A cable insulation aging evaluation method based on high-voltage frequency domain dielectric response characteristic quantity

Abstract

The application provides a cable insulation aging evaluation method based on high-voltage frequency domain dielectric response characteristic quantity, under the condition that the insulation electric field intensity is 0.5kV / mm-1kV / mm, an ultralow frequency sweep test signal is applied to the measured cable under three different voltages, and the dielectric loss tangent value frequency domain response curve corresponding to each voltage level is obtained; at least two dielectric characteristic quantities are calculated based on the frequency domain response curve, the dielectric characteristic quantities include a first characteristic quantity representing the discrete degree of the dielectric loss tangent value under different voltages, and a second characteristic quantity representing the frequency domain integral characteristic of the dielectric loss tangent value under the highest test voltage; based on a preset correlation evaluation model taking the cable insulation mechanical performance aging index as a benchmark, the insulation health state score of the measured cable is calculated by combining the dielectric characteristic quantities, and the insulation aging state is determined according to the health state score.

Description

Technical Field

[0001] This invention belongs to the field of power equipment condition monitoring and fault diagnosis technology, specifically relating to a cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics. Background Technology

[0002] High-voltage power cables are the core infrastructure for power transmission in power systems, widely used in urban power grids, inter-regional power transmission, industrial power supply, and new energy grid connection. Their operational reliability directly affects the safety and stability of the power system. With the expansion of the power grid and the increasing service life of cables, the proportion of faults caused by cable insulation aging continues to rise. According to relevant operational statistics, in 6-500kV power cable lines, equipment aging faults account for 21.1% of the total faults, with main insulation aging faults accounting for 60.9% of all aging faults. The main insulation of high-voltage cables is primarily made of cross-linked polyethylene (XLPE). Its aging process is affected by a combination of electrical, thermal, environmental, and mechanical factors. Deterioration of insulation performance easily leads to accidents such as partial discharge and insulation breakdown, causing power outage losses and increased operation and maintenance costs. Therefore, accurately assessing the aging status of XLPE cable insulation has significant engineering value for achieving condition-based maintenance and ensuring the safe operation of the power grid.

[0003] Existing methods for assessing the aging of XLPE cable insulation are mainly divided into two categories: destructive testing and non-destructive testing. Destructive testing requires sampling and analysis of the cable insulation. While it can obtain precise data on the microscopic properties of the insulation material, it causes irreversible damage to the cable, making it unsuitable for on-site assessment of in-service cables. Furthermore, it involves long testing cycles and complex operations, failing to meet the needs of rapid on-site testing. Non-destructive testing enables the non-destructive assessment of in-service cables. However, conventional methods such as insulation resistance and power frequency dielectric loss testing, partial discharge detection, and temperature monitoring have limitations, including insensitivity to early aging, weak anti-interference capabilities, and the inability to directly quantify the degree of aging, making it difficult to achieve accurate assessment of the aging state of cable insulation.

[0004] Frequency domain dielectric response (FDS) technology, as a non-destructive insulation diagnostic method, has been gradually applied to the field of XLPE cable insulation aging assessment due to its advantages such as non-destructive testing, large information capacity, and sensitivity to changes in insulation microstructure. It obtains the dielectric response characteristics of the insulating medium through frequency sweep testing over a wide frequency range, reflecting the polarization relaxation process and microscopic defect information of the insulating material. However, existing related technologies are mostly concentrated on testing and analysis under normal pressure and low frequency conditions, and still face significant bottlenecks in practical engineering applications: First, existing assessments mostly use conventional parameters such as dielectric constant and dielectric loss factor as indicators, which lack sufficient correlation with the insulation aging mechanism and are easily affected by test conditions and cable structure, failing to accurately characterize the aging depth and distinguish different aging types; second, existing dielectric response analysis models are mostly developed for oil-paper insulation equipment, with poor applicability to XLPE extruded insulation materials, making it difficult to efficiently analyze complex dielectric spectrum data, and the assessment process is highly subjective; third, existing solutions mostly remain at the qualitative judgment of insulation aging status, making it difficult to achieve quantitative assessment of the degree of aging, lacking mature quantitative assessment standards and models, and failing to provide accurate technical support for cable operation and maintenance decisions. Summary of the Invention

