Cable state evaluation method and apparatus, and terminal device and storage medium

Abstract

A cable state evaluation method and apparatus, and a terminal device and a storage medium. The method comprises: acquiring sensing data by means of an optical fiber sensor; dividing sampling points, and dividing sampling points on the same optical fiber segment into one group; integrating sensing data of each group of sampling points and inputting same into an optical fiber segment state evaluation model, so as to obtain a state evaluation result of a corresponding optical fiber segment; and on the basis of state evaluation results of all optical fiber segments, obtaining a cable state evaluation result of an entire optical fiber. Data collection, processing and evaluation can be automatically performed, thereby reducing manual intervention and improving the work efficiency and the accuracy of cable state evaluation results.

Description

Cable state evaluation method and device, terminal equipment and storage medium TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent cable monitoring and evaluation, and particularly relates to a cable state evaluation method and device, terminal equipment and storage medium. BACKGROUND

[0002] With the continuous expansion and upgrading of power and communication networks, cables, as one of the key infrastructures, bear the important task of transmitting information and energy. However, cables are susceptible to stress concentration, temperature abnormalities, vibration damage and other problems during operation due to external environmental factors, load changes and other factors, which can affect the safety and reliability of the cables. In traditional cable monitoring methods, a combination of regular inspection and remote monitoring is usually used. This method has the following main problems:

[0003] Firstly, regular inspection has the defects of long cycle and many blind spots, and cannot achieve real-time monitoring of the cable state. In complex environments, cable damage or abnormal state may not be easily discovered, and once a fault occurs, a lot of time and manpower is often needed to troubleshoot and repair.

[0004] Secondly, although remote monitoring can improve the efficiency of cable monitoring to some extent, due to the limitations of monitoring equipment and the limitations of data processing and analysis, the actual safety state of the cable cannot be accurately determined.

[0005] Therefore, in order to solve the problems existing in the traditional cable safety evaluation method, there is an urgent need for an intelligent and real-time monitoring cable safety evaluation technology. SUMMARY

[0006] The present application provides a cable state evaluation method, device, terminal equipment and storage medium to solve the technical problems of long time-consuming regular inspection and inaccurate remote monitoring.

[0007] In order to solve the above technical problems, the present application provides a cable state evaluation method, the cable is arranged on a plurality of towers, and an optical fiber is arranged in the cable; a plurality of optical fiber sensors are arranged in the optical fiber; the optical fiber includes a plurality of optical fiber segments;

[0008] The cable state evaluation method comprises:

[0009] The optical fiber sensor acquires sensing data of all sampling points on the entire optical fiber; wherein the sensing data includes strain data, temperature data, vibration frequency data and vibration amplitude data;

[0010] According to the vibration frequency data, all the sampling points are divided to obtain a plurality of sampling point groups; wherein the sampling points in each sampling point group are located on the same optical fiber segment;

[0011] For each sampling point group, all the sensing data of the sampling point group are integrated to obtain integrated data of the corresponding optical fiber segment;

[0012] For each integrated data, the integrated data is input into the trained optical fiber segment state evaluation model to obtain an optical fiber segment state evaluation result of the corresponding optical fiber segment; wherein the optical fiber segment state evaluation result is: normal, icing, dancing, lightning stroke, heating or broken strand;

[0013] According to the optical fiber segment state evaluation results of all the optical fiber segments, a cable state evaluation result of the whole optical fiber is obtained;

[0014] The training process of the optical fiber segment state evaluation model comprises:

[0015] Obtain sample sensing data of a plurality of sample optical fiber segments; wherein the sample optical fiber segments include optical fiber segments with states of normal, icing, dancing, lightning stroke, heating and broken strand;

[0016] For each optical fiber segment, the sample sensing data of the optical fiber segment are integrated to obtain sample integrated data; and the sample integrated data are labeled according to the state of the optical fiber segment; wherein the types of the labels include: normal, icing, dancing, lightning stroke, heating and broken strand;

[0017] The optical fiber segment state evaluation model is trained according to the sample integrated data with labels to obtain the trained optical fiber segment state evaluation model.

