Cable aging and humidity monitoring method, system and device and storage medium

Abstract

The invention discloses a cable aging and humidity monitoring method, system and device and a storage medium, and the method comprises the steps: obtaining the response data of a cable under different humidity and aging conditions based on the polarization current measurement data of the cable, constructing a dielectric response function model, and obtaining the aging and humidity characteristic parameters of the cable based on the dielectric response function model; cable aging feature extraction is performed on the polarization current measurement data of the cable, and the cable aging state is classified and identified by taking the strongly correlated cable aging features as the input of the state identification model; establishing a mathematical relationship between the humidity characteristic parameters and the humidity content based on the humidity characteristic parameters, and predicting the humidity state of the cable; and evaluating the health condition of the cable based on the cable humidity state and the cable aging state. Through polarization current measurement and simplified signal processing, the equipment cost and complexity are reduced, and the feasibility of practical application is improved.

Description

Technical Field

[0001] This invention relates to the technical field of cable condition monitoring, and in particular to a method, system, device, and storage medium for monitoring cable aging and humidity. Background Technology

[0002] With the continuous development of power distribution systems and the increase in electricity demand, power cables, as key power transmission components, are widely used in various medium-voltage and low-voltage power networks. In particular, cross-linked polyethylene (XLPE) cables, due to their excellent electrical, thermal, and mechanical properties, have become the main choice for medium-voltage (6-30kV) power systems. However, as the service life of cables increases, the cable insulation material gradually ages. Especially due to environmental factors (such as temperature and humidity) and current load, the insulation layer of XLPE cables exhibits significant physical and chemical degradation, leading to a decline in cable performance and even cable failure. Therefore, how to effectively and accurately assess the aging state and humidity content of cables has become an important research topic in the power industry.

[0003] Traditional XLPE cable aging assessment methods primarily rely on physical or chemical testing, requiring cable cutting or disassembly. This makes real-time on-site testing impossible and is complex, impacting the cable's continued usability. Furthermore, some testing techniques introduce environmental interference during testing, leading to inaccurate or unstable results. Currently, many high-precision aging assessment technologies rely on expensive equipment. In practical applications, their large size and complex operational requirements increase operational difficulty and maintenance costs, limiting their large-scale promotion and application. Humidity accumulation is a significant factor contributing to XLPE cable aging. Moisture penetration into the cable insulation layer triggers water treeing, accelerating cable degradation. While traditional humidity assessment methods can detect water trees, their quantitative assessment of humidity content is inaccurate. Moreover, the complex and irregular morphology of water trees makes accurate quantification using traditional detection methods difficult. In actual power systems, cable aging and humidity intrusion often occur simultaneously, and their interaction accelerates cable insulation degradation. However, existing methods often rely on single-factor analysis, lacking modeling of the synergistic effects of both factors, resulting in incomplete assessment results and a high risk of misjudgment or omission. Most existing cable aging and humidity assessment technologies require long testing cycles and are typically limited to specific conditions, failing to achieve real-time online monitoring. Current aging assessment methods rely on complex data analysis algorithms, such as principal component analysis and support vector machines. These methods demand high data quality and processing accuracy, often requiring significant computing power and skilled technical personnel for data analysis. While machine learning and artificial intelligence technologies have provided some assistance in cable aging assessment, the data processing workflows of traditional methods remain cumbersome and lack support for real-time processing of large-scale data, posing significant challenges in practical applications. Summary of the Invention

[0004] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a method, system, device, and storage medium for monitoring cable aging and humidity, addressing the problems of high cost, equipment complexity, inability to simultaneously assess the effects of thermal aging and humidity, and low data analysis efficiency in the prior art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a method for monitoring cable aging and humidity, comprising:

[0007] Based on the polarization current measurement data of the cable, the response data of the cable under different humidity and aging conditions are obtained and a dielectric response function model is constructed. Based on the dielectric response function model, the characteristic parameters of cable aging and humidity are obtained.

