Cable aging degree detection method and system

Abstract

The invention provides a cable aging degree detection method, which comprises the following steps: acquiring polarization current of a cable to be detected to fit a polarization current curve of the cable to be detected; determining to-be-extracted characteristic quantities, and screening out key characteristic quantities from the to-be-extracted characteristic quantities based on the polarization current curve of the to-be-detected cable; and importing the selected key characteristic quantity into a trained classifier to obtain the aging degree of the cable to be detected. The invention also provides a cable aging degree detection system. By implementing the method, the accuracy of a cable aging degree detection result can be improved, and the moisture content of the cable can be effectively detected.

Description

Technical Field

[0001] The present invention relates to the technical field of power cable detection, and particularly relates to a method and system for detecting the aging degree of a cable. Background Art

[0002] Existing literature shows that thermal aging has a significant impact on the insulation performance of cables, mainly reflected in aspects such as breakdown strength, mechanical stiffness, and conductivity. Thermal aging not only changes the physical and chemical properties of the cable, but also triggers partial discharge phenomena inside the cable, resulting in the formation of microcavities. In addition, the intrusion of moisture, especially in wet soil, may exacerbate the formation of electrical trees and water trees (water trees are tree-like structures formed due to the penetration of moisture into the cable insulation, which promotes uneven electric fields), ultimately leading to the formation of electrical breakdown paths, and even triggering cable failures in severe cases. Therefore, the potential risks brought by cable insulation failure will not only affect the stability of the power system, but may also cause serious damage to society and the economy. Thus, it can be seen that how to accurately detect the health status of cables has become an important issue that must be focused on in ensuring the safety of power supply.

[0003] However, in the existing technology, the aging degree of cables is mainly detected by focusing on surface factors such as service life, failure rate, and usage environment, but lacks in-depth research on the aging degree of cables based on the polarization properties of cables (such as polarization current, etc.). This not only makes the detection results of the cable aging degree have large errors, but also cannot effectively detect the moisture content of the cable. Summary of the Invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide a method and system for detecting the aging degree of a cable, which can not only improve the accuracy of the detection results of the cable aging degree, but also effectively detect the moisture content of the cable.

[0005] To solve the above technical problem, the embodiments of the present invention provide a method for detecting the aging degree of a cable, and the method includes the following steps:

[0006] Obtain the polarization current of the cable to be measured to fit the polarization current curve of the cable to be measured;

[0007] Determine the feature quantities to be extracted, and based on the polarization current curve of the cable to be measured, screen out the key feature quantities from the feature quantities to be extracted;

[0008] Import the selected key feature quantities into a trained classifier to obtain the aging degree of the cable to be measured.

[0009] Wherein, the specific steps of determining the feature quantities to be extracted and screening out the key feature quantities from the feature quantities to be extracted based on the polarization current curve of the cable to be measured include:

[0010] The overall fitting formula based on the cable polarization current is It is determined that the feature quantities to be extracted are τ, A 0 , σ dc , m, and n; where I(t) is the polarization current curve of the cable to be measured; C 0 is the capacitance of the cable to be measured, which is determined by the manufacturing material of the cable to be measured and has a fixed value; U 0 is the DC power supply voltage, which has a fixed value;

[0011] According to the polarization current curve of the cable to be measured, and through the preset analysis of variance method ANOVA, the feature quantities to be extracted τ, A 0 , σ dc , m, and n are analyzed to obtain the three feature quantities with the highest scores and output them as key feature quantities.

[0012] Among them, the classifier is constructed by one of the support vector machine (SVM), extreme gradient boosting (XGBoost), and random forest algorithm (RFA).

[0013] Among them, the method further includes:

[0014] When it is determined that A is included in the key feature quantities 0 , through the formula and combined with the obtained aging degree of the cable to be measured, the water content k of the cable to be measured is calculated; where x 0 , a, b, and c are all fixed coefficients, and the values of the four are determined by the obtained aging degree of the cable to be measured.

[0015] Among them, the aging degree includes thermal aging, mechanical damage aging, moisture aging, electrical aging, and chemical corrosion aging.

