A method and system for monitoring and maintaining the operation status of power cables

Abstract

The present invention relates to the field of power monitoring technology, and specifically to a method and system for monitoring and maintaining the operating status of power cables, comprising the following steps: sending a test signal of a predetermined frequency and strength to a target cable; capturing the test signal transmitted through the cable at the other end of the cable, and predicting the degree of cable aging through an attenuation index; using TDR technology to analyze the time delay and strength of the reflected signal in the cable to determine the preliminary location of potential damage or discontinuity in the cable; performing a multi-frequency signal test on the cable to analyze the transmission characteristics of the cable for each frequency signal, thereby obtaining the fault type; and using a multi-task learning model for comprehensive analysis to determine the aging status of the cable and the specific damage location. The present invention can not only detect obvious damage on the outside of the cable, but also identify minor damage or aging phenomena inside the cable that are invisible to the naked eye, reducing the need for manual operation and thus improving the efficiency and reliability of detection.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring, and in particular to a method and system for monitoring and maintaining the operating status of a power cable. Background Art

[0002] Power cables, as a vital component of the power system, play a crucial role in the transmission and distribution of electrical energy. As power systems evolve towards higher reliability and efficiency, cable health monitoring and maintenance have become critical to ensuring stable operation. However, traditional cable monitoring methods rely primarily on regular physical inspections and localized electrical testing. These methods are not only time-consuming and labor-intensive, but also often fail to detect even minor damage or early signs of aging within the cable, increasing the risk of potential cable failure.

[0003] Furthermore, traditional methods have limitations in locating cable damage. Because cables are often buried underground or hidden within other facilities, direct access for inspection is difficult and restrictive. This results in inefficient manual inspections and makes it difficult to conduct a comprehensive inspection of the entire cable line. Even minor damage can worsen over time, ultimately affecting cable performance and safety, and potentially causing unexpected power system outages.

[0004] With technological advancements, emerging monitoring technologies, such as time-domain reflectometry (TDR) and frequency response analysis, are being applied to cable health monitoring. These technologies can provide more detailed information about cable health, including damage locations and aging levels, but they often require complex data processing and analysis methods. Therefore, effectively integrating and analyzing data from these advanced technologies, as well as improving the accuracy and efficiency of cable damage detection and aging assessment, have become pressing technical challenges in the field of cable health monitoring. Summary of the Invention

[0005] Based on the above objectives, the present invention provides a method and system for monitoring and maintaining the operating status of a power cable.

[0006] A method for monitoring and maintaining the operating status of a power cable comprises the following steps:

[0007] S1: Signal transmission, sending a test signal of predetermined frequency and intensity to the target cable;

[0008] S2: Signal capture and attenuation analysis: Capture the test signal transmitted through the cable at the other end of the cable or at a test point set up in a segment. Compare the attenuation between the original signal and the received signal, and predict the degree of cable aging based on the attenuation index.

[0009] S3: Time Domain Reflectometry (TDR) technology uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to determine the preliminary location of potential damage or discontinuity in the cable;

[0010] S4: Frequency response analysis: Perform multi-frequency signal testing on the cable to analyze the cable's transmission characteristics for each frequency signal. This helps identify the impact of cable aging or minor damage on the transmission of different frequency signals, thereby determining the fault type.

[0011] S5: Based on the analysis results of S2, S3, and S4, a multi-task learning model is used for comprehensive analysis to determine the aging condition of the cable and the specific damage location.

[0012] Furthermore, the S1 specifically includes:

[0013] S11, configuring a signal generator, wherein the signal generator includes an adjustable frequency and an adjustable output intensity, and the frequency range and intensity level of the test signal are preset according to the characteristics of the cable and the detection requirements;

[0014] S12, connect the signal generator to the target cable through a cable test head or adapter that matches the cable;

[0015] S13, generating a pulse waveform using a modulation function of a signal generator;

[0016] S13, test sequence initialization: start the test sequence through the signal generator, send test signals of predetermined frequency and intensity in sequence, and monitor the stability and continuity of the signal to ensure the signal quality of the entire test process.

[0017] Furthermore, the S2 specifically includes:

[0018] S21, set up detection points at the other end of the cable or in sections according to the cable length and structure, and install an oscilloscope at each detection point;

[0019] S22, while transmitting the test signal, ensuring that the oscilloscope at the detection point synchronously records the signal through the synchronous output port of the signal generator;

[0020] S23, record the received signal through an oscilloscope, compare it with the original signal, analyze the attenuation degree between the two, use the attenuation model to convert the attenuation degree into the cable aging index, and judge the aging degree of the cable.