[0018] As a preferred solution, according to the vibration frequency data, all the sampling points are divided to obtain a plurality of sampling point groups, comprising:

[0019] According to the jump positions of the vibration frequency data, the sampling points are divided by taking the jump positions as division points, and the sampling points between two adjacent jump positions form a group, to obtain a plurality of sampling point groups.

[0020] As a preferred solution, the expression of the strain data of the sampling point group is: [ε1,ε2,ε3,...,ε i ,...,ε N ];

[0021] In the formula, N represents the number of sampling points of the sampling point group; ε i represents the strain data collected by the i-th sampling point;

[0022] The expression of the temperature data of the sampling point group is: [T1,T2,T3,...,Ti ..., T N ];

[0023] wherein, T i represents temperature data collected at the i-th sampling point;

[0024] The expression of the vibration frequency data of the sampling point group is: [f1, f2, f3,..., f i ..., f N ];

[0025] wherein, f i represents vibration frequency data collected at the i-th sampling point;

[0026] The expression of the vibration amplitude data of the sampling point group is: [A1, A2, A3,..., A i ..., A N ];

[0027] wherein, A i represents vibration amplitude data collected at the i-th sampling point;

[0028] The integration of all sensing data of the sampling point group to obtain the integrated data corresponding to the optical fiber segment includes:

[0029] The strain data, temperature data, vibration frequency data and vibration amplitude data of the sampling point group are integrated into a vector;

[0030] The vector is subjected to interpolation processing or interception processing so that the data length of the vector is a preset data length, and the processed vector is taken as the integrated data corresponding to the optical fiber segment.

[0031] As a preferred solution, the optical fiber segment state evaluation model is a decision tree model; and the training algorithm of the optical fiber segment state evaluation model is a LightGBM algorithm.

[0032] As a preferred solution, the parameter settings of the LightGBM algorithm include: the maximum number of leaf nodes is 128; the minimum number of samples of leaf nodes is 20; the learning goal is multi-classification; the number of classes of multi-classification is 6; the maximum depth is -1; the learning rate is 0.01; and the minimum sample weight sum of child nodes is 8.

[0033] As a preferred solution, the optical fiber segment is a segment of optical fiber between two adjacent towers;

[0034] After dividing all sampling points according to the vibration frequency data to obtain a plurality of sampling point groups, the method further includes:

[0035] According to a preset tower detail table, the tower number corresponding to each sampling group is determined.

[0036] The cable state evaluation result of the whole optical fiber is obtained according to the optical fiber section state evaluation results of all optical fiber sections, and the method comprises the following steps:

[0037] It is judged whether the optical fiber section state evaluation results of all optical fiber sections are all normal, if yes, it is determined that the cable state evaluation result of the whole optical fiber is normal, and if not, the corresponding tower number is inquired according to the optical fiber section with the abnormal optical fiber section state evaluation result, and the tower number and the corresponding optical fiber section state evaluation result are output.

[0038] On the basis of the above-mentioned embodiment, another embodiment of the present application provides a cable state evaluation device, which comprises an optical fiber sensor, a data processing module and a state evaluation module.

[0039] The optical fiber sensor is used for acquiring sensing data of all sampling points on the whole optical fiber, wherein the sensing data comprises strain data, temperature data, vibration frequency data and vibration amplitude data.

[0040] The data processing module is used for dividing all sampling points according to the vibration frequency data, so as to obtain a plurality of sampling point groups; and for each sampling point group, all sensing data of the sampling point group are integrated, so as to obtain integrated data of the corresponding optical fiber section; wherein the sampling points in each sampling point group are located on the same optical fiber section.

[0041] The state evaluation module is used for inputting the integrated data into the trained optical fiber section state evaluation model for each integrated data, so as to obtain an optical fiber section state evaluation result of the corresponding optical fiber section; and a cable state evaluation result of the whole optical fiber is obtained according to the optical fiber section state evaluation results of all optical fiber sections; wherein the optical fiber section state evaluation result is normal, icing, dancing, lightning stroke, heating or broken strand; the training process of the optical fiber section state evaluation model comprises the following steps: sample sensing data of a plurality of sample optical fiber sections are acquired; wherein the sample optical fiber sections comprise optical fiber sections with normal, icing, dancing, lightning stroke, heating and broken strand states; for each optical fiber section, sample sensing data of the optical fiber section are integrated, so as to obtain sample integrated data; and a label is set for the sample integrated data according to the state of the optical fiber section; wherein the types of the label comprise normal, icing, dancing, lightning stroke, heating and broken strand; the optical fiber section state evaluation model is trained according to the sample integrated data with the set label, so as to obtain the trained optical fiber section state evaluation model.