[0005] To address the shortcomings and deficiencies of existing technologies, this invention provides a high-voltage cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics. This method primarily solves the industry pain points of existing frequency domain dielectric response assessment technologies, such as weak correlation between characteristic quantities and insulation aging mechanisms, poor compatibility with XLPE extruded insulation materials, and the inability to accurately and quantitatively assess the degree of aging. This invention employs an ultra-low frequency sweep test scheme with at least three different voltage levels to acquire frequency domain dielectric response data of cable insulation. Based on this, multi-dimensional core aging characteristic quantities are extracted, including the dispersion of the dielectric loss tangent, which reflects the nonlinear characteristics of the insulation dielectric response under different electric field strengths; the integral value of the frequency domain dielectric curve, which reflects the overall polarization and loss level of the insulation under strong fields; and the high-voltage dielectric loss tangent, which reflects the sensitivity of dielectric loss to voltage and frequency changes. This constructs a characteristic quantity system deeply correlated with the aging mechanism of cable insulation. Using the retention rate of elongation at break, a mechanical performance aging index specified in industry standards for cable insulation materials, as a benchmark, this invention establishes a high-precision quantitative correlation model with a goodness of fit of no less than 95% between each core characteristic quantity and the insulation aging state, providing an authoritative benchmark for the quantitative assessment of insulation aging. Furthermore, this invention uses a combined weighting method that integrates subjective and objective factors to determine the weight ratio of each characteristic quantity. Through reverse normalization and comprehensive calculation, a cable insulation health status score is obtained. Based on the score interval, insulation aging levels and corresponding operation and maintenance strategies are divided, achieving a non-destructive, accurate, and quantitative assessment of the aging state of high-voltage cable insulation. This invention does not require causing destructive damage to the cable, has strong anti-interference capabilities, and is highly sensitive to early insulation aging. It is suitable for on-site condition assessment of cross-linked polyethylene insulated high-voltage cables and can provide reliable technical support for condition-based maintenance and operation and maintenance decisions for power cables.

[0006] The specific technical solution adopted by this invention to solve its technical problem is as follows:

[0007] A cable insulation aging assessment method based on high voltage frequency domain dielectric response characteristics involves applying an ultra-low frequency sweep test signal to the cable under test at at least three different voltages to obtain the frequency domain response curves of the dielectric loss tangent corresponding to each voltage level.

[0008] At least two dielectric characteristics are calculated based on the frequency domain response curve. The dielectric characteristics include a first characteristic that characterizes the dispersion of the dielectric loss tangent under different voltages, and a second characteristic that characterizes the frequency domain integral characteristics of the dielectric loss tangent under the highest test voltage.

[0009] Based on a pre-set correlation evaluation model that uses the aging index of cable insulation mechanical properties as a benchmark, the insulation health status score of the tested cable is calculated in combination with the dielectric characteristic quantity, and the insulation aging status is determined based on the health status score.

[0010] Furthermore, the frequency range of the ultra-low frequency sweep test signal is 0.01Hz to 0.1Hz, and the test frequency points cover at least 4 frequency points within this frequency band; the at least three different voltages must all make the insulation electric field strength between 0.5kV / mm and 1kV / mm; during the test, the cable under test is first pre-pressurized until the dielectric loss tangent value stabilizes, and then the test is repeated multiple times for each test frequency point, and the average value is taken as the dielectric loss tangent value of that frequency point, and the acquisition time is not less than 3 complete test signal cycles.

[0011] Furthermore, the first characteristic quantity is the dispersion of the dielectric loss tangent value, and its calculation formula is as follows:

[0012]

[0013] In the formula, D is the dispersion of the dielectric loss tangent, and N is the total number of test frequency points. For testing voltage and , For the nth test frequency point, This represents the dielectric loss tangent measured at voltage U and frequency f.

[0014] The second characteristic quantity is the integral value of the high-voltage frequency domain dielectric curve, and the calculation formula is as follows:

[0015]

[0016] In the formula, A is the integral value of the high-voltage frequency domain dielectric curve. Here, f represents the dielectric loss tangent at the highest test voltage, and f is the test frequency. to This refers to the ultra-low frequency test band range.

[0017] In this invention, at least two dielectric characteristic quantities, namely the dispersion of dielectric loss tangent and the integral value of the high-voltage frequency domain dielectric curve, can be used to establish a strong quantitative correlation with the degree of cable insulation aging, thereby achieving accurate assessment of the insulation aging state. As a preferred embodiment, the high-voltage dielectric loss tangent can be added as a third characteristic quantity to further improve the anti-interference capability of the assessment results and the assessment accuracy under extreme scenarios.