[0021] Furthermore, the attenuation model is expressed as: ,in, Is in the distance The signal attenuation at is the basic attenuation, taking into account factors such as cable connection and initial loss, is the attenuation coefficient per unit length, The length of the cable is converted into the aging index of the cable, which includes:

[0022] Establish a relationship model between attenuation and aging: Based on historical monitoring data, establish a relationship model between cable attenuation and aging degree;

[0023] Attenuation parameter calibration: When the cable is first put into use, measure and record the basic attenuation and the attenuation coefficient per unit length , serving as a benchmark for future attenuation measurements;

[0024] Periodic attenuation measurement: Regularly measure the actual attenuation of the cable using a test signal , and record the distance ;

[0025] Aging degree calculation: Based on the relationship model, the actual measured attenuation Converting to an aging index for the cable involves calculating the difference between the actual attenuation and the base attenuation, and correlating the difference with the degree of aging.

[0026] Furthermore, the S3 specifically includes:

[0027] S31, a short pulse or step signal is sent to the cable through the TDR device, and the signal transmission time is recorded at the device end;

[0028] S32, Reflected Signal Reception and Recording: Use the same TDR equipment to capture reflected signals caused by discontinuities in the cable (such as damaged or aged areas) and record the arrival time and strength of the reflected signals;

[0029] S33, calculate the time delay between the reflected signal and the original transmitted signal, the time delay Obtained directly from TDR equipment;

[0030] S34, Determine the location of the break or discontinuity : ,in, is the distance from the start of the cable to the discontinuity, is the propagation speed of the signal in the cable (depending on the dielectric constant of the cable), is the time delay from transmitting the signal to receiving the reflected signal;

[0031] S35, by analyzing the intensity of the reflected signal, a preliminary judgment is made on the nature of the discontinuity. A high reflection intensity indicates severe damage or discontinuity.

[0032] Furthermore, the S4 specifically includes:

[0033] S41, determine the test frequency, starting from low frequency and gradually increasing to high frequency to evaluate the frequency response characteristics of the cable;

[0034] S42, for each selected test frequency, using a signal generator to generate a test signal of the corresponding frequency, and transmitting the test signal through the cable, and capturing the transmitted signal at the other end of the cable or at a detection point set in a segment;

[0035] S43, Phase Analysis, for each test frequency, measures and records the phase change of the signal transmitted through the cable. The phase change is measured by comparing the phase difference between the transmitted signal and the received signal.

[0036] S44, drawing a frequency response curve of the cable based on the phase analysis result and the attenuation analysis of the attenuation model, including a curve showing attenuation variation with frequency and a curve showing phase variation with frequency;

[0037] S45 identifies the impact of cable aging or minor damage on signal transmission by analyzing the frequency response curve characteristics. Different types of faults (such as insulation aging, minor cracks, or connector damage) exhibit different characteristics on the frequency response curve.

[0038] Furthermore, the phase change is calculated as: ;in, is the frequency Phase change (degrees) under and are the phases of the signals at the receiving and transmitting ends respectively.

[0039] Furthermore, the S5 specifically includes:

[0040] S51, Data Preprocessing: Preprocess attenuation, TDR results, frequency response curve characteristics, and phase change, including denoising, normalization, and feature extraction to ensure that the data is suitable for model input;

[0041] S52, define the task: define cable aging condition analysis and damage location determination as two related learning tasks. The aging condition is considered a classification problem (for example, classifying it into several levels based on the degree of aging), and the damage location is considered a localization problem (for example, predicting the specific location of the damage or the range where the damage exists).

[0042] S53, Design of Multi-task Learning Model: A multi-task learning model processes multiple types of input data and can output the results of multiple tasks. It uses a shared hidden layer to extract common features of all tasks and then provides a dedicated output layer for each task. This is achieved through a multi-task learning model based on deep learning.

[0043] S54, training model: Use a training dataset containing attenuation, TDR results, frequency response curve characteristics, and phase change to train the model. During the training process, it is necessary to simultaneously optimize the loss functions of the two tasks. This is achieved through a weighted loss function, where the weights reflect the importance of different tasks.