[0042] As a preferred scheme, the data analysis module is further used for determining the tower number corresponding to each sampling group according to a preset tower detail table.

[0043] The cable state evaluation result of the whole optical fiber is obtained according to the optical fiber section state evaluation results of all optical fiber sections, and the method comprises the following steps:

[0044] If the fiber segment state evaluation results of all the fiber segments are normal, it is determined that the cable state evaluation result of the whole fiber is normal, and if not, the corresponding tower number is queried according to the fiber segment whose fiber segment state evaluation result is not normal, and the tower number and the corresponding fiber segment state evaluation result are output.

[0045] On the basis of the above-mentioned embodiments, a terminal device is provided in another embodiment of the application, which comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the cable state evaluation method described in the above-mentioned embodiments of the application when executing the computer program.

[0046] On the basis of the above-mentioned embodiments, a storage medium is provided in another embodiment of the application, which comprises a stored computer program, wherein the device where the storage medium is located executes the cable state evaluation method described in the above-mentioned embodiments of the application when the computer program runs.

[0047] Compared with the prior art, the embodiments of the application have the following beneficial effects:

[0048] The application obtains sensing data through the optical fiber sensor, divides the sampling points, divides the sampling points on the same fiber segment into a group, integrates the sensing data of each sampling point group, inputs the sensing data into a fiber segment state evaluation model to obtain the state evaluation result of the corresponding fiber segment, and obtains the cable state evaluation result of the whole fiber according to the state evaluation results of all the fiber segments. The application can automatically collect, process and evaluate data, reduces manual intervention, improves work efficiency and the accuracy of the cable state evaluation result. BRIEF DESCRIPTION OF DRAWINGS

[0049] Fig. 1 is a flowchart of a cable state evaluation method according to an embodiment of the application;

[0050] Fig. 2 is a structural schematic diagram of a cable state evaluation device according to an embodiment of the application. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0052] Embodiment one

[0053] Please refer to Figure 1, a flowchart of a cable state evaluation method provided by an embodiment of the present application, the cable is arranged on a plurality of towers, and an optical fiber is arranged in the cable; a plurality of optical fiber sensors are arranged in the optical fiber; the optical fiber comprises a plurality of optical fiber segments;

[0054] The cable state evaluation method comprises:

[0055] S1, obtaining sensing data of all sampling points on the whole optical fiber by the optical fiber sensors; wherein the sensing data comprises strain data, temperature data, vibration frequency data and vibration amplitude data.

[0056] In step S1, the optical fiber sensors are arranged in the cable system to actually collect strain, temperature and vibration state parameter data of the cable.

[0057] It should be noted that the optical fiber sensor comprises a strain detection module and a temperature detection module, adopts Brillouin optical time domain reflectometry (BOTDR), utilizes two channels, one channel monitors a strain sensing core without excess length in the cable center tube, and the other channel monitors a communication core with excess length. The core with excess length is used to measure temperature, and the core without excess length is used for temperature compensation, so as to realize strain measurement of the core without excess length. Finally, strain and temperature data are collected by the two cores without and with excess length. The optical fiber sensor further comprises a vibration monitoring module, which adopts distributed acoustic sensing technology (DAS) and is used to collect vibration frequency and amplitude information of the cable.

[0058] S2, dividing all sampling points according to the vibration frequency data to obtain a plurality of sampling point groups; wherein the sampling points in each sampling point group are located on the same optical fiber segment.

[0059] In step S3, the cable is arranged on a plurality of towers, and the present application divides the optical fiber sensors between adjacent two towers and located on the same optical fiber segment into the same group according to the vibration frequency data, that is, the sampling points, to obtain a plurality of sampling point groups.

[0060] In a preferred embodiment, the dividing all sampling points according to the vibration frequency data to obtain a plurality of sampling point groups comprises:

[0061] According to the jump position of the vibration frequency data, the sampling points are divided by taking the jump position as a division point, the sampling points between adjacent two jump positions are a group, and a plurality of sampling point groups are obtained.

