0163] S42: on each optical cable, calculating and obtaining risk probability values ​​of R types of operation status characterization parameters of the optical cable operation status category of particular concern, to form a risk probability value set, , where a is the index of the cable operation status category of special concern, Represents the risk probability value in the R-type operating status characterization parameters of the optical cable operating status category of special concern a.

[0164] Preferably, in S42, the calculation strategy of the risk probability value is:

[0165] S421: Preset curve deviation thresholds corresponding to R types of operating state characterization parameters;

[0166] S422: Optical cable The unit time step of the interference point is taken as the horizontal axis, and the rth operating state characterization parameter is taken as the vertical axis. The actual geometric characteristic curve of the rth operating state characterization parameter is drawn, and the At an interference point, the actual geometric characteristic curve of the rth operating state characterization parameter is compared with the standard geometric characteristic curve of the rth operating state characterization parameter in the factory state of the optical cable, and the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is obtained. When the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is less than the rth curve deviation threshold, the rth operating state characterization parameter of the interference point is marked as risk;

[0167] Exemplarily, in this embodiment, a method for obtaining a standard geometric characteristic curve of an optical cable temperature value is provided, including:

[0168] The temperature change of the optical cable is monitored in real time by using the measuring device Fiber Bragg Grating Sensor (FBG);

[0169] Place fiber grating sensors or other temperature sensors on the optical cable to ensure that the sensors can evenly cover all positions of the optical cable;

[0170] The time step of temperature data measurement is set to once every minute, and the measurement range is along the length of the optical cable, with temperature measurement performed once every meter;

[0171] In a constant environment, with the time step as the horizontal axis and the temperature value as the vertical axis, record the temperature data of each position of the optical cable. Specifically, ensure that the measurement lasts for a long enough time, such as 24 hours, to obtain the distribution of the optical cable temperature in different time periods.

[0172] Process the collected data to remove noise and outliers. Specific methods include:

[0173] Moving average method: Smooth the temperature data to reduce measurement errors.

[0174] Data calibration: Compare the measured data with the data from the environmental temperature and humidity recorder to ensure the accuracy of the measurement.

[0175] According to the processed data, a standard geometric characteristic curve of the optical cable temperature value is drawn, with the time step as the horizontal axis and the temperature value as the vertical axis, to draw a standard geometric characteristic curve of the optical cable temperature changing with time;

[0176] Preferably, a spatial distribution curve of the optical cable temperature can also be drawn.

[0177] S423: Calculate the sub-risk probability value of the rth operating state characterization parameter , the sub-risk probability value The calculation strategy is: the number of interference points marked as risk and the total number of interference points on this optical cable The ratio of

[0178] S424: Repeat steps S422-S423 to obtain the sub-risk probability values ​​of the R types of operating status characterization parameters, and select the maximum sub-risk probability value as the risk probability value in the R type operating status characterization parameters of the a-th type of optical cable operating status category of special concern .

[0179] In S422, the calculation strategy of the curve deviation is:

[0180] ;

[0181] in, For the The curve deviation of the geometric characteristic curve of the rth operating state characterization parameter at the interference point;

[0182] It represents the data volume of the parameters representing the rth operating state at the time point T in the unit time step;

[0183] It represents the mean value of the data volume of the rth operating state characterization parameter when the optical cable leaves the factory.

[0184] S5: Determine risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualize the risk characterization parameters, and generate an optical cable monitoring report.

[0185] S5 includes the following specific steps:

[0186] S51: Extract the risk probability value of the optical cable operation status category of special concern a The corresponding operating status characterization parameter is selected as the risk characterization parameter of the optical cable operating status category of special concern a;

[0187] S52: Visualize the risk characterization parameters of all optical cable operation status categories of special concern, and generate an optical cable monitoring report, wherein the optical cable monitoring report also includes: the optical cable position of the interference point corresponding to the maximum curve deviation in the curve deviation set composed of the curve deviation of each risk characterization parameter.

[0188] Embodiment 2

[0189] like Figure 2 As shown, the optical cable online monitoring system based on time series big data analysis according to an embodiment of the present invention is as follows: Figure 2 As shown, it includes the following modules:

[0190] Fiber fingerprint data packet division module, model building module, interference superposition analysis module, risk probability value calculation module and optical cable monitoring report generation module;

[0191] The optical fiber fingerprint data packet division module is used to continuously obtain the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area and the historical optical fiber fingerprint data packets of the historical operating status, cut the historical optical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments;

[0192] The model building module is used to establish an optical cable timing data model, and to perform coupling analysis on real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model;

[0193] The interference superposition analysis module performs interference superposition analysis on interference points of the same optical cable operation status category on each optical cable according to the coupling analysis result;

[0194] The risk probability value calculation module is used to extract the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculate the risk probability value of each operation status characterization parameter;

[0195] The optical cable monitoring report generating module determines the risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualizes the risk characterization parameters, and generates an optical cable monitoring report.