[0044] S55, Model Evaluation and Adjustment: Use the validation set to evaluate the model's performance on aging condition analysis and damage location determination tasks, and adjust the model structure, parameters, and task weights based on the evaluation results;

[0045] S56, apply the model for analysis: Apply the trained deep learning-based multi-task learning model to new test data, comprehensively analyze the aging condition of the cable and the specific damage location, and output an assessment result including the aging condition and prediction of the damage location.

[0046] Furthermore, the multi-task learning model based on deep learning specifically includes:

[0047] Shared layer: consists of several convolutional layers or recurrent layers, used to extract common features of all input data;

[0048] Task-specific layers: Each task (aging analysis and damage location determination) has its own set of fully connected layers to learn the task representation from the features extracted by the shared layers;

[0049] The input data includes attenuation data, TDR results, frequency response characteristics and phase difference information;

[0050] Shared layer feature extraction: As input data, the feature extraction of the shared layer is expressed as ,in is the function of the shared layer, are the parameters of the shared layer;

[0051] Task-specific layer output: For the aging analysis task, let its output be ,in, is the function of the aging status analysis task layer, is the parameter of the task layer. For the damage location determination task, let its output be ,in, is a function of the damage location determination task layer, is the parameter of the task layer;

[0052] Loss function: The model loss function is the weighted sum of the two task loss functions, expressed as: ,in and are the loss functions for aging condition analysis and damage location determination tasks, and are the true labels of the corresponding tasks, and is the weight used to balance the importance of different tasks.

[0053] A system for monitoring and maintaining the operating status of a power cable, used to implement the above-mentioned method for monitoring and maintaining the operating status of a power cable, includes the following modules:

[0054] Signal transmitting module: sends a test signal of predetermined frequency and strength to the target cable;

[0055] Signal capture and attenuation analysis module: Set at the other end or segment of the cable, it is used to capture the test signal transmitted through the cable and analyze the attenuation between the original signal and the received signal, and predict the aging degree of the cable based on the attenuation index;

[0056] Time Domain Reflectometry (TDR) module: uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to preliminarily determine the location of potential damage or discontinuity in the cable;

[0057] Frequency response analysis module: performs multi-frequency signal testing on the cable and analyzes the transmission characteristics of the cable for signals of different frequencies. It identifies the impact of cable aging or minor damage on signal transmission and determines the type of fault.

[0058] Comprehensive analysis and diagnosis module: A multi-task learning model is used to perform comprehensive analysis to determine the aging condition of the cable and the specific damage location.

[0059] Beneficial effects of the present invention:

[0060] By sending predefined test signals to the cable and analyzing its attenuation, phase change, frequency response and other characteristics during cable transmission, it is possible to automatically and accurately identify aging and damage in the cable. Compared with traditional manual inspections, such as physical inspections or simple visual inspections, it has significant advantages. It can not only detect obvious damage on the outside of the cable, but also identify tiny damage or aging inside the cable that is invisible to the naked eye, reducing the need for manual operation and thus improving the efficiency and reliability of detection.

[0061] This invention achieves comprehensive monitoring of cable status by integrating cable attenuation data, time domain reflectometry (TDR) results, frequency response characteristics, and phase difference information. Compared with single-metric analysis, this multi-dimensional analysis method can more comprehensively capture subtle changes in cable aging or damage, thereby significantly improving the accuracy of diagnosing cable aging status and damage location. TDR technology can accurately locate physical damage to the cable, while frequency response analysis helps to reveal changes in cable material characteristics caused by aging. The combination of the two can provide a more accurate assessment of cable health status.

[0062] This paper uses a multi-task learning (MTL) model to simultaneously handle cable aging analysis and damage location determination within the same learning framework. This not only significantly improves data analysis efficiency, but also enhances the model's generalization ability across tasks by jointly learning the correlations between tasks. This shared information learning approach is particularly suitable for cable monitoring, where cable aging and specific damage are often interrelated. The MTL model can better capture these inherent connections, improving the accuracy and stability of predictions. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0064] Figure 1 This is a flowchart of an operation and maintenance method according to an embodiment of the present invention;

[0065] Figure 2 This is a schematic diagram of the functional modules of the operation and maintenance system according to an embodiment of the present invention;

[0066] Figure 3 Schematic diagram of a curve showing attenuation variation with frequency according to an embodiment of the present invention;

[0067] Figure 4 Schematic diagram of a curve showing phase variation versus frequency according to an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0069] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0070] like Figure 1-4 As shown, a method for monitoring and maintaining the operating status of a power cable includes the following steps:

[0071] S1: Signal transmission, sending a test signal of predetermined frequency and intensity to the target cable;

[0072] S2: Signal capture and attenuation analysis: Capture the test signal transmitted through the cable at the other end of the cable or at a test point set up in a segment. Compare the attenuation between the original signal and the received signal, and predict the degree of cable aging based on the attenuation index.