[0016] Among them, the polarization current of the cable to be measured is measured by a preset measuring device; where

[0017] The measuring device includes a DC power supply, an electrometer, and aluminum tape; one end of the DC power supply is connected to the core of the cable to be measured, and the other end is connected to one end of the electrometer, and is used to output a certain voltage of 200V to be applied to the cable to be measured;

[0018] The other end of the electrometer is connected to the outer semiconductor layer of the cable to be measured, and is used to continuously record the polarization current of the cable to be measured until its value is constant at a fixed value;

[0019] The aluminum tape is wound around the end face of the cable to be measured to form a protection ring, which is used to eliminate the influence of the surface leakage current on the electrometer.

[0020] An embodiment of the present invention also provides a cable aging degree detection system, including:

[0021] A cable polarization current acquisition unit, configured to acquire the polarization current of the cable to be measured, so as to fit the polarization current curve of the cable to be measured;

[0022] A key feature quantity screening unit, configured to determine the feature quantities to be extracted, and based on the polarization current curve of the cable to be measured, screen out the key feature quantities from the feature quantities to be extracted;

[0023] A cable aging degree detection unit, configured to import the selected key feature quantities into a trained classifier to obtain the aging degree of the cable to be measured.

[0024] Wherein, the key feature quantity screening unit includes:

[0025] A to-be-extracted feature quantity determination module, configured to determine that the feature quantities to be extracted are τ, A , σ 0 , m, and n based on the overall fitting formula of the cable polarization current as dc ; where I(t) is the polarization current curve of the cable to be measured; C 0 is the capacitance of the cable to be measured, which is determined by the production material of the cable to be measured and has a fixed value; U 0 is the DC power supply voltage, which has a fixed value;

[0026] A key feature quantity screening module, configured to analyze the feature quantities to be extracted, namely τ, A 0 , σ dc , m, and n according to the polarization current curve of the cable to be measured and through the preset analysis of variance method ANOVA, so as to obtain the three feature quantities with the highest scores and output them as key feature quantities.

[0027] Wherein, it further includes:

[0028] A cable water content detection unit, configured to calculate the water content k of the cable to be measured through the formula 0 and in combination with the obtained aging degree of the cable to be measured when it is determined that A is included in the key feature quantities; where x , a, b, and c are all fixed coefficients, and the values of the four are determined by the obtained aging degree of the cable to be measured. 0

[0029] Wherein, the aging degree includes thermal aging, mechanical damage aging, moisture aging, electrical aging, and chemical corrosion aging.

[0030] Implementing the embodiments of the present invention has the following beneficial effects:

[0031] ​1. Based on the polarization current, the present invention screens out key characteristic quantities and inputs them into a classifier to quickly obtain the aging degree of the cable, thereby not only improving the accuracy of the cable aging degree detection result;

[0032] 2. The present invention also quickly obtains the water content of the cable according to the key characteristic quantities and the detection result of the cable aging degree, so as to effectively detect the moisture content of the cable. Description of the Drawings

[0033] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, obtaining other drawings based on these drawings still belongs to the scope of the present invention.

[0034] Figure 1 It is a flowchart of a method for detecting the aging degree of a cable provided by an embodiment of the present invention;

[0035] Figure 2 It is a schematic connection diagram between a cable to be measured and a measuring device in a method for detecting the aging degree of a cable provided by an embodiment of the present invention;

[0036] Figure 3 It is a schematic structural diagram of a system for detecting the aging degree of a cable provided by an embodiment of the present invention. Detailed Embodiments

[0037] To make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings.

[0038] As Figure 1 shown, in an embodiment of the present invention, a method for detecting the aging degree of a cable is provided. The method includes the following steps:

[0039] Step S1: Obtain the polarization current of the cable to be measured to fit the polarization current curve of the cable to be measured;

[0040] Step S2: Determine the characteristic quantities to be extracted, and based on the polarization current curve of the cable to be measured, screen out the key characteristic quantities from the characteristic quantities to be extracted;

[0041] Step S3: Input the selected key characteristic quantities into a trained classifier to obtain the aging degree of the cable to be measured.