[0196] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0197] It should be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and / or other information.

[0198] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or / and a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0199] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0200] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0201] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0202] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0203] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0204] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0205] In summary, compared with the prior art, the technical effects of the present invention are as follows:

[0206] 1. The present invention combines time series big data analysis and machine learning models to achieve accurate monitoring of the real-time operating status of optical cables and rapid fault response, significantly improving the stability and reliability of optical cable networks.

[0207] 2. The present invention adopts a multi-dimensional feature fusion analysis method to comprehensively evaluate the operating status of the optical cable, greatly enhancing the comprehensiveness and depth of fault diagnosis and reducing the false alarm and missed alarm rates.

[0208] 3. The present invention introduces a fault classification method based on risk probability value to achieve quantitative assessment and hierarchical management of optical cable fault risks, optimize maintenance strategies and resource allocation, and significantly reduce operation and maintenance costs.

[0209] 4. The present invention visualizes the risk characterization parameters to achieve an intuitive display of the fault location and risk level, greatly facilitating the troubleshooting and management of operation and maintenance personnel and improving the overall operation and maintenance efficiency.

[0210] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An optical cable online monitoring method based on time series big data analysis, characterized in that: The method comprises the following specific steps: S1: continuously obtain the real-time fiber fingerprint data packets of the real-time operation status of all optical cables in the monitoring area and the historical fiber fingerprint data packets of the historical operation status, cut the historical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments; S2: Establishing an optical cable timing data model, and performing coupling analysis on the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model; S3: Based on the coupling analysis results, interference superposition analysis is performed on the interference points of the same cable operation status category on each optical cable; S4: extracting the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculating and obtaining the risk probability value of each operation status characterization parameter; S5: Determine the risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualize the risk characterization parameters, and generate an optical cable monitoring report; The S3 comprises the following specific steps: S31: matching all the scattered real-time cutting time sequence data segments to the corresponding optical cables according to the index of the optical cable on the real-time cutting time sequence data segment; S32: extracting all the cable operation status categories on each section of each optical cable to form an optical cable operation status category set, wherein the optical cable operation status category set of the g-th optical cable is , , is the set of interference points on the g-th optical cable whose operating status category is the y-th category; S33: Get The straight-line distances between all interference points and the optical fiber transmitting end are calculated, and the straight-line distances between each interference point and the optical fiber transmitting end are arranged in ascending order; S34: Interference point collection The first interfering subpoint As the reference point, The other adjacent interference points in the interference stacking analysis are combined to perform interference stacking analysis, and the interference stacking analysis specifically includes: ； in, represents the comprehensive interference superposition coefficient between the u-th and u+1-th interference points, represents the feature weight of the x-th multivariate signal feature; It represents the single interference superposition coefficient of the u-th and u+1-th interference points on the x-th multivariate signal feature; The calculation strategy of the risk probability value is: S421: Preset curve deviation thresholds corresponding to R types of operating state characterization parameters; S422: Optical cable The unit time step of the interference point is taken as the horizontal axis, and the rth operating state characterization parameter is taken as the vertical axis. The actual geometric characteristic curve of the rth operating state characterization parameter is drawn, and the At an interference point, the actual geometric characteristic curve of the rth operating state characterization parameter is compared with the standard geometric characteristic curve of the rth operating state characterization parameter in the factory state of the optical cable, and the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is obtained. When the curve deviation of the geometric characteristic curve of the rth operating state characterization parameter is less than the rth curve deviation threshold, the rth operating state characterization parameter of the interference point is marked as risk; S423: Calculate the sub-risk probability value of the rth operating state characterization parameter , the sub-risk probability value The calculation strategy is: the number of interference points marked as risk and the total number of interference points on this optical cable The ratio of S424: Repeat steps S422-S423 to obtain the sub-risk probability values ​​of the R types of operating status characterization parameters, and select the maximum sub-risk probability value as the risk probability value in the R type operating status characterization parameters of the a-th type of optical cable operating status category of special concern .