[0073] S3: Time Domain Reflectometry (TDR) technology uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to determine the preliminary location of potential damage or discontinuity in the cable;

[0074] S4: Frequency response analysis: Perform multi-frequency signal testing on the cable to analyze the cable's transmission characteristics for each frequency signal. This helps identify the impact of cable aging or minor damage on the transmission of different frequency signals, thereby determining the fault type.

[0075] S5: Based on the analysis results of S2, S3, and S4, a multi-task learning model is used for comprehensive analysis to determine the aging condition of the cable and the specific damage location.

[0076] S1 specifically includes:

[0077] S11, configure the signal generator, which includes adjustable frequency and adjustable output intensity. Preset the test signal frequency range and intensity level based on the cable characteristics and testing requirements. Select the Keysight 33600A series waveform generator.

[0078] S12, connect the signal generator to the target cable through a cable test head or adapter that matches the cable;

[0079] S13, utilizing the modulation function of the signal generator to generate a pulse waveform. The signal waveform may include a sine wave, square wave, pulse wave, or a custom waveform. Considering that the present invention is aimed at monitoring and maintaining the operating status of power cables, including detecting aging and minor damage, a pulse wave is the most suitable choice. The application of pulse waves in TDR technology can efficiently and accurately locate fault points in cables, which is crucial for maintenance and repair work.

[0080] S13, test sequence initialization: start the test sequence through the signal generator, send test signals of predetermined frequency and intensity in sequence, and monitor the stability and continuity of the signal to ensure the signal quality of the entire test process.

[0081] S2 specifically includes:

[0082] S21, set up detection points at the other end of the cable or in sections according to the cable length and structure, and install an oscilloscope at each detection point;

[0083] S22, while transmitting the test signal, ensuring that the oscilloscope at the detection point synchronously records the signal through the synchronous output port of the signal generator;

[0084] S23, record the received signal through an oscilloscope, compare it with the original signal, analyze the attenuation degree between the two, use the attenuation model to convert the attenuation degree into the cable aging index, and judge the aging degree of the cable.

[0085] The attenuation model is used to describe the reduction in signal strength when passing through a cable. It is related to the physical properties, length, material, and aging of the cable. The attenuation model is expressed as a linear relationship. The attenuation model is expressed as: ,in, Is in the distance The signal attenuation at is the basic attenuation, taking into account factors such as cable connection and initial loss, is the attenuation coefficient per unit length, The length of the cable is converted into the aging index of the cable, which includes:

[0086] Establish a relationship model between attenuation and aging: Based on historical monitoring data, establish a relationship model between cable attenuation and aging degree;

[0087] Attenuation parameter calibration: When the cable is first put into use, measure and record the basic attenuation and the attenuation coefficient per unit length , serving as a benchmark for future attenuation measurements;

[0088] Periodic attenuation measurement: Regularly measure the actual attenuation of the cable using a test signal , and record the distance ;

[0089] Aging degree calculation: Based on the relationship model, the actual measured attenuation Converting to an aging index for the cable involves calculating the difference between the actual attenuation and the base attenuation, and correlating the difference with the degree of aging.

[0090] In order to convert the attenuation degree into the aging index of the cable, a linear model is set. There is a direct linear relationship between the aging index of the cable and the attenuation degree of the signal passing through the cable. The model is expressed as: ,in, It is the aging index of the cable, which can be a percentage from 0 to 100, where 0% represents no aging and 100% represents complete aging; Is the intercept of the linear model, representing the baseline aging index of the cable when there is no aging. Ideally, if the cable has no aging at all, Should be close to is the slope of the linear model, indicating the degree of attenuation and aging indicators The strength of the relationship between them. The value of determines the impact of the attenuation change on the aging index. It is the degree of attenuation, which can be expressed by the difference between the attenuation calculated by the above attenuation model and the initial attenuation (i.e. the attenuation when the cable is new).