[0042] Specifically, in step S1, first, the cable to be measured is loaded on a preset measuring device to measure the polarization current. In one example, the production material of the cable to be measured is selected as cross-linked polyethylene (XLPE) material.

[0043] As shown Figure 2 in the figure, the measuring device includes a DC power supply, an electrometer, and aluminum tape; one end of the DC power supply is connected to the core of the cable under test, and the other end is connected to one end of the electrometer, which is used to output a certain voltage of 200V to be applied to the cable under test; the other end of the electrometer is connected to the outer semiconductor layer of the cable under test, which is used to continuously record the polarization current of the cable under test until its value is constant at a fixed value (the current value is stable after about 600s); the aluminum tape is wound around the end face of the cable under test to form a protective ring (i.e., Figure 2 the aluminum ring in

[0044] ), which is used to eliminate the influence of surface leakage current on the electrometer.

[0045] Secondly, for the polarization current recorded by the electrometer, which may contain noise, missing values, or abnormal data, through operations such as filtering, interpolation, and denoising, to obtain the polarization current of high-quality data.

[0046] Finally, based on the two-segment power-law model, the polarization current of high-quality data is fitted to obtain the polarization current curve I(t). where I(t) is the polarization current curve of the cable under test, and I 0 is the relaxation polarization current caused by dipole and interface polarization under a stable DC electric field; C 0 is the capacitance of the cable under test, which is determined by the manufacturing material of the cable under test and takes a fixed value; U 0 is the DC power supply voltage, which takes a fixed value (such as 200V).

[0047] Secondly, according to the above generalized dielectric response function h(t), a model is established, and the overall fitting formula for the cable polarization current is where τ, A 0 , σ dc , m, and n are all variables, that is, the characteristic quantities to be extracted.

[0048] Finally, according to the polarization current curve of the cable under test, and using the analysis of variance method ANOVA, the characteristic quantities to be extracted are τ, A 0 , σ dc , m, and n are analyzed to obtain the three characteristic quantities with the highest scores and output them as key characteristic quantities, specifically:

[0049] Step 1: The feature dataset was tested using the Shapiro-Wilk test (a well-known normality test). This test provides a specific parameter called the p-value, which indicates the normality of the dataset. If the p-value is greater than 0.05, the dataset is considered normal. In this example, after confirming normality, it was found that the p-values of τ, A 0 , σ dc , m, and n were 0.18, 0.06, 0.26, 0.76, and 0.65 respectively;

[0050] Step 2: ANOVA was applied to all feature sets to select the best features. ANOVA analysis is a statistical method used to determine whether there are any statistically significant differences between the means of two or more groups, which is done with the help of a parameter called the F-score.

[0051] Among them, the F-score is calculated as

[0052]

[0053] The higher the F-score, the more significant the impact of the feature on the target variable.

[0054] Step 3: Based on these F-scores, the 3 feature quantities with the highest F-scores were selected as the key feature quantities for subsequent further analysis. In one example, as shown in Table 1, the 3 feature quantities with the highest F-scores were τ, A 0 and n.

[0055] Table 1

[0056]

[0057] In step S3, the key feature quantities selected in step S2 were imported into the corresponding ML classifier to obtain the aging degree of the cable to be tested. Among them, the classifier was constructed based on one of support vector machine (SVM), extreme gradient boosting (XGBoost), and random forest algorithm (RFA). The aging degree includes but is not limited to thermal aging, mechanical damage aging, moisture aging, electrical aging, and chemical corrosion aging.

[0058] It should be noted that the input variables of the classifier can be reconstructed or adjusted according to different combinations of key feature quantities, and the training method of the classifier is usually implemented by common technical means in this field, which will not be elaborated here one by one. At the same time, the aging degree can also be further classified based on the above categories, and the specific design can be flexibly adjusted according to the actual situation, which will not be elaborated here either.

[0059] For example, the selected feature quantities τ, A 0And \(m\) and \(n\) are key feature quantities, which are fed into three well-known ML classifiers to establish a classification model for classifying the thermal aging degree of the cable. At this time, these three ML classifiers are respectively constructed by Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost) and Random Forest Algorithm (RFA), and 80% of the data set in each stage is set as the training set, and the remaining 20% of the data set is set as the test set.