[0018] Therefore, the dielectric characteristic further includes a third characteristic, which is the dielectric loss tangent measured at the highest test voltage and the lowest test frequency.

[0019] Furthermore, the aging index of the cable insulation mechanical properties is the elongation at break retention rate of the cable insulation material, which is the percentage of the current elongation at break of the cable insulation material to the initial elongation at break; the correlation evaluation model is a polynomial fitting model obtained by fitting multiple sets of cable sample test data with different aging degrees.

[0020] Furthermore, when calculating the insulation health status score, a combined weighting method integrating subjective and objective factors is used to determine the weight of each dielectric characteristic quantity. The combined weighting method is as follows: the subjective weight of each dielectric characteristic quantity is obtained by using the ordered binary comparison quantization method, the objective weight of each dielectric characteristic quantity is obtained by using the entropy weight method, and the final combined weight of each dielectric characteristic quantity is obtained by multiplication synthesis and normalization.

[0021] Furthermore, before calculating the insulation health status score, the measured values ​​of each dielectric characteristic quantity are first reverse normalized, and then the insulation health status score is calculated using the following formula:

[0022]

[0023] In the formula, S is the insulation health status score, and M is the total number of dielectric characteristics. The combined weights of the corresponding dielectric characteristics, , This is the value after inverse normalization of the corresponding dielectric characteristic.

[0024] Furthermore, based on the insulation health status score, the aging status of cable insulation is divided into three levels: insulation failure risk level, mid-aging level, and good insulation level. Different levels correspond to different operation and maintenance strategies.

[0025] Furthermore, the cable under test is a cross-linked polyethylene (XLPE) insulated high-voltage power cable; before testing, both ends of the cable under test are disconnected from other electrical equipment, the cable ends are treated to expose the main insulation layer and the surface is cleaned, during testing, the high-voltage end of the testing device is connected to the cable core, the grounding end is reliably grounded to the cable's metal sheath, and the signal acquisition end is connected to the host computer through optical fiber to achieve high and low voltage isolation.

[0026] And, a cable insulation aging assessment system for performing the cable insulation aging assessment method as described above, including an ultra-low frequency test module, a characteristic quantity calculation module, an aging assessment module, and a data storage module;

[0027] The ultra-low frequency test module is used to apply ultra-low frequency sweep test signals of multiple voltage levels to the cable under test and acquire the frequency domain response curve of the dielectric loss tangent value corresponding to each voltage level.

[0028] The characteristic quantity calculation module is used to calculate the corresponding dielectric characteristic quantity based on the frequency domain response curve;

[0029] The aging assessment module is used to calculate the insulation health status score and output the insulation aging status assessment result based on the preset correlation assessment model and combined weights.

[0030] The data storage module is used to store the preset correlation evaluation model and combined weight parameters.

[0031] Compared to existing technologies, this invention and its preferred embodiment achieve non-destructive assessment of the aging state of high-voltage cable insulation, without causing irreversible damage to the cable itself. It can be directly applied to on-site condition monitoring of in-service cables, overcoming the application scenario limitations of traditional destructive testing methods. It constructs a multi-dimensional feature system deeply correlated with the cable insulation aging mechanism, which, compared to conventional dielectric parameters commonly used in existing technologies, can more comprehensively and accurately characterize the aging degree of insulation materials, exhibiting higher sensitivity in identifying early insulation degradation. This effectively solves the problems of weak correlation between existing assessment indicators and aging mechanisms, and susceptibility to external interference. It uses the insulation mechanical performance aging indicators specified by industry standards. A quantitative correlation model was established based on this benchmark, enabling precise quantitative assessment of the aging degree of cable insulation. This breaks through the limitations of existing technologies that mostly rely on qualitative judgments. The assessment results have clear standard basis, significantly improving reliability and authority. A comprehensive assessment system was constructed using a weighting method that integrates subjective and objective factors, effectively reducing assessment bias and subjective interference caused by a single weighting method. The assessment results are more stable. At the same time, through graded aging state determination and corresponding operation and maintenance guidelines, it can directly provide practical technical support for condition-based inspection and preventive maintenance of power cables. The overall testing scheme and assessment process are adaptable to complex on-site conditions, easy to operate, and have good engineering applicability and promotion value. Attached Figure Description

[0032] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0033] Figure 1 This is a schematic diagram illustrating the correlation of aging assessment of high-voltage XLPE cables in an embodiment of the present invention;

[0034] Figure 2 This is a schematic diagram of the XLPE dumbbell-shaped sample structure in an embodiment of the present invention;

[0035] Figure 3 This is a graph showing the correlation between the dispersion of the dielectric loss tangent and the retention rate of the elongation at break in an embodiment of the present invention.