[0091] Here is an example:

[0092] Through experiments and historical data analysis, a specific cable aging model was obtained, whose parameters are (Even new cables may have of “baseline aging”), (i.e., for every 1dB increase in attenuation, the aging index increases .

[0093] If the initial attenuation of a cable segment is , the measured attenuation after a period of time is , then the attenuation increases , according to the model, the aging index of this cable can be calculated as: , which means that the cable has aged more than the initial state. .

[0094] S3 specifically includes:

[0095] S31, a short pulse or step signal is sent to the cable through the TDR device, and the signal transmission time is recorded at the device end;

[0096] S32, Reflected Signal Reception and Recording: Use the same TDR equipment to capture reflected signals caused by discontinuities in the cable (such as damaged or aged areas) and record the arrival time and strength of the reflected signals;

[0097] S33, calculate the time delay between the reflected signal and the original transmitted signal, the time delay Obtained directly from TDR equipment;

[0098] S34, Determine the location of the break or discontinuity : ,in, is the distance from the start of the cable to the discontinuity, is the propagation speed of the signal in the cable (depending on the dielectric constant of the cable), is the time delay from transmitting the signal to receiving the reflected signal;

[0099] S35, by analyzing the intensity of the reflected signal, a preliminary judgment is made on the nature of the discontinuity. A high reflection intensity indicates severe damage or discontinuity.

[0100] S4 specifically includes:

[0101] S41, determine the test frequency, starting from low frequency and gradually increasing to high frequency to evaluate the frequency response characteristics of the cable;

[0102] S42, for each selected test frequency, using a signal generator to generate a test signal of the corresponding frequency, and transmitting the test signal through the cable, and capturing the transmitted signal at the other end of the cable or at a detection point set in a segment;

[0103] S43, Phase Analysis, for each test frequency, measures and records the phase change of the signal transmitted through the cable. The phase change is measured by comparing the phase difference between the transmitted signal and the received signal.

[0104] S44, drawing a frequency response curve of the cable based on the phase analysis result and the attenuation analysis of the attenuation model, including a curve showing attenuation variation with frequency and a curve showing phase variation with frequency;

[0105] S45 identifies the impact of cable aging or minor damage on signal transmission by analyzing frequency response curve characteristics, such as abnormal increases in attenuation or phase mutations within a specific frequency range. Different types of faults (such as insulation aging, minor cracks, or connector damage) exhibit different characteristics on the frequency response curve.

[0106] The following are common fault types and their characteristics on the frequency response curve:

[0107] 1. Cable insulation aging.

[0108] Characteristics: Aging of insulation materials will cause the overall attenuation to gradually increase with increasing frequency. In the high-frequency area, this increase is more significant.

[0109] Frequency response curve performance: In the low-frequency area, the attenuation remains relatively stable, but as the frequency increases, the slope of the attenuation curve gradually increases, showing that the curve bends upward.

[0110] 2. Minor cracks or local damage.

[0111] Characteristics: Small cracks or localized damage can cause significant attenuation peaks at specific frequencies because these damages cause signal reflections and scattering.

[0112] Frequency response curve performance: At a specific frequency (or nearby frequencies), the attenuation curve will have a sharp peak, reflecting the local reflection of the signal at the damage site.

[0113] 3. The connector or connector is damaged.

[0114] Characteristics: Damage to a connector or connector will result in an overall increase in attenuation at all frequencies, especially if the connector loses good electrical contact.

[0115] Frequency response curve performance: In the entire frequency range, the attenuation curve moves upward as a whole, showing a higher attenuation than normal.

[0116] 4. The shielding layer is damaged.

[0117] Characteristics: If the shielding layer of the cable is damaged, especially at high frequencies, it will lead to increased electromagnetic interference and thus increase signal attenuation.

[0118] Frequency response curve performance: In the high-frequency region, the attenuation curve rises faster than that of an undamaged cable, and an abnormal attenuation increase occurs in the high-frequency range.

[0119] 5. Moisture intrusion.

[0120] Characteristics: Moisture intrusion causes increased attenuation of the cable within a specific frequency range, especially at frequencies where moisture absorption is stronger.

[0121] Frequency response curve performance: In certain frequency regions, such as the microwave band, the attenuation curve may show a more obvious upward trend than normal.