[0062] It should be noted that the vibration frequency of the optical fiber between adjacent towers is different due to different spans, and the jump position of the vibration frequency appears at the tower. The vibration frequency data of DAS can be used to locate the tower of the cable. The span refers to the straight-line distance between adjacent two towers.

[0063] In the embodiment, according to the jump positions of the vibration frequency data, the positions of the towers can be inferred, and the sampling points between two adjacent jump positions are set as a group to obtain a plurality of sampling point groups, each of which is located between two adjacent towers, i.e., on the same fiber segment.

[0064] S3, for each sampling point group, integrating all the sensing data of the sampling point group to obtain integrated data of the corresponding fiber segment.

[0065] In step S3, the collected data can also be first removed of noise and outliers to ensure the accuracy and integrity of the data, and then integrated.

[0066] In a preferred embodiment, the expression of the strain data of the sampling point group is: [ε1, ε2, ε3,..., ε i ,...,ε N ];

[0067] In the formula, N represents the number of sampling points of the sampling point group; ε i represents the strain data collected by the i th sampling point;

[0068] The expression of the temperature data of the sampling point group is: [T1, T2, T3,..., T i ,...,T N ];

[0069] In the formula, T i represents the temperature data collected by the i th sampling point;

[0070] The expression of the vibration frequency data of the sampling point group is: [f1, f2, f3,..., f i ,...,f N ];

[0071] In the formula, f i represents the vibration frequency data collected by the i th sampling point;

[0072] The expression of the vibration amplitude data of the sampling point group is: [A1, A2, A3,..., A i ,...,A N ];

[0073] In the formula, A i represents the vibration amplitude data collected by the i th sampling point;

[0074] The integration of all the sensing data of the sampling point group to obtain the integrated data of the corresponding fiber segment includes:

[0075] The strain data, temperature data, vibration frequency data, and vibration amplitude data of the sampling point group are integrated into a single vector;

[0076] The vector is interpolated or truncated to make its data length a preset length, and the processed vector is used as the integrated data of the corresponding fiber segment.

[0077] In this embodiment, the length of the optical fiber between two adjacent towers is used as one data unit (the data length is the number of sampling points N in the corresponding sampling point group). The strain data [ε1,ε2,ε3,...,ε] are then used. i ,...,ε N Temperature data [T1,T2,T3,...,T] i ,...,T N Vibration frequency data [f1,f2,f3,...,f] i ,...,f N ] and vibration amplitude data [A1,A2,A3,...,A i ,...,A N ], forming an N×4 matrix:

[0078] Further transform this matrix into a vector, forming a 1×4N vector X. i =[ε1,ε2,ε3,...,ε N ,T1,T2,T3,...,T N f1, f2, f3, ..., f N A1, A2, A3, ..., A N Then, the vector is interpolated or truncated to make the length data of all gears consistent.

[0079] X i The data is input into the trained fiber segment state assessment model to obtain the state classification prediction value, which is the result of the fiber segment state assessment.

[0080] S4. For each integrated data, the integrated data is input into the trained fiber segment status assessment model to obtain the fiber segment status assessment result for the corresponding fiber segment; wherein, the fiber segment status assessment result is: normal, icing, galloping, lightning strike, overheating or broken strand.

[0081] The training process of the fiber segment status assessment model includes:

[0082] Acquire sample sensing data of several sample optical fiber segments; wherein, the sample optical fiber segments include optical fiber segments in the states of normal, icing, galloping, lightning strike, overheating, and broken strands;

[0083] For each fiber segment, the sample sensing data of the fiber segment is integrated to obtain sample integrated data; and according to the state of the fiber segment, the sample integrated data is tagged; wherein, the types of tags include: normal, icing, dancing, lightning strike, overheating, and broken strand;

[0084] The fiber segment status assessment model is trained by integrating sample data with pre-set labels, resulting in a trained fiber segment status assessment model.

[0085] In step S4, during the training of the fiber optic segment condition assessment model, a large amount of sample data is collected for different cable conditions, including: normal, icing, galloping, accumulation, heating, and strand breakage. For each condition, strain data, temperature data, vibration frequency data, and vibration amplitude data of several fiber optic segments are collected, and corresponding labels are assigned to the data, including: normal, icing, galloping, accumulation, heating, and strand breakage. The sample data is randomly shuffled, and 90% is selected as the validation set, with the remainder used as the training set.