2. The optical cable online monitoring method based on time series big data analysis according to claim 1 is characterized in that: S1 includes the following specific steps: S11: continuously record the real-time fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the sensor network, and store them in the local server of the optical cable monitoring platform, wherein the total number of optical cables in the monitoring area is G, and g is the index of the optical cable; The historical fiber fingerprint data packet of the historical operation status includes: a historical fiber fingerprint data master packet and a historical fiber fingerprint data sub-packet; the historical fiber fingerprint data master packet includes: optical signal data of the optical cable at different time nodes; the historical fiber fingerprint data sub-packet includes: R types of operation status characterization parameters of the optical cable at different time nodes; S12: Preset a unit time step, extract the historical fiber fingerprint data packet in the local server of the optical cable monitoring platform, cut the historical fiber fingerprint data packet according to the unit time step, and obtain a historical cutting time series data segment, wherein the unit time step is 10 seconds; the historical cutting time series data segment includes: a historical cutting time series data mother segment and a historical cutting time series data sub-segment; S13: At the same time, the mother segment of the historical cutting time series data is scattered; S14: Time domain performance of optical signal data in the mother segment of statistical historical cutting time series data , where t is the time axis consisting of historical unit time steps.

3. The optical cable online monitoring method based on time series big data analysis according to claim 2 is characterized in that: S1 also includes the following specific steps: S15: Convert the optical signal data in the mother segment of the historical cutting time series data into the frequency domain through Fourier transform, and count the frequency domain performance of the optical signal data in the mother segment of the historical cutting time series data , where f is the frequency axis composed of historical frequency components corresponding to the historical unit time step; S16: performing wavelet transformation and denoising on the mother segment of the historical cutting time series data, wherein the wavelet transformation includes: ； in, is the wavelet coefficient of the optical signal data recorded by the i-th sensor on the g-th optical cable at the scale j; is the wavelet function, j is the scale, and k represents the position; The denoising process comprises: ; for Wavelet coefficients after denoising; Indicates the noise threshold when the scale is j; S17: According to steps S14-S16, the multivariate signal features of the mother segment of the historical cutting time series data are extracted respectively, and the multivariate signal features include: time domain features , frequency domain characteristics and time-frequency domain features ; The extraction strategy of the multivariate signal features specifically includes: ； x is the multivariate signal feature index, ; When x=1, The time domain representation of the optical signal data in the mother segment of the historical cutting time series data , Time domain features ; When x=2, The frequency domain representation of the optical signal data in the mother segment of the historical cutting time series data , is the frequency domain feature ; When x=3, is the wavelet coefficient of the denoised optical signal data in the mother segment of the historical cutting time series data , is the time-frequency domain feature ; When the index of the multivariate signal feature is x, The mean of When the index of the multivariate signal feature is x, The standard deviation of When the index of the multivariate signal feature is x, The maximum and minimum amplitudes of When the index of the multivariate signal feature is x, skewness; When the index of the multivariate signal feature is x, The kurtosis of .

4. The optical cable online monitoring method based on time series big data analysis according to claim 3 is characterized in that S2 include: S21: dividing the historical fiber fingerprint data mother package and the corresponding historical cutting time series data mother segment into a 9:1 ratio to form a training set and a test set; S22: Establishing an optical cable time series data model, and inputting the historical optical fiber fingerprint data mother package, the historical cutting time series data mother section and the multivariate signal features in the training set into the input layer of the optical cable time series data model; S23: extracting context information of the historical optical fiber fingerprint data mother package through a recurrent neural network, and outputting a context feature vector of the historical optical fiber fingerprint data mother package; S24: extracting local features of the mother segment of the historical cutting time series data through a convolutional neural network, and outputting an initial feature vector of the mother segment of the historical cutting time series data; S25: extract three high-dimensional feature vectors corresponding to the multivariate signal features through the fully connected layer; S26: The initial feature vector of the mother segment of the historical cutting time series data is fused through the attention mechanism to extract high-dimensional fusion features; S27: Perform nonlinear transformation on the input high-dimensional fusion features through the fully connected layer to extract high-level abstract representation of the features, and use the last fully connected layer to classify and map each mother section of the historical cutting time series data into Y optical cable operation status categories, wherein the activation function of the last fully connected layer is a softmax activation function, and the output of the last fully connected layer is Y probability values; S28: Train the optical cable timing data model and use the test set to test the classification accuracy of the optical cable timing data model. When the accurately classified historical cutting timing data mother section reaches 90%, the training of the optical cable timing data model is completed.