[0008] Cable aging features are extracted from the polarization current measurement data of the cable, and the strongly correlated cable aging features are used as input to the state recognition model to classify and identify the cable aging state.

[0009] Based on the humidity characteristic parameters, a mathematical relationship between the humidity characteristic parameters and the humidity content is established to predict the cable humidity status;

[0010] The health status of the cable is assessed based on its humidity and aging conditions.

[0011] As a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the following steps are included: acquiring the cable's response data under different humidity and aging conditions and constructing a dielectric response function model:

[0012] A constant voltage is applied to the cable, and the current response in the time domain is measured, the current response including the conduction current and the polarization current;

[0013] Based on the polarization current response slopes on the first and second time scales, a dielectric response function model is constructed.

[0014] Based on the polarization current change states at the first and second time scales, characteristic parameters of cable aging and humidity are obtained.

[0015] The characteristic parameters include humidity-sensitive parameters, the slope of the polarization current response at the first time scale, the slope of the polarization current response at the second time scale, and the time of transition between the polarization current change state at the first time scale and the polarization current change state at the second time scale.

[0016] As a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the extraction of cable aging characteristics from the polarization current measurement data of the cable includes:

[0017] Based on the measured polarization current data, the dielectric response function is calculated to obtain the dynamic process of dipole and interface polarization inside the cable insulation material. For each time point, the response characteristics at different time scales are obtained.

[0018] When the dipole or interface polarization within the cable insulation material reaches its maximum state at any given time point, the polarization current response intensity at that time point and the current time point are used as characteristic parameters for evaluating cable aging.

[0019] In a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the strongly correlated cable aging characteristics are used as input to the state recognition model to classify and identify the cable aging state, including:

[0020] The importance of each feature is assessed by comparing the variance between different groups of cable aging features with the variance within the same group. Cable aging features with the minimum calculated value are removed, and cable aging status is classified and identified based on the remaining feature parameters.

[0021] As a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the classification and identification of cable aging status based on other feature parameters includes: classifying cables under different aging conditions to maximize the interval between different feature categories, optimizing the status identification model parameters, constructing multiple independent decision trees, and using a voting mechanism to determine the final classification result.

[0022] The state recognition model is represented as:

[0023]

[0024] y i (w T x i +b)≥1,i=1,2,...N

[0025]

[0026] y = mode(f1(x),f2(x),...,f t (x))

[0027] Where w is the normal vector of the hyperplane, ||w|| is the norm of the normal vector, and x i It is the input feature, y i Here are the corresponding category labels, b is the bias term, and L(θ) represents the overall loss of the cable condition assessment. It is a loss function, representing the true state y of the cable. i and predicted state The difference between them; Ω(f) k ) is the regularization term, f t (x) is the prediction result of the t-th tree, and mode represents the majority vote.

[0028] The beneficial effects of this preferred technical solution are that, from feature selection to the use of machine learning models to classify the aging state of cables, the degree of aging can be accurately assessed based on the specific electrical characteristics of the cables, thereby enabling the implementation of corresponding maintenance measures and improving the reliability and safety of the power system.

[0029] As a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the following steps are taken: Based on the humidity characteristic parameters, establishing a mathematical relationship between the humidity characteristic parameters and the humidity content to predict the cable humidity status includes: the mathematical relationship between the humidity characteristic parameters and the humidity content is expressed as follows:

[0030]

[0031] Where, m c y0 represents the cable humidity content, A0 is the humidity-sensitive parameter, and y0, a, b, and c are regression coefficients.

[0032] As a preferred embodiment of the cable aging and humidity monitoring method of the present invention, the assessment of the cable's health status based on the cable's humidity state and aging state includes:

[0033] If either the cable humidity or cable aging condition exceeds the preset safety threshold, an alarm mechanism will be triggered to prompt the operator to perform inspection or maintenance.