[0086] In a preferred embodiment, the fiber segment state assessment model is a decision tree model; the training algorithm for the fiber segment state assessment model is the LightGBM algorithm.

[0087] In a preferred embodiment, the parameter settings of the LightGBM algorithm include: a maximum of 128 leaf nodes; a minimum of 20 leaf node samples; a multi-class classification learning objective; 6 categories in the multi-class classification; a maximum depth of -1; a learning rate of 0.01; and a minimum sum of weights for child nodes of 8.

[0088] In this embodiment, the initial model is a simple decision tree model, which uses the initial data of the cable to split nodes and generate a tree structure. Although the performance of the initial model may be low, it provides a foundation for subsequent gradient boosting.

[0089] The LightGBM algorithm was used for training the data, with the following parameters set:

[0090] In a single base classifier (decision tree),

[0091] The maximum number of leaf nodes, num_leaves, is 128. To prevent overfitting, the "leaf-wise tree growth algorithm" is used here.

[0092] Set the minimum number of subsamples in the leaf node to min_data_in_leaf = 20.

[0093] The learning task and corresponding learning objective are specified as multi-class (objective = multiclass).

[0094] The number of categories in the multi-class classification is num_class = 6.

[0095] The maximum depth is set to the default value of -1, meaning there is no limit (max_depth = -1).

[0096] The learning rate is set to learning_rate = 0.01.

[0097] Set the minimum sum of sample weights in the child nodes. If the sum of sample weights in a leaf node is less than this value, the splitting process ends. min_sum_hessian_in_leaf = 8.

[0098] The boosting parameter is set to GBDT, which has a classic and stable effect. To prevent overfitting and improve training speed, the parameters are set to feature_fraction = 0.9, bagging_freq = 1, and bagging_fraction = 0.8. The regularization parameter lambda_l1 is set to 0.1 to reduce overfitting. The other parameters are set to verbosity = -1, nthread = 15, metric = 'multi_logloss', and random_state = 2019.

[0099] With the parameters set as above, the LightGBM model is used to train the model and obtain the fiber segment status assessment model.

[0100] Using the gradient boosting algorithm, more complex models are built by continuously optimizing the residuals. The specific steps are as follows:

[0101] (1) Calculate the residual: Calculate the difference between the current model prediction and the actual value;

[0102] (2) Fitting residuals: Use the new decision tree to fit the residuals to explain as much of the current model’s error as possible;

[0103] (3) Update the model: Add the new tree to the model and use the learning rate to control the update step size to prevent overfitting;

[0104] (4) Repeat the above steps until the preset number of iterations or other stopping conditions are reached.

[0105] A leaf-wise growth strategy is used, selecting the leaf node with the highest gain for splitting each time. This strategy can find the optimal split point more quickly, improving the training speed and accuracy of the model.

[0106] Histogram algorithms are used to discretize features, reducing data memory usage and computational complexity. The specific steps are as follows:

[0107] (1) Feature discretization: Divide continuous features into a fixed number of bins, each bin representing an interval;

[0108] (2) Histogram construction: Calculate the number of samples and gradient statistics in each bucket based on the training data;

[0109] (3) Quickly find the split point: Quickly find the split point with the maximum gain on the histogram;

[0110] (4) This optimization method is applicable to large-scale datasets and can significantly improve model training efficiency.

[0111] During model training, cross-validation is used to evaluate the model's performance and generalization ability. The specific steps are as follows:

[0112] (1) Data partitioning: Divide the dataset into K subsets, one of which is used as the validation set and the remaining subsets are used as the training set;

[0113] (2) Iterative validation: Repeat K times, each time selecting a different subset as the validation set, and using the remaining subset as the training set;

[0114] (3) Performance evaluation: Calculate the performance index for each validation and take the average value as the overall performance of the model. Further optimize the model by adjusting hyperparameters (such as learning rate, tree depth, number of leaf nodes, etc.) to improve accuracy and prevent overfitting.