5. The optical cable online monitoring method based on time series big data analysis according to claim 4 is characterized in that: S2 also includes the following specific steps: S29: extracting a real-time optical fiber fingerprint data packet of the real-time operating status of all optical cables in the monitoring area, and cutting the real-time optical fiber fingerprint data packet according to the unit time step to obtain a real-time cutting time series data segment, and breaking up the real-time cutting time series data mother segment, wherein the real-time cutting time series data segment includes: a real-time cutting time series data mother segment and a real-time cutting time series data sub-segment; S210: Inputting the real-time cutting time series data mother section into the trained optical cable time series data model, and outputting the optical cable operation status category to which the real-time cutting time series data mother section belongs.

6. The optical cable online monitoring method based on time series big data analysis according to claim 5 is characterized in that: S3 also includes the following specific steps: S35: Evaluate the set of interference points The comprehensive interference superposition coefficient of each adjacent interference point combination is calculated, and the average level of the comprehensive interference superposition coefficient is calculated, and each adjacent interference point combination whose comprehensive interference superposition coefficient is greater than the average level is selected, and at the same time, the optical cable operation status category of an interference point farther from the optical fiber transmitting end in the adjacent interference point combination is reduced from the yth category to the y-1th category; S36: Traverse the cable operation status category set of the g-th optical cable , conduct interference superposition analysis; S37: After the traversal is completed, the optical cable operation status categories of all interference points on the g-th optical cable are output.

7. The optical cable online monitoring method based on time series big data analysis according to claim 6 is characterized in that: S4 includes the following steps: S41: extracting the real-time cutting time series data sub-segment of the interference point corresponding to the optical cable operation status category of special concern on each optical cable, wherein the confirmation strategy of the optical cable operation status category of special concern on each optical cable includes: arranging the number of occurrences of each optical cable operation status category in the historical optical fiber fingerprint data packet on each optical cable in descending order, and updating the sequence of the number of occurrences of each optical cable operation status category in real time, intercepting the top three optical cable operation status categories with the highest number of occurrences on each optical cable as the optical cable operation status categories of special concern, and the number of interference points corresponding to each optical cable operation status category of special concern is respectively ; S42: on each optical cable, calculating and obtaining risk probability values ​​of R types of operation status characterization parameters of the optical cable operation status category of particular concern, to form a risk probability value set, , where a is the index of the cable operation status category of special concern, Represents the risk probability value in the R-type operating status characterization parameters of the optical cable operating status category of special concern a.

8. The optical cable online monitoring method based on time series big data analysis according to claim 7 is characterized in that: In S422, the calculation strategy of the curve deviation is: ； in, For the The curve deviation of the geometric characteristic curve of the rth operating state characterization parameter at the interference point; It represents the data volume of the parameters representing the rth operating state at the time point T in the unit time step; It represents the mean value of the data volume of the rth operating state characterization parameter when the optical cable leaves the factory.

9. The optical cable online monitoring method based on time series big data analysis according to claim 8 is characterized in that: S5 includes the following specific steps: S51: Extract the risk probability value of the optical cable operation status category of special concern a The corresponding operating status characterization parameter is selected as the risk characterization parameter of the optical cable operating status category of special concern a; S52: Visualize the risk characterization parameters of all optical cable operation status categories of special concern, and generate an optical cable monitoring report, wherein the optical cable monitoring report also includes: the optical cable position of the interference point corresponding to the maximum curve deviation in the curve deviation set composed of the curve deviation of each risk characterization parameter.

10. An optical cable online monitoring system based on time series big data analysis, used to implement the optical cable online monitoring method based on time series big data analysis as claimed in any one of claims 1 to 9, characterized in that: The system includes the following modules: Fiber fingerprint data packet division module, model building module, interference superposition analysis module, risk probability value calculation module and optical cable monitoring report generation module; The optical fiber fingerprint data packet division module is used to continuously obtain the real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area and the historical optical fiber fingerprint data packets of the historical operating status, cut the historical optical fiber fingerprint data packets, obtain the historical cutting time series data segments, and pre-process the historical cutting time series data segments; The model building module is used to establish an optical cable timing data model, and to perform coupling analysis on real-time optical fiber fingerprint data packets of the real-time operating status of all optical cables in the monitoring area through the optical cable timing data model; The interference superposition analysis module performs interference superposition analysis on interference points of the same optical cable operation status category on each optical cable according to the coupling analysis result; The risk probability value calculation module is used to extract the real-time cutting time series data sub-segments of the interference points corresponding to the optical cable operation status categories of particular concern on each optical cable, and calculate the risk probability value of each operation status characterization parameter; The optical cable monitoring report generating module determines the risk characterization parameters of the real-time operation status of the optical cable according to the risk probability value, visualizes the risk characterization parameters, and generates an optical cable monitoring report.

Images