[0122] By testing the cable with multiple frequency signals, the following Figure 3 、 Figure 4 curve;

[0123] exist Figure 3In the figure, the Total Attenuation curve represents the total attenuation of the cable under the influence of aging and specific damage. It can be seen that at about 500Hz, there is an obvious attenuation peak due to cable damage, which is consistent with the characteristics of small cracks or local damage mentioned in the above analysis. At the same time, the trend of overall attenuation increasing with increasing frequency indicates the aging of cable insulation. The Normal Attenuation curve (dashed line) represents the normal attenuation of the cable when there is no aging or damage.

[0124] exist Figure 4 In the figure, the phase variation curve shows periodic fluctuations as the frequency changes, which is a manifestation of phase variation.

[0125] The phase change is calculated as: ;in, is the frequency Phase change (degrees) under and are the phases of the signals at the receiving and transmitting ends respectively.

[0126] S5 specifically includes:

[0127] S51, Data Preprocessing: Preprocess attenuation, TDR results, frequency response curve characteristics, and phase change, including denoising, normalization, and feature extraction to ensure that the data is suitable for model input;

[0128] S52, define the task: define cable aging condition analysis and damage location determination as two related learning tasks. The aging condition is considered a classification problem (for example, classifying it into several levels based on the degree of aging), and the damage location is considered a localization problem (for example, predicting the specific location of the damage or the range where the damage exists).

[0129] S53, Design of Multi-task Learning Model: A multi-task learning model processes multiple types of input data and can output the results of multiple tasks. It uses a shared hidden layer to extract common features of all tasks and then provides a dedicated output layer for each task. This is achieved through a multi-task learning model based on deep learning.

[0130] S54, training model: Use a training dataset containing attenuation, TDR results, frequency response curve characteristics, and phase change to train the model. During the training process, it is necessary to simultaneously optimize the loss functions of the two tasks. This is achieved through a weighted loss function, where the weights reflect the importance of different tasks.

[0131] S55, Model Evaluation and Adjustment: Use the validation set to evaluate the model's performance on aging condition analysis and damage location determination tasks, and adjust the model structure, parameters, and task weights based on the evaluation results;

[0132] S56, apply the model for analysis: Apply the trained deep learning-based multi-task learning model to new test data, comprehensively analyze the aging condition of the cable and the specific damage location, and output an assessment result including the aging condition and prediction of the damage location.

[0133] The multi-task learning model based on deep learning specifically includes:

[0134] Shared layer: consists of several convolutional layers or recurrent layers, used to extract common features of all input data;

[0135] Task-specific layers: Each task (aging analysis and damage location determination) has its own set of fully connected layers to learn the task representation from the features extracted by the shared layers;

[0136] The input data includes attenuation data, TDR results, frequency response characteristics and phase difference information;

[0137] Shared layer feature extraction: As input data, the feature extraction of the shared layer is expressed as ,in is the function of the shared layer, are the parameters of the shared layer;

[0138] Task-specific layer output: For the aging analysis task, let its output be ,in, is the function of the aging status analysis task layer, is the parameter of the task layer. For the damage location determination task, let its output be ,in, is a function of the damage location determination task layer, is the parameter of the task layer;

[0139] Loss function: The model loss function is the weighted sum of the two task loss functions, expressed as: ,in and are the loss functions for aging condition analysis and damage location determination tasks, and are the true labels of the corresponding tasks, and is the weight used to balance the importance of different tasks.

[0140] During the training phase, by minimizing the loss function To update the shared layer parameters and each task-specific layer parameter and ,In the application phase, given new input data, the model can simultaneously provide ,assessment of aging conditions and prediction of damage locations.

[0141] like Figure 2 As shown, a system for monitoring and maintaining the operation status of a power cable is used to implement the above-mentioned method for monitoring and maintaining the operation status of a power cable, and includes the following modules:

[0142] Signal transmitting module: sends a test signal of predetermined frequency and strength to the target cable;

[0143] Signal capture and attenuation analysis module: Set at the other end or segment of the cable, it is used to capture the test signal transmitted through the cable and analyze the attenuation between the original signal and the received signal, and predict the aging degree of the cable based on the attenuation index;

[0144] Time Domain Reflectometry (TDR) module: uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to preliminarily determine the location of potential damage or discontinuity in the cable;

[0145] Frequency response analysis module: performs multi-frequency signal testing on the cable and analyzes the transmission characteristics of the cable for signals of different frequencies. It identifies the impact of cable aging or minor damage on signal transmission and determines the type of fault.