[0034] If either the cable humidity or cable aging status indicator does not exceed the preset safety threshold, the cable is in a healthy state.

[0035] The beneficial effects of this preferred technical solution are that real-time online monitoring and intelligent early warning of cross-linked polyethylene cables greatly improve the reliability of power system operation. By combining polarization current measurement and machine learning algorithms, the health status of the cable can be accurately assessed, and early warning of faults can be achieved.

[0036] Secondly, the present invention provides a cable aging and humidity monitoring system, comprising:

[0037] The model building module is used to acquire the response data of the cable under different humidity and aging conditions based on the polarization current measurement data of the cable and to construct a dielectric response function model. Based on the dielectric response function model, the characteristic parameters of cable aging and humidity are obtained.

[0038] The cable aging condition identification module is used to extract cable aging features from the polarization current measurement data of the cable, and to classify and identify the cable aging condition by using strongly correlated cable aging features as input to the condition identification model.

[0039] The cable humidity condition prediction module is used to establish a mathematical relationship between the humidity characteristic parameters and the humidity content based on the humidity characteristic parameters, and to predict the cable humidity condition.

[0040] The evaluation module is used to assess the health status of the cable based on its humidity and aging conditions.

[0041] Thirdly, the present invention provides an electronic device, comprising:

[0042] Memory and processor;

[0043] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the cable aging and humidity monitoring method.

[0044] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the cable aging and humidity monitoring method.

[0045] Compared with existing technologies, the present invention offers the following advantages: It obtains cable response data under different humidity and aging conditions through polarization current measurement and extracts humidity-sensitive parameters using a dielectric response model. By establishing a mathematical relationship between humidity-sensitive parameters and humidity content, the humidity status of the cable is predicted. Based on this, cable samples are classified according to their aging state. The most relevant features are selected to train the classifier, ensuring the accuracy and reliability of the classification results. The entire process enables real-time monitoring of the cable, timely detection of aging or humidity anomalies, and provides effective early warning and maintenance decision-making support. It boasts high monitoring efficiency, low cost, and significant practical value and application prospects. Attached Figure Description

[0046] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0047] Figure 1 This is a schematic flowchart illustrating a cable aging and humidity monitoring method, system, device, and storage medium according to an embodiment of the present invention.

[0048] Figure 2 This invention relates to a cable aging and humidity monitoring method, system, device, and storage medium, as described in one embodiment of the present invention.

[0049] Figure 3 This is a graph showing the variation of characteristic parameters of a cable aging and humidity monitoring method, system, device, and storage medium with humidity and aging status, according to an embodiment of the present invention. Detailed Implementation

[0050] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0051] Example 1, referring to Figures 1-3 As one embodiment of the present invention, this embodiment provides a method for monitoring cable aging and humidity, including:

[0052] S100: Based on the polarization current measurement data of the cable, obtain the response data of the cable under different humidity and aging conditions and construct a dielectric response function model. Based on the dielectric response function model, obtain the characteristic parameters of cable aging and humidity.

[0053] S200: Extract cable aging features from the polarization current measurement data of the cable, use the strongly correlated cable aging features as input to the state recognition model, and classify and identify the cable aging state.

[0054] S300: Based on humidity characteristic parameters, establish the mathematical relationship between humidity characteristic parameters and humidity content to predict the humidity status of cables;

[0055] S400: Assess the health of a cable based on its humidity and aging conditions.