[0115] The trained LightGBM model is then evaluated using a validation set to ensure its stability and accuracy. Evaluation methods include:

[0116] (1) Performance metrics calculation: Use the validation set to calculate the model's performance metrics, such as accuracy, recall, F1 score, etc.

[0117] (2) Model stability test: Verify the model’s performance on different datasets to ensure its stability and robustness;

[0118] (3) Deployment can be carried out based on the evaluated model.

[0119] S5. Based on the fiber segment status assessment results of all fiber segments, obtain the cable status assessment results of the entire fiber.

[0120] It should be noted that you should not buy any device that can determine the safety status of cables, promptly identify abnormalities in the cable system, and take necessary measures, such as repairing, replacing, or reinforcing cable sections, to improve the safety and reliability of the cable system.

[0121] In a preferred embodiment, the optical fiber segment is a section of optical fiber between two adjacent towers;

[0122] After dividing all sampling points into several sampling point groups based on the vibration frequency data, the method further includes:

[0123] Based on the pre-set tower list, determine the tower number corresponding to each sampling group;

[0124] The process of obtaining the overall cable status assessment result based on the status assessment results of all fiber segments includes:

[0125] Determine whether the fiber segment status assessment results for all fiber segments are normal. If yes, then determine that the overall cable status assessment result for the entire fiber is normal. If not, then for the fiber segments whose status assessment results are abnormal, query the corresponding tower number and output the tower number and the corresponding fiber segment status assessment result.

[0126] In this embodiment, the physical span between towers is obtained through a preset tower schedule. By connecting towers for positioning, the fiber length can be mapped one-to-one with the tower number, span by span. The tower number is recorded in the tower schedule.

[0127] During the cable status assessment, if all fiber optic segments show normal status, the entire cable is considered normal, confirming the overall cable status assessment is normal. If any fiber optic segments are abnormal, the status assessment results for those segments and their corresponding tower numbers are output so maintenance personnel can pinpoint the location of the anomaly and proceed with subsequent repairs.

[0128] It should be noted that span is the straight-line distance between two adjacent towers. The cable between the two towers has a certain sag, therefore the cable length is greater than the span. The optical fiber also has a certain excess length within the cable, therefore the optical fiber length is greater than the cable length, and also greater than the span. As the transmission distance increases, the accumulated error between the optical fiber and the span affects the positioning of distant towers in existing technologies. However, the span-by-span continuous tower positioning proposed in this invention can ensure that the optical fiber length corresponds to the tower span for each span, and the error between the optical fiber and the span will not accumulate. Therefore, the positioning of distant towers will not be affected, and it is more conducive to accurately locating abnormal positions.

[0129] Example 2

[0130] Please refer to Figure 2, which is a structural schematic diagram of a cable condition assessment device provided in an embodiment of the present invention, including: an optical fiber sensor, a data processing module, and a condition assessment module;

[0131] The fiber optic sensor is used to acquire sensing data from all sampling points along the entire fiber optic cable; wherein the sensing data includes: strain data, temperature data, vibration frequency data, and vibration amplitude data.

[0132] The data processing module is used to divide all sampling points into several sampling point groups based on the vibration frequency data; for each sampling point group, it integrates all the sensing data of the sampling point group to obtain the integrated data of the corresponding optical fiber segment; wherein the sampling points in each sampling point group are located on the same optical fiber segment.

[0133] The state assessment module is used to input the integrated data into the trained fiber segment state assessment model for each integrated data point to obtain the fiber segment state assessment result for the corresponding fiber segment; based on the fiber segment state assessment results of all fiber segments, the cable state assessment result for the entire fiber is obtained; wherein, the fiber segment state assessment result is: normal, icing, galloping, lightning strike, overheating, or broken strand; the training process of the fiber segment state assessment model includes: acquiring sample sensor data of several sample fiber segments; wherein, the sample fiber segments include fiber segments with states of normal, icing, galloping, lightning strike, overheating, and broken strand; for each fiber segment, integrating the sample sensor data of the fiber segment to obtain sample integrated data; and setting labels for the sample integrated data according to the state of the fiber segment; wherein, the types of labels include: normal, icing, galloping, lightning strike, overheating, and broken strand; training the fiber segment state assessment model based on the sample integrated data with set labels to obtain the trained fiber segment state assessment model.