[0146] Comprehensive analysis and diagnosis module: A multi-task learning model is used to perform comprehensive analysis to determine the aging condition of the cable and the specific damage location.

[0147] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Within the scope of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0148] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for monitoring and maintaining the operating status of a power cable, characterized in that: The following steps are involved: S1: Signal transmission, sending a test signal of predetermined frequency and intensity to the target cable; S2: Signal capture and attenuation analysis: Capture the test signal transmitted through the cable at the other end of the cable or at a test point set up in a segment. Compare the attenuation between the original signal and the received signal, and predict the degree of cable aging based on the attenuation index. S3: Time Domain Reflectometry (TDR) technology uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to determine the preliminary location of potential damage or discontinuity in the cable; S4: Frequency response analysis: Perform multi-frequency signal testing on the cable to analyze the cable's transmission characteristics for each frequency signal. This helps identify the impact of cable aging or minor damage on the transmission of different frequency signals, thereby determining the fault type. S5: Based on the analysis results of S2, S3, and S4, a multi-task learning model is used for comprehensive analysis to determine the aging status of the cable and the specific damage location; The S2 specifically includes: S21, set up detection points at the other end of the cable or in sections according to the cable length and structure, and install an oscilloscope at each detection point; S22, while transmitting the test signal, ensuring that the oscilloscope at the detection point synchronously records the signal through the synchronous output port of the signal generator; S23, recording the received signal with an oscilloscope, comparing it with the original signal, analyzing the attenuation between the two, and converting the attenuation into a cable aging index using an attenuation model to determine the cable aging degree; The attenuation model is expressed as: ,in, Is in the distance The signal attenuation at is the basic attenuation, taking into account factors such as cable connection and initial loss, is the attenuation coefficient per unit length, The length of the cable is converted into the aging index of the cable, which includes: Establish a relationship model between attenuation and aging: Based on historical monitoring data, establish a relationship model between cable attenuation and aging degree; Attenuation parameter calibration: When the cable is first put into use, measure and record the basic attenuation and the attenuation coefficient per unit length , serving as a benchmark for future attenuation measurements; Periodic attenuation measurement: Regularly measure the actual attenuation of the cable using a test signal , and record the distance ; Aging degree calculation: Based on the relationship model, the actual measured attenuation Converting to an aging index for the cable involves calculating the difference between the actual attenuation and the base attenuation, and correlating the difference with the degree of aging.

2. A method for monitoring and maintaining the operation status of a power cable according to claim 1, characterized in that: Said S1 specifically includes: S11, configuring a signal generator, wherein the signal generator includes an adjustable frequency and an adjustable output intensity, and the frequency range and intensity level of the test signal are preset according to the characteristics of the cable and the detection requirements; S12, connect the signal generator to the target cable through a cable test head or adapter that matches the cable; S13, generating a pulse waveform using a modulation function of a signal generator; S13, test sequence initialization: start the test sequence through the signal generator, send test signals of predetermined frequency and intensity in sequence, and monitor the stability and continuity of the signal to ensure the signal quality of the entire test process.

3. A method for monitoring and maintaining the operation status of a power cable according to claim 1, characterized in that: The S3 specifically includes: S31, a short pulse or step signal is sent to the cable through the TDR device, and the signal transmission time is recorded at the device end; S32, Reflected signal reception and recording: Use the same TDR device to capture the reflected signal caused by the discontinuity in the cable and record the arrival time and intensity of the reflected signal; S33, calculate the time delay between the reflected signal and the original transmitted signal, the time delay Obtained directly from TDR equipment; S34, Determine the location of the break or discontinuity : ,in, is the distance from the start of the cable to the discontinuity, is the propagation speed of the signal in the cable, is the time delay from transmitting the signal to receiving the reflected signal; S35, by analyzing the intensity of the reflected signal, a preliminary judgment is made on the nature of the discontinuity. A high reflection intensity indicates severe damage or discontinuity.