[0056] It should be noted that existing aging assessment methods cannot effectively distinguish the effects of thermal aging and humidity accumulation, and humidity detection often presents challenges in quantitative analysis. Due to the complex and irregular morphology of water trees, traditional methods struggle to accurately quantify humidity content; traditional detection methods are mostly offline, requiring long testing cycles and failing to provide real-time monitoring of cable conditions. This results in the inability to provide timely warnings before faults occur and to effectively predict cable health; existing equipment, such as high-voltage dielectric spectroscopy and PDC technology, typically requires expensive instruments and is complex to operate, making it unsuitable for large-scale deployment; traditional methods rely on complex data analysis and high computing resources, and have poor real-time processing capabilities for large-scale data, impacting application efficiency. Existing XLPE cable aging and humidity assessment methods have several limitations, including high cost, equipment complexity, inability to simultaneously assess the effects of thermal aging and humidity, and complex data analysis. Therefore, there is an urgent need for a new, more efficient, and more economical method that can monitor and accurately assess cable aging and humidity in real time. This invention reduces equipment costs and complexity and improves the feasibility of practical applications by measuring polarization current and simplifying signal processing. By combining machine learning algorithms with simplified data processing procedures, it improves processing efficiency and adapts to the practical application needs of large-scale power systems.

[0057] In this embodiment of the application, step S100, which involves acquiring the cable's response data under different humidity and aging conditions and constructing a dielectric response function model, includes:

[0058] A constant voltage is applied to the cable, and the current response in the time domain is measured. The current response includes the conduction current and the polarization current.

[0059] Based on the polarization current response slopes on the first and second time scales, a dielectric response function model is constructed.

[0060] Based on the polarization current change states at the first and second time scales, characteristic parameters of cable aging and humidity are obtained.

[0061] The characteristic parameters include humidity-sensitive parameters, the slope of the polarization current response at the first time scale, the slope of the polarization current response at the second time scale, and the time of transition between the polarization current change state at the first time scale and the polarization current change state at the second time scale.

[0062] It should be noted that the dielectric response function model determines the shape and characteristics of the polarization current spectrum. Different materials exhibit different degrees of degradation. This application uses the dielectric response function model to describe the effects of factors such as thermal aging or moisture penetration on insulating materials. Specifically, the dielectric response is modeled as the sum of two power-law characteristics with negative slopes. The dielectric response function model can be expressed as the following formula:

[0063]

[0064] Among them, the exponents are 0 < m < 1 and 1 < n < 2. When the response of the polarization current is plotted on a logarithmic scale, two straight lines will be shown. One has a slope of "m" in the short-time range, and the other has a slope of "n" in the long-time range. These two regions indicate that there are two independent and continuous physical processes in the time-domain measurement of the dielectric material; the transition between these two states occurs at time t = τ, which is the reciprocal of the loss-peak frequency.

[0065] Furthermore, when analyzing the measured polarization current, first determine the dielectric response function h(t). Once h(t) is determined, the formula of h(t) can be used for modeling, and the overall fitting equation of the polarization current I pol (t) is as follows:

[0066]

[0067] In the dielectric response model, characteristic parameters such as A0, m, n, and τ can be extracted as useful indicators for insulation aging and humidity.

[0068] In an optional embodiment, by applying a constant 200 VDC voltage to an XLPE cable sample, record the change of current with time to obtain the polarization current measurement data of the cable; during the measurement process, the current is divided into two parts: one is the conduction current (caused by the DC resistance of the material), and the other is the polarization current (caused by dielectric polarization and interfacial polarization). The expression of the current is as follows:

[0069] I pol (t) = I d (t) + I con

[0070] Among them, I d (t) is the polarization current, and I con is the conduction current.

[0071] The polarization current can be expressed by the following formula:

[0072] I d (t) = C0U0h(t)

[0073] Among them, C0 is the geometric capacitance of the test object, h(t) is the dielectric response function, and U0 is the amplitude of the charging voltage.

[0074] In the embodiment of the present application, the extraction of cable aging characteristics from the polarization current measurement data of the cable in step S200 includes:

[0075] Based on the measured polarization current data, the dielectric response function is calculated to obtain the dynamic process of dipole and interface polarization inside the cable insulation material. For each time point, the response characteristics at different time scales are obtained.