[0036] Figure 4 This is a graph showing the correlation between the integral value of the high-voltage frequency domain dielectric curve and the elongation at break retention rate in an embodiment of the present invention.

[0037] Figure 5 This is a graph showing the correlation between the high-voltage dielectric loss tangent and the elongation at break retention rate in an embodiment of the present invention.

[0038] Figure 6 This is a schematic diagram of the insulation condition assessment process for high-voltage XLPE cables according to an embodiment of the present invention;

[0039] Figure 7This is a schematic diagram of the on-site testing environment for evaluating the insulation aging of high-voltage cables according to an embodiment of the present invention. Detailed Implementation

[0040] To make the features and advantages of the present invention more apparent and understandable, specific embodiments are described below in detail:

[0041] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0042] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0043] This invention aims to overcome the shortcomings of existing technologies and provide a more comprehensive, precise, and interference-resistant method for assessing cable insulation aging. This method should be able to accurately obtain the dielectric properties of the insulation material during frequency domain dielectric response testing, extract characteristic quantities directly related to the insulation aging mechanism, establish a quantitative relationship between these characteristic quantities and the degree of aging, construct a dielectric response analysis model suitable for XLPE cable insulation, achieve accurate analysis of dielectric spectrum data, improve the interference resistance and data reliability of field testing, and realize quantitative assessment of the degree of aging and prediction of remaining life.

[0044] The implementation of the present invention will be further illustrated and described below with reference to the accompanying drawings through more specific embodiments:

[0045] This invention provides a cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics. According to GB / T 11026.2-2012 "Electrical Insulation Materials - Heat Resistance Part 2: Selection of Test Judgment Standards", for XLPE materials, the recommended endpoint values ​​for elongation at break and breakdown field strength are 50%. This embodiment constructs the correlation between frequency domain dielectric characteristics and the insulation state of XLPE cables based on this mechanical performance evaluation standard, that is, establishing the correlation between high-voltage cable frequency domain dielectric characteristics and elongation at break. An ordered binary comparison quantization method is used to objectively assign weights to three characteristics: the dispersion D of dielectric loss tangent, the integral value A of the high-voltage frequency domain dielectric curve, and the high-voltage dielectric loss tangent T, using the entropy weight method. Figure 1 As shown, the assignment results have high objectivity and evaluation accuracy.

[0046] Its implementation process specifically includes the following steps:

[0047] 1. Test Data Acquisition

[0048] The entire cable underwent thermal aging. The cable had a voltage rating of 220 kV, an insulation thickness of 28 mm, and a length of 5 m. Thermal aging was performed in a closed environment at room temperature, with the aging temperature set at 130℃. Ultra-low frequency dielectric loss parameters were measured every 14 days, for a total of 70 days. During testing, a 50 mm long cable sample was cut and rotary-cut, and the elongation at break was measured.

[0049] When conducting ultra-low frequency dielectric loss testing, the ultra-low frequency test voltage parameters need to be determined based on the voltage level (insulation thickness) of the cable under test. Tests show that when the electric field strength of the insulation is between 0.5kV / mm and 1kV / mm, ultra-low frequency testing can effectively reflect the aging condition of the cable. Therefore, when selecting voltages, it is advisable to choose three or more voltages that allow the insulation electric field strength to remain within this range. The voltage frequency should be adjusted according to the actual situation on site (0.01 Hz - 0.1 Hz). The data acquisition system adopts a synchronous acquisition architecture of a high-voltage divider + high-precision non-inductive resistor + high-speed data acquisition card. As a further preferred implementation, the high-voltage divider has an accuracy class of no less than 0.2 and a voltage division ratio of 1000:1; the high-precision non-inductive resistor uses a high-stability metal film non-inductive resistor with a resistance of 100Ω, an accuracy of ±0.05%, and a temperature coefficient not exceeding ±5ppm / ℃; the high-speed data acquisition card has a sampling rate of no less than 200kS / s, a resolution of no less than 16 bits, and at least two synchronous acquisition channels.