4. A method for monitoring and maintaining the operation status of a power cable according to claim 1, characterized in that: The S4 specifically includes: S41, determine the test frequency, starting from low frequency and gradually increasing to high frequency to evaluate the frequency response characteristics of the cable; S42, for each selected test frequency, using a signal generator to generate a test signal of the corresponding frequency, and transmitting the test signal through the cable, and capturing the transmitted signal at the other end of the cable or at a detection point set in a segment; S43, Phase Analysis, for each test frequency, measures and records the phase change of the signal transmitted through the cable. The phase change is measured by comparing the phase difference between the transmitted signal and the received signal. S44, drawing a frequency response curve of the cable based on the phase analysis result and the attenuation analysis of the attenuation model, including a curve showing attenuation variation with frequency and a curve showing phase variation with frequency; S45, by analyzing the characteristics of the frequency response curve, can identify the impact of cable aging or minor damage on signal transmission. Different types of faults show different characteristics on the frequency response curve.

5. A method for monitoring and operating the power cable according to claim 4, characterized in that: The phase change is calculated as: ;in, is the frequency The phase change under and are the phases of the signals at the receiving and transmitting ends respectively.

6. A method for monitoring and maintaining the operation status of a power cable according to claim 5, characterized in that: The S5 specifically includes: S51, Data Preprocessing: Preprocess attenuation, TDR results, frequency response curve characteristics, and phase change, including denoising, normalization, and feature extraction to ensure that the data is suitable for model input; S52, define the task: define cable aging condition analysis and damage location determination as two related learning tasks, where the aging condition is considered a classification problem and the damage location determination is considered a positioning problem; S53, Design of Multi-task Learning Model: A multi-task learning model processes multiple types of input data and can output the results of multiple tasks. It uses a shared hidden layer to extract common features of all tasks and then provides a dedicated output layer for each task. This is achieved through a multi-task learning model based on deep learning. S54, training model: Use a training dataset containing attenuation, TDR results, frequency response curve characteristics, and phase change to train the model. During the training process, it is necessary to simultaneously optimize the loss functions of the two tasks. This is achieved through a weighted loss function, where the weights reflect the importance of different tasks. S55, Model Evaluation and Adjustment: Use the validation set to evaluate the model's performance on aging condition analysis and damage location determination tasks, and adjust the model structure, parameters, and task weights based on the evaluation results; S56, apply the model for analysis: Apply the trained deep learning-based multi-task learning model to new test data, comprehensively analyze the aging condition of the cable and the specific damage location, and output an assessment result including the aging condition and prediction of the damage location.

7. A method for monitoring and maintaining the operation status of a power cable according to claim 6, characterized in that: The multi-task learning model based on deep learning specifically includes: Shared layer: consists of several convolutional layers or recurrent layers, used to extract common features of all input data; Task-specific layers: Each task has its own set of fully connected layers to learn the representation of the task from the features extracted by the shared layers; The input data includes attenuation data, TDR results, frequency response characteristics and phase difference information; Shared layer feature extraction: As input data, the feature extraction of the shared layer is expressed as ,in is the function of the shared layer, are the parameters of the shared layer; Task-specific layer output: For the aging analysis task, let its output be ,in, is the function of the aging status analysis task layer, is the parameter of the task layer. For the damage location determination task, let its output be ,in, is a function of the damage location determination task layer, is the parameter of the task layer; Loss function: The model loss function is the weighted sum of the two task loss functions, expressed as: ,in and are the loss functions for aging condition analysis and damage location determination tasks, and are the true labels of the corresponding tasks, and It is the weight used to balance the importance of different tasks.

8. A system for monitoring and maintaining the operation status of a power cable, used to implement a method for monitoring and maintaining the operation status of a power cable according to any one of claims 1 to 7, characterized in that: Includes the following modules: Signal transmitting module: sends a test signal of predetermined frequency and strength to the target cable; Signal capture and attenuation analysis module: Set at the other end or segment of the cable, it is used to capture the test signal transmitted through the cable and analyze the attenuation between the original signal and the received signal, and predict the aging degree of the cable based on the attenuation index; Time Domain Reflectometry (TDR) module: uses TDR technology to analyze the time delay and strength of the reflected signal in the cable to preliminarily determine the location of potential damage or discontinuity in the cable; Frequency response analysis module: performs multi-frequency signal testing on the cable and analyzes the transmission characteristics of the cable for signals of different frequencies. It identifies the impact of cable aging or minor damage on signal transmission and determines the type of fault. Comprehensive analysis and diagnosis module: A multi-task learning model is used to perform comprehensive analysis to determine the aging condition of the cable and the specific damage location.

Images