[0076] When the dipole or interface polarization within the cable insulation material reaches its maximum state at any given time point, the polarization current response intensity at that time point and the current time point are used as characteristic parameters for evaluating cable aging.

[0077] In an alternative embodiment, the dipole relaxation current is measured during the period when the insulation layer is subjected to stress for a considerable time (typically 1 hour); in the representation of the current-time plot, the current recorded by isothermal relaxation current analysis (IRC) is plotted relative to ln(t) after multiplying by the recorded time. Typically, the parameters extracted from the IRC plot can serve as useful indicators of insulation aging.

[0078] Replace I with the calculated dielectric response function data h(t). d (t) is used for IRC analysis, saving measurement time. The measured relaxation current depends on the size C0 and the applied DC voltage amplitude U0, eliminating the influence of voltage and insulation geometry, making h(t) a true representation of the insulation characteristics. Considering these modifications to traditional IRC analysis, the characteristic obtained by plotting h(t)×t against ln(t) is called the Normalized Dipole Isothermal Relaxation Current Plot (NDIRC). From the NDIRC plot, the NDIRC value can be obtained. peak and ln(t) peak These two characteristic parameters are used as useful indicators of insulation aging.

[0079] In this embodiment of the application, step S200, which uses strongly correlated cable aging characteristics as input to the state recognition model, classifies and identifies the cable aging state, including:

[0080] The importance of each feature is assessed by comparing the variance between different groups of cable aging features with the variance within the same group. Cable aging features with the minimum calculated value are removed, and cable aging status is classified and identified based on the remaining feature parameters.

[0081] It should be noted that features related to cable aging and humidity were extracted from polarization current measurement data. Features obtained through dielectric response modeling and NDIRC analysis include A0, m, n, τ, and NDIRC. peak and ln(t) peakSince the impact of any single parameter on cable aging and humidity cannot be determined, this embodiment uses the Analysis of Variance (ANOVA) algorithm to select the features most relevant to cable aging and humidity. The ANOVA algorithm evaluates the importance of each feature by calculating the F-score, selecting the feature with the highest F-score for subsequent machine learning classification. The formula for calculating the F-score is:

[0082]

[0083] Among them, F i is the F-value of feature i; between-group variance i represents the variance between different groups; within-group variance i represents the variance within the same group.

[0084] A high F-value indicates that this characteristic has a significant impact on cable aging and humidity. The F-values ​​of the characteristic parameters are shown in Table 1. The five characteristic parameters with the highest F-values ​​were selected for further analysis.

[0085] Table 1. Characteristic parameter F-values

[0086]

[0087] In this embodiment, step S200, classifying and identifying the cable aging state based on the remaining feature parameters, includes: classifying cables under different aging conditions to maximize the interval between different feature categories, optimizing the state identification model parameters, constructing multiple independent decision trees, and using a voting mechanism to determine the final classification result; the state identification model is represented as:

[0088]

[0089] y i (w T x i +b)≥1,i=1,2,...N

[0090]

[0091] y = mode(f1(x),f2(x),...,f t (x))

[0092] Where w is the normal vector of the hyperplane, ||w|| is the norm of the normal vector, and x i It is the input feature, y i Here are the corresponding category labels, b is the bias term, and L(θ) represents the overall loss of the cable condition assessment. It is a loss function, representing the true state y of the cable. i and predicted state The difference between them; Ω(f) k ) is the regularization term, f t(x) is the prediction result of the t-th tree, and mode represents the majority vote.

[0093] In an optional embodiment, the state recognition model includes three classifier models: support vector machine, extreme gradient boosting, and random forest algorithm.

[0094] Support Vector Machine (SVM) is a supervised learning algorithm commonly used in classification and regression analysis. Its goal is to find the optimal hyperplane to maximize the margin between different classes, so that cables under different aging conditions are correctly classified.