[0050] The current in the circuit is calculated by acquiring the voltage signal across the non-inductive resistor, and the voltage signal can be inversely calculated from the voltage across the resistor. Simultaneously, the acquisition time should be ensured to be no less than three complete test cycles to ensure the statistical validity of the data. In this invention, when measuring dielectric loss, a pre-pressure is first applied for 20 minutes to stabilize the dielectric loss value and avoid disturbance errors. Then, the dielectric loss value is repeatedly measured ten times at frequencies of 0.01, 0.02, 0.05, and 0.1 Hz, and the average of the ten tests is used as the dielectric loss value result for this group of aged samples for subsequent data processing.

[0051] When measuring elongation at break, XLPE slices were prepared into dumbbell-shaped specimens according to GB / T 1040.2-2006 standard, such as... Figure 2As shown. Tensile tests were conducted on XLPE cable specimens under different thermal stresses using an electronic universal tensile testing machine, with the machine's tensile speed set to 250 mm / min. Each group of specimens at different aging levels was tested ten times, and the average of the ten tests was taken as the experimental result for that group of aged specimens.

[0052] 2. Extraction of core features

[0053] From the ultra-low frequency dielectric loss test data obtained in step S1, three core feature quantities that can characterize the insulation aging state in multiple dimensions are extracted:

[0054] (1) Dispersion of dielectric loss tangent (D)

[0055] The sum of the relative deviations of the dielectric loss factor in the low-frequency range (preferably 0.01 Hz to 0.1 Hz) under three selected different voltages (voltages that cause the insulation electric field strength to be between 0.5 kV / mm and 1 kV / mm) is calculated to reflect the nonlinearity of the dielectric response of the insulation under strong fields and characterize the separation level of the frequency domain dielectric curves under different voltages. Its expression is:

[0056] (1)

[0057] In the formula, D represents the dispersion of dielectric loss tangent values ​​at different aging degrees, used to characterize the degree of separation of the frequency domain curves of dielectric loss tangent values ​​under different voltage levels. It reflects the nonlinear characteristics of the dielectric response of the insulating material under different electric field strengths. The larger the value, the higher the degree of separation of the dielectric loss curves under different voltages, and the more severe the insulation aging. n is the nth test frequency point; tan1δ, tan2δ, and tan3δ represent the three frequency domain dielectric curves tested, respectively; f n For testing the frequency domain, the value range is 0.01-0.1 Hz. A larger D indicates a greater degree of separation between the three frequency domain dielectric curves, and vice versa.

[0058] (2) Integral value of dielectric curve in high voltage frequency domain (A)

[0059] Calculate the area enclosed by the dielectric loss factor curve and the frequency axis in the low-frequency range (preferably 0.01 Hz to 0.1 Hz) under the highest test voltage to reflect the overall polarization and loss level of the insulation under a strong field. Its expression is:

[0060] (2)

[0061] In the formula, A is the integral value of the high-voltage frequency domain dielectric curve measured at the highest test voltage; tanδ is the frequency domain dielectric curve when the voltage level is U; f is the test frequency, with a value range of 0.01-0.1 Hz.

[0062] (3) High voltage dielectric loss tangent (T)

[0063] Generally, the dielectric loss tangent tends to increase with increasing voltage level or decreasing test frequency, and it becomes more sensitive to changes in voltage and frequency with increasing aging time. Therefore, the dielectric loss tangent is selected as an important criterion for cable aging. It is typically the dielectric loss value measured at the lowest test frequency and highest applied voltage.

[0064] (3)

[0065] 3. Establish a correlation model

[0066] A mathematical correlation model is established between the above three core characteristic quantities (D, A, T) and the physical quantity reflecting the degree of degradation of the mechanical properties of cable insulation (elongation at break H, i.e., the percentage of the current elongation at break relative to the initial value). The specific calculation process is as follows:

[0067] (1) Correlation between the dispersion of dielectric loss tangent and the percentage of failure value of elongation at break

[0068] By fitting the dispersion of dielectric loss tangent to the percentage of failure value at elongation at break, the correlation between the dispersion of dielectric loss tangent and the aging state of high-voltage XLPE cable insulation can be indirectly obtained, thereby assessing the degree of aging of the cable insulation. The correlation between the dispersion of dielectric loss tangent and the percentage of failure value at elongation at break is as follows: Figure 3 As shown.