[0134] In a preferred embodiment, the data analysis module is further configured to determine the tower number corresponding to each sampling group based on a preset tower list.

[0135] The process of obtaining the overall cable status assessment result based on the status assessment results of all fiber segments includes:

[0136] Determine whether the fiber segment status assessment results for all fiber segments are normal. If yes, then determine that the overall cable status assessment result for the entire fiber is normal. If not, then for the fiber segments whose status assessment results are abnormal, query the corresponding tower number and output the tower number and the corresponding fiber segment status assessment result.

[0137] Example 3

[0138] Accordingly, embodiments of the present invention provide a terminal device, the terminal device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the cable status assessment method described in the above embodiments of the invention.

[0139] Example 4

[0140] Accordingly, embodiments of the present invention provide a storage medium, the storage medium including a stored computer program, wherein, when the computer program is running, it controls the device where the storage medium is located to execute the cable status assessment method described in the above embodiments of the invention.

[0141] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0142] Those skilled in the art will clearly understand that, for convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0143] The terminal device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and memory.

[0144] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the device, connecting various parts of the device via various interfaces and lines.

[0145] The memory can be used to store the computer program. The processor implements various functions of the device by running or executing the computer program stored in the memory and calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0146] The storage medium is a storage medium in which the computer program is stored. When executed by a processor, the computer program can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0147] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A cable condition assessment method characterized by, The cable is arranged on a plurality of towers, and an optical fiber is arranged in the cable; a plurality of optical fiber sensors are arranged in the optical fiber; and the optical fiber comprises a plurality of optical fiber segments; The cable state evaluation method comprises: Obtaining sensing data of all sampling points on the whole optical fiber through the optical fiber sensors; wherein the sensing data comprises strain data, temperature data, vibration frequency data and vibration amplitude data; According to the vibration frequency data, all sampling points are divided to obtain a plurality of sampling point groups; wherein the sampling points in each sampling point group are located on the same optical fiber segment; For each sampling point group, all sensing data of the sampling point group are integrated to obtain integrated data of the corresponding optical fiber segment; For each integrated data, the integrated data is input into the trained optical fiber segment state evaluation model to obtain an optical fiber segment state evaluation result of the corresponding optical fiber segment; wherein the optical fiber segment state evaluation result is normal, icing, dancing, lightning stroke, heating or broken strand; According to the optical fiber segment state evaluation results of all optical fiber segments, a cable state evaluation result of the whole optical fiber is obtained; The training process of the optical fiber segment state evaluation model comprises: Obtaining sample sensing data of a plurality of sample optical fiber segments; wherein the sample optical fiber segments include optical fiber segments with states of normal, icing, dancing, lightning stroke, heating and broken strand; For each optical fiber segment, the sample sensing data of the optical fiber segment are integrated to obtain sample integrated data; and a label is set for the sample integrated data according to the state of the optical fiber segment; wherein the types of the label include normal, icing, dancing, lightning stroke, heating and broken strand; The optical fiber segment state evaluation model is trained according to the sample integrated data with the set label to obtain a trained optical fiber segment state evaluation model.

2. The cable condition assessment method of claim 1, wherein, According to the vibration frequency data, all sampling points are divided to obtain a plurality of sampling point groups, which comprises: According to the jump positions of the vibration frequency data, the sampling points are divided by taking the jump positions as division points, and the sampling points between adjacent two jump positions form a group to obtain a plurality of sampling point groups.

3. The cable condition assessment method of claim 1, wherein, The expression of the strain data of the sampling point group is: [ε1, ε2, ε3, …, ε i ,...,ε N ] In the formula, N represents the number of sampling points of the sampling point group; ε i represents the strain data collected by the i th sampling point; The temperature data expression of the sampling point group is: [T1, T2, T3,..., T i ,...,T N ] ; In the formula, T i represents the temperature data collected at the ith sampling point; The vibration frequency data expression of the sampling point group is: [f1,f2,f3,...,f i ,...,f N ] ; In the formula, f i represents the vibration frequency data collected at the i th sampling point; The vibration amplitude data expression of the sampling point group is: [A1,A2,A3,...,A i ,...,A N ] In the formula, A i represents the vibration amplitude data collected at the i th sampling point; The integration of all sensing data of the sampling point group to obtain the integrated data of the corresponding optical fiber segment comprises: The strain data, temperature data, vibration frequency data and vibration amplitude data of the sampling point group are integrated into a vector; The vector is subjected to interpolation processing or interception processing to make the data length of the vector be a preset data length, and the processed vector is taken as the integrated data of the corresponding optical fiber segment.