[0095] For nonlinear data, SVM uses kernel tricks to map the data to a higher-dimensional space, making the data linearly separable:

[0096] K(x,x f )=φ(x) T φ(x f )

[0097] Where K(x,x) f ) is the kernel function, and φ(x) is the mapping function that maps data to a high-dimensional space.

[0098] Extreme Gradient Boosting (XGBoost) is an ensemble learning technique that combines multiple weak classifiers (typically decision trees) to provide more accurate results. Each new model is an optimization based on the previous one. XGBoost updates the model parameters using gradient boosting. XGBoost iteratively optimizes the objective function by weighting the outputs of each tree.

[0099]

[0100] Where η is the learning rate, f t (x i ) is the prediction result for the t-th tree.

[0101] Random forest is an ensemble learning method that classifies data by building multiple decision trees and voting on the results. Each decision tree is trained independently of the others, and each tree is trained using only a random subset of the data, thus reducing overfitting. Each tree is trained using a random subset of features on the training data, and the tree generation process is recursive until a stopping condition is met (such as the tree depth reaching a maximum or the number of samples falling below a certain threshold). Based on different aging conditions and humidity values, all cable samples can be divided into five aging states, as shown in Table 2:

[0102] Table 2 Classification of Aging Status of Cable Samples

[0103]

[0104] Among them, the cable sample is A x M y The naming convention is as follows: "x" refers to the aging time (days), and "y" refers to the number of artificial holes.

[0105] In this embodiment of the application, step S300, which establishes a mathematical relationship between humidity characteristic parameters and humidity content based on humidity characteristic parameters to predict the cable humidity status, includes: the mathematical relationship between humidity characteristic parameters and humidity content is expressed as follows:

[0106]

[0107] Where, m c The denoted 'A0' represents the humidity content of the cable, 'Y0', 'A', 'B', and 'C' are regression coefficients that depend on the aging level of the cable.

[0108] In an optional embodiment, humidity-sensitive parameters are extracted from the cable's polarization current data using a quantitative humidity content analysis method, and the cable's humidity content is predicted by establishing a mathematical relationship between these parameters and humidity content. Since the amplitude of the polarization current is very low, it is susceptible to external noise; therefore, the minute changes in characteristic parameters with humidity and aging state can be obtained by performing six polarization measurements on the cable sample. Specifically, A0 varies with aging degree and humidity as follows: Figure 3 As shown, by Figure 3 It can be seen that, at a fixed moisture content, A0 increases with increasing aging degree. This indicates that under similar aging conditions, the humidity content follows a stable trend; therefore, without considering the effects of aging, A0 can be used to assess the humidity of the cable. A0 maintains a good polynomial fit with the cable humidity values ​​at all aging levels. The relationship between cable humidity content and humidity-sensitive parameters is established using a polynomial regression method. The fitting coefficients for the five aging levels are shown in Table 3, where R... 2 The fitting accuracy of the polynomial relationship for different aging levels is represented by the mean error (MAE) (%), which represents the average error between the actual cable humidity value and the predicted value.

[0109] Table 3 Regression analysis coefficients

[0110]

[0111] The results show that this application can effectively utilize the polarization current measurement data of the cable to assess the humidity status of the cable, thereby providing accurate data support for cable health monitoring.

[0112] In this embodiment of the application, step S400, which assesses the health status of the cable based on its humidity and aging conditions, includes:

[0113] If either the cable humidity or cable aging condition exceeds the preset safety threshold, an alarm mechanism will be triggered to prompt the operator to perform inspection or maintenance.

[0114] If either the cable humidity or cable aging status indicator does not exceed the preset safety threshold, the cable is in a healthy state.

[0115] It should be noted that, in the embodiments of this application, the preset safety threshold can be set to 200-500 PPM.