[0060] In the embodiment of the present invention, after careful analysis of the above five feature quantities, it is determined that the feature quantity \(A\) 0 can be used as the target parameter for moisture evaluation because it maintains a well-fitted polynomial relationship with the water content under all aging levels. Therefore, the method further includes: when it is determined that \(A\) is included in the key feature quantities 0 , through the formula and combined with the obtained aging degree of the cable to be measured, the water content \(k\) of the cable to be measured is calculated; where \(x\) 0 , \(a\), \(b\), and \(c\) are all fixed coefficients, and the values of the four are determined by the obtained aging degree of the cable to be measured, that is, the values of the above four fixed coefficients for different aging degrees are different.

[0061] As Figure 3 shown, in the embodiment of the present invention, a cable aging degree detection system is provided, including:

[0062] A cable polarization current acquisition unit 110, configured to acquire the polarization current of the cable to be measured to fit the polarization current curve of the cable to be measured;

[0063] A key feature quantity screening unit 120, configured to determine the feature quantities to be extracted, and based on the polarization current curve of the cable to be measured, screen out the key feature quantities from the feature quantities to be extracted;

[0064] A cable aging degree detection unit 130, configured to import the selected key feature quantities into a trained classifier to obtain the aging degree of the cable to be measured.

[0065] Among them, the key feature quantity screening unit 120 includes:

[0066] A to-be-extracted feature quantity determination module, configured to determine the to-be-extracted feature quantities as \(\tau\), \(A\) , \(\sigma\) 0 , \(m\) and \(n\) based on the overall fitting formula of the cable polarization current as dc ; where \(I(t)\) is the polarization current curve of the cable to be measured; \(C\) 0 is the capacitance of the cable to be measured, which is determined by the production material of the cable to be measured and has a fixed value; \(U\) 0 is the DC power supply voltage, and its value is a fixed value;

[0067] A key feature quantity screening module, configured to analyze the feature quantities to be extracted, namely τ, A, σ, m, and n, according to the polarization current curve of the cable to be measured and by using a preset analysis of variance (ANOVA) method, so as to obtain the top three feature quantities with the highest scores and output them as key feature quantities. 0 σ dc m, and n, and output the three feature quantities with the highest scores as key feature quantities.

[0068] Further included are:

[0069] A cable water content detection unit, configured to calculate the water content k of the cable to be measured through the formula and in combination with the aging degree of the cable to be measured when it is determined that A is included in the key feature quantities. Here, x, a, b, and c are all fixed coefficients, and their values are determined by the aging degree of the cable to be measured. 0 When and in combination with the obtained aging degree of the cable to be measured. Here, x, a, b, and c are all fixed coefficients, and their values are determined by the aging degree of the cable to be measured. 0 a, b, and c are all fixed coefficients, and their values are determined by the aging degree of the cable to be measured.

[0070] The aging degree includes thermal aging, mechanical damage aging, moisture aging, electrical aging, and chemical corrosion aging.

[0071] Implementing the embodiments of the present invention has the following beneficial effects:

[0072] 1. Based on the polarization current, the present invention screens out key feature quantities and inputs them into a classifier to quickly obtain the aging degree of the cable, thereby not only improving the accuracy of the cable aging degree detection result;

[0073] 2. The present invention also quickly obtains the water content of the cable according to the key feature quantities and the detection result of the cable aging degree, thereby effectively detecting the water content of the cable.

[0074] It should be noted that in the above system embodiments, the included system modules are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional modules are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0075] Those of ordinary skill in the art can understand that all or part of the steps in implementing the above method embodiments can be completed by instructing relevant hardware through a program. The program can be stored in a computer-readable storage medium, such as a ROM / RAM, disk, optical disc, etc.

[0076] The above-disclosed are only the preferred embodiments of the present invention. Of course, the scope of the rights of the present invention cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A method for detecting cable aging degree, characterized in that: The method comprises the following steps: Obtaining the polarization current of the cable to be tested to fit the polarization current curve of the cable to be tested; Determining feature quantities to be extracted, and screening out key feature quantities from the feature quantities to be extracted based on the polarization current curve of the cable to be tested; The selected key feature quantities are imported into the trained classifier to obtain the aging degree of the cable to be tested.