4. The cable condition assessment method of claim 1, wherein, The optical fiber segment state evaluation model is a decision tree model; and the training algorithm of the optical fiber segment state evaluation model is a LightGBM algorithm.

5. The cable condition assessment method of claim 4, wherein, The parameter setting of the LightGBM algorithm comprises: the maximum number of leaf nodes is 128; the minimum number of leaf nodes is 20; the learning goal is multi-classification; the number of classes of multi-classification is 6; the maximum depth is -1; the learning rate is 0.01; and the minimum sample weight sum of child nodes is 8.

6. The cable condition assessment method of claim 1, wherein, The optical fiber segment is a segment of optical fiber between two adjacent towers; After dividing all the sampling points according to the vibration frequency data to obtain a plurality of sampling point groups, the method further comprises: According to a preset tower detail table, determine the tower number corresponding to each sampling group; The cable state evaluation result of the whole optical fiber is obtained according to the optical fiber segment state evaluation results of all the optical fiber segments, comprising: determining whether the optical fiber segment state evaluation results of all the optical fiber segments are normal, if so, determining that the cable state evaluation result of the whole optical fiber is normal, if not, according to the optical fiber segments with abnormal optical fiber segment state evaluation results, querying the corresponding tower number, and outputting the tower number and the corresponding optical fiber segment state evaluation result.

7. A cable condition evaluation apparatus characterized by comprising: comprising: an optical fiber sensor, a data processing module and a state evaluation module; The optical fiber sensor is configured to acquire sensing data of all sampling points on the whole optical fiber; wherein the sensing data comprises strain data, temperature data, vibration frequency data and vibration amplitude data; The data processing module is configured to divide all the sampling points according to the vibration frequency data to obtain a plurality of sampling point groups; for each sampling point group, integrate all the sensing data of the sampling point group to obtain integrated data of the corresponding optical fiber segment; wherein the sampling points in each sampling point group are located on the same optical fiber segment; The state evaluation module is configured to input each integrated data into a trained optical fiber segment state evaluation model to obtain an optical fiber segment state evaluation result of the corresponding optical fiber segment; and obtain a cable state evaluation result of the whole optical fiber according to the optical fiber segment state evaluation results of all the optical fiber segments; wherein the optical fiber segment state evaluation result is normal, icing, dancing, lightning strike, heating or broken strand; the training process of the optical fiber segment state evaluation model comprises: acquiring sample sensing data of a plurality of sample optical fiber segments; wherein the sample optical fiber segments include optical fiber segments with normal, icing, dancing, lightning strike, heating and broken strand states; for each optical fiber segment, integrate the sample sensing data of the optical fiber segment to obtain sample integrated data; and set a label for the sample integrated data according to the state of the optical fiber segment; wherein the types of the label include normal, icing, dancing, lightning strike, heating and broken strand; train the optical fiber segment state evaluation model according to the sample integrated data with the set label to obtain the trained optical fiber segment state evaluation model.

8. The cable condition assessment apparatus of claim 7, wherein, The data analysis module is further configured to determine the tower number corresponding to each sampling group according to a preset tower detail table; The cable state evaluation result of the whole optical fiber is obtained according to the optical fiber segment state evaluation results of all the optical fiber segments, comprising: determining whether the optical fiber segment state evaluation results of all the optical fiber segments are normal, if so, determining that the cable state evaluation result of the whole optical fiber is normal, if not, according to the optical fiber segments with abnormal optical fiber segment state evaluation results, querying the corresponding tower number, and outputting the tower number and the corresponding optical fiber segment state evaluation result.

9. A terminal device, comprising: A computer-readable storage medium storing a computer program configured to, when executed by a processor, cause the processor to perform the cable condition evaluation method according to any one of claims 1 to 6.

10. A storage medium, characterized by The storage medium includes a stored computer program, wherein the computer program, when executed, controls a device in which the storage medium is located to perform the cable condition evaluation method according to any one of claims 1 to 6.

Images