[0116] In one optional implementation, the real-time online monitoring system can monitor the health status of cables in the power system in real time through polarization current measurement and signal analysis. Polarization current sensors are installed at key locations in the cable system. These sensors collect polarization current data from the cables in real time and transmit the data to a remote monitoring system for analysis. The data is transmitted to the monitoring platform wirelessly or via wired connection for real-time analysis. The monitoring platform can assess the health status of the cables based on the real-time data collected, especially the humidity content and thermal aging condition. The system performs real-time assessments based on set thresholds. Once the humidity or aging condition of the cable exceeds the set safety threshold, the system will automatically issue an alarm to prompt operators to inspect or maintain the cable.

[0117] It should be noted that this application, by combining polarization current measurement, dielectric response modeling, quantitative analysis of humidity content, and machine learning classification techniques, can accurately assess the thermal aging and humidity status of XLPE cables. First, polarization current measurement is used to obtain the cable's response data under different humidity and aging conditions, and a dielectric response model is used to extract humidity-sensitive parameters. Next, by establishing a mathematical relationship between the humidity-sensitive parameters and humidity content, the cable's humidity status is predicted. Based on this, machine learning classification algorithms, such as Support Vector Machine (SVM), XGBoost, and Random Forest (RFA), are used to classify cable samples according to their aging status. The classifier is trained by selecting the most relevant features and combining methods such as cross-validation to ensure the accuracy and reliability of the classification results. The entire process enables real-time monitoring of the cable and timely detection of aging or humidity anomalies, providing effective early warning and maintenance decision-making basis. This patent provides an efficient and low-cost method for monitoring the aging and humidity of XLPE cables, with high practical value and application prospects.

[0118] Example 2: The above example is an illustrative scheme of a cable aging and humidity monitoring method. It should be noted that the technical solution of this cable aging and humidity monitoring system belongs to the same concept as the technical solution of the above-described cable aging and humidity monitoring method. Details not described in detail in this example can be found in the description of the above-described cable aging and humidity monitoring method.

[0119] This embodiment provides a cable aging and humidity monitoring system, comprising:

[0120] The model building module is used to obtain the response data of the cable under different humidity and aging conditions based on the polarization current measurement data of the cable and to build a dielectric response function model. Based on the dielectric response function model, the characteristic parameters of cable aging and humidity are obtained.

[0121] The cable aging condition identification module is used to extract cable aging features from the polarization current measurement data of the cable, and to classify and identify the cable aging condition by using strongly correlated cable aging features as input to the condition identification model.

[0122] The cable humidity condition prediction module is used to establish a mathematical relationship between humidity characteristic parameters and humidity content based on humidity characteristic parameters, and to predict the cable humidity condition.

[0123] The assessment module is used to evaluate the health status of the cable based on its humidity and aging conditions.

[0124] This embodiment also provides an electronic device suitable for cable aging and humidity monitoring methods, including:

[0125] The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the cable aging and humidity monitoring method proposed in the above embodiments.

[0126] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the cable aging and humidity monitoring method proposed in the above embodiments.

[0127] The storage medium proposed in this embodiment and the cable aging and humidity monitoring method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0128] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0129] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for monitoring cable aging and humidity, characterized in that, include: Based on the polarization current measurement data of the cable, the response data of the cable under different humidity and aging conditions are obtained and a dielectric response function model is constructed. Based on the dielectric response function model, the characteristic parameters of cable aging and humidity are obtained. Cable aging features are extracted from the polarization current measurement data of the cable, and the strongly correlated cable aging features are used as input to the state recognition model to classify and identify the cable aging state. Based on the humidity characteristic parameters, a mathematical relationship between the humidity characteristic parameters and the humidity content is established to predict the cable humidity status; The health status of the cable is assessed based on its humidity and aging conditions.