2. The cable aging degree detection method according to claim 1, characterized in that: The specific steps of determining the feature quantity to be extracted and screening out the key feature quantity from the feature quantity to be extracted based on the polarization current curve of the cable to be tested include: The overall fitting formula based on the cable polarization current is: Determine the features to be extracted as τ, A0, σ dc , m and n; where I(t) is the polarization current curve of the cable to be tested; C0 is the capacitance of the cable to be tested, which is determined by the material of the cable to be tested and has a fixed value; U0 is the DC power supply voltage, which has a fixed value; According to the polarization current curve of the cable to be tested, and through the preset variance analysis method ANOVA, the feature quantities to be extracted are τ, A0, σ dc , m and n are analyzed to obtain the three feature quantities with the highest scores and output them as key feature quantities.

3. The cable aging degree detection method according to claim 2, characterized in that: The classifier is constructed based on one of support vector machine (SVM), extreme gradient boosting (XGBoost) and random forest algorithm (RFA).

4. The cable aging degree detection method according to claim 2, characterized in that: The method further comprises: When it is determined that the key feature quantity contains A0, the formula Combined with the obtained aging degree of the cable to be tested, the moisture content k of the cable to be tested is calculated; wherein x0, a, b, and c are all fixed coefficients, and the values ​​of the four are determined by the obtained aging degree of the cable to be tested.

5. The cable aging degree detection method according to claim 4, characterized in that: The aging degree includes heat aging, mechanical damage aging, moisture aging, electrical aging and chemical corrosion aging.

6. The cable aging degree detection method according to claim 1, characterized in that: The polarization current of the cable to be tested is measured by a preset measuring device; wherein, The measuring device comprises a DC power supply, an electrometer and an aluminum tape; one end of the DC power supply is connected to the cable core of the cable to be tested, and the other end is connected to one end of the electrometer, and is used to output a certain voltage of 200V and apply it to the cable to be tested; The other end of the electrometer is connected to the outer semiconductor layer of the cable to be tested, and is used to continuously record the polarization current of the cable to be tested until its value is constant at a fixed value; The aluminum tape is wound around the end surface of the cable to be tested to form a protective ring, which is used to eliminate the influence of surface leakage current on the electrometer.

7. A cable aging degree detection system, characterized in that: include: A cable polarization current acquisition unit is used to acquire the polarization current of the cable to be tested so as to fit the polarization current curve of the cable to be tested; A key feature quantity screening unit, used to determine the feature quantity to be extracted, and screen the key feature quantity from the feature quantity to be extracted based on the polarization current curve of the cable to be tested; The cable aging degree detection unit is used to import the selected key feature quantity into the trained classifier to obtain the aging degree of the cable to be tested.

8. The cable aging degree detection system according to claim 7, characterized in that: The key feature quantity screening unit comprises: The module for determining the feature quantity to be extracted is used for the overall fitting formula based on the cable polarization current: Determine the features to be extracted as τ, A0, σ dc , m and n; where I(t) is the polarization current curve of the cable to be tested; C0 is the capacitance of the cable to be tested, which is determined by the material of the cable to be tested and has a fixed value; U0 is the DC power supply voltage, which has a fixed value; The key feature quantity screening module is used to select the feature quantities to be extracted, τ, A0, σ, according to the polarization current curve of the cable to be tested and through the preset variance analysis method ANOVA. dc , m and n are analyzed to obtain the three feature quantities with the highest scores and output them as key feature quantities.

9. The cable aging degree detection system according to claim 8, characterized in that: Also includes: The cable moisture content detection unit is used to determine that the key characteristic quantity contains A0 through the formula Combined with the obtained aging degree of the cable to be tested, the moisture content k of the cable to be tested is calculated; wherein x0, a, b, and c are all fixed coefficients, and the values ​​of the four are determined by the obtained aging degree of the cable to be tested.

10. The cable aging degree detection system according to claim 9, characterized in that: The aging degree includes heat aging, mechanical damage aging, moisture aging, electrical aging and chemical corrosion aging.

Images