2. The cable aging and humidity monitoring method as described in claim 1, characterized in that, Acquiring cable response data under different humidity and aging conditions and constructing a dielectric response function model includes: A constant voltage is applied to the cable, and the current response in the time domain is measured, the current response including the conduction current and the polarization current; Based on the polarization current response slopes on the first and second time scales, a dielectric response function model is constructed. Based on the polarization current change states at the first and second time scales, characteristic parameters of cable aging and humidity are obtained. The characteristic parameters include humidity-sensitive parameters, the slope of the polarization current response at the first time scale, the slope of the polarization current response at the second time scale, and the time of transition between the polarization current change state at the first time scale and the polarization current change state at the second time scale.

3. The cable aging and humidity monitoring method as described in claim 2, characterized in that, Extracting cable aging characteristics from the polarization current measurement data of the cable includes: Based on the measured polarization current data, the dielectric response function is calculated to obtain the dynamic process of dipole and interface polarization inside the cable insulation material. For each time point, the response characteristics at different time scales are obtained. When the dipole or interface polarization within the cable insulation material reaches its maximum state at any given time point, the polarization current response intensity at that time point and the current time point are used as characteristic parameters for evaluating cable aging.

4. The cable aging and humidity monitoring method as described in claim 3, characterized in that, Using strongly correlated cable aging characteristics as input to the state recognition model, the cable aging state is classified and identified, including: The importance of each feature is assessed based on the ratio of the variance between different groups of cable aging features to the variance within the same group. Cable aging features with the minimum calculated value are eliminated, and cable aging status is classified and identified based on the remaining feature parameters.

5. The cable aging and humidity monitoring method as described in claim 4, characterized in that, The classification and identification of cable aging status based on other feature parameters includes: classifying cables under different aging conditions to maximize the interval between different feature categories, optimizing the status identification model parameters, constructing multiple independent decision trees, and using a voting mechanism to determine the final classification result. The state recognition model is represented as: y i (w T x i +b)≥1,i=1,2,...N y=mode(f1(x),f2(x),...,f t (x)) Where w is the normal vector of the hyperplane, ||w|| is the norm of the normal vector, and x i It is the input feature, y i Here are the corresponding category labels, b is the bias term, and L(θ) represents the overall loss of the cable condition assessment. It is a loss function, representing the true state y of the cable. i and predicted state The difference between them; Ω(f) k ) is the regularization term, f t (x) is the prediction result of the t-th tree, and mode represents the majority vote.

6. The cable aging and humidity monitoring method as described in claim 5, characterized in that, Based on the humidity characteristic parameters, a mathematical relationship between the humidity characteristic parameters and the humidity content is established to predict the cable humidity status, including: the mathematical relationship between the humidity characteristic parameters and the humidity content is expressed as follows: Where, m c y0 represents the cable humidity content, A0 is the humidity-sensitive parameter, and y0, a, b, and c are regression coefficients.

7. The cable aging and humidity monitoring method as described in claim 6, characterized in that, The health status of the cable is assessed based on its humidity and aging conditions, including: If either the cable humidity or cable aging condition exceeds the preset safety threshold, an alarm mechanism will be triggered to prompt the operator to perform inspection or maintenance. If either the cable humidity or cable aging status indicator does not exceed the preset safety threshold, the cable is in a healthy state.

8. A cable aging and humidity monitoring system, applied to the method described in any one of claims 1-7, characterized in that, include: The model building module is used to acquire the response data of the cable under different humidity and aging conditions based on the polarization current measurement data of the cable and to construct a dielectric response function model. Based on the dielectric response function model, the characteristic parameters of cable aging and humidity are obtained. The cable aging condition identification module is used to extract cable aging features from the polarization current measurement data of the cable, and to classify and identify the cable aging condition by using strongly correlated cable aging features as input to the condition identification model. The cable humidity condition prediction module is used to establish a mathematical relationship between the humidity characteristic parameters and the humidity content based on the humidity characteristic parameters, and to predict the cable humidity condition. An evaluation module is used to assess the health status of the cable based on its humidity and aging conditions.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the cable aging and humidity monitoring method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the cable aging and humidity monitoring method according to any one of claims 1 to 7.

Images