[0069] The measurement results of the two methods were fitted, and the fitting results are shown in Equation 4. The goodness of fit (R²) is... 2 The value is approximately 95.5%, which can be used to demonstrate the correlation between the dispersion of dielectric loss tangent and the percentage of failure value at elongation at break. Based on Equation 4, the correlation between the dispersion of dielectric loss tangent and the aging condition of high-voltage cable insulation can be established through elongation at break. Substituting the failure value at elongation at break into Equation 4, the cable insulation reaches the failure value when the dispersion of dielectric loss tangent is 10.52.

[0070] (4)

[0071] (2) Correlation between the integral value of the high-voltage frequency domain dielectric curve and the percentage of failure value of elongation at break

[0072] Similarly, the relationship between the integral value of the high-voltage frequency domain dielectric curve and the insulation state of the high-voltage XLPE cable is established using the percentage of failure value at break elongation. Figure 4 As shown.

[0073] The fitting formula for the integral value A of the high-voltage frequency domain dielectric curve and the percentage of failure value H at break elongation is shown in Equation 5. Its goodness of fit (R) 2The value is approximately 97%, which can be used to demonstrate the correlation between the integral value of the high-voltage frequency domain dielectric curve and the percentage of failure value based on elongation at break. Substituting the failure value based on elongation at break into Equation 5, the insulation failure value is calculated to be 0.0027 when the integral value of the high-voltage frequency domain dielectric curve is used as the criterion.

[0074] (5)

[0075] (3) Correlation between high-pressure dielectric loss tangent and percentage of failure value with elongation at break

[0076] The relationship between the high-pressure dielectric loss tangent T and the percentage of failure value H based on elongation at break is as follows: Figure 5 As shown.

[0077] The fitting formula for the high-pressure dielectric loss tangent T and the percentage of failure value H based on the elongation at break is shown in Equation 6, and its goodness of fit (R) 2 The value is approximately 96%, which can be used to illustrate the relationship between the high-voltage dielectric loss tangent and the percentage of failure value based on elongation at break. When the elongation at break reaches the failure value, the corresponding failure value obtained from the high-voltage dielectric loss tangent is 0.042.

[0078] (6)

[0079] 4. Overall Health Status Score

[0080] Measuring the elongation at break of cable insulation can effectively assess its insulation condition. According to the insulation diagnosis specifications in standard GB / T 11026.2-2012, when the elongation at break drops to 50% of its initial value, the insulation is considered to have reached its failure point, indicating insulation performance failure. Therefore, the elongation at break can be used as a criterion for judging insulation aging and establishing the relationship between high-voltage frequency domain dielectric characteristics and cable insulation aging. Using the elongation at break as the basis for insulation aging assessment, and to make a more accurate judgment on the cable insulation condition, the percentage of failure value H based on the elongation at break is further subdivided. When H ≤ 50%, the insulation condition of the high-voltage cable is considered failed; when 50% < H ≤ 75%, the insulation condition is considered good; and when H > 75%, the insulation condition is considered excellent.

[0081] Substituting the insulation state boundary values ​​of 50% and 75% into equations 4, 5, and 6 yields the boundary values ​​when each characteristic quantity is used as the basis for insulation state evaluation. The evaluation basis for each characteristic quantity is shown in Table 1. By confirming the boundary values, the insulation state evaluation interval can be divided.

[0082] Table 1. Insulation state boundary values ​​corresponding to each characteristic quantity

[0083]

[0084] Since the extracted features have varying abilities to characterize the cable insulation condition, their roles in the cable insulation aging assessment process will also differ. Therefore, weights can be assigned to each feature, and the combined weights can be used to assess the cable insulation condition.

[0085] The ordered binary comparison quantization method is used to assign subjective weights to each feature quantity. Subjective weighting methods such as expert surveys and the analytic hierarchy process rely on experts providing precise initial weights. However, research on high-voltage frequency domain dielectric features is still limited, making it impossible to provide explicit initial weights. The ordered binary comparison quantization method only requires the ranking of the importance of each feature quantity to determine the subjective weights. A 0.1-0.9 scale is used to perform binary comparisons of the importance of each feature quantity. When two feature quantities have the same importance, the corresponding element in the comparison matrix is ​​set to 0.5. The calculation process is as follows:

[0086] Assume the feature set consists of the three features mentioned above, and assume that the three features are of equal importance. Using a scaling method, a binary comparison of the importance of each feature is performed, constructing a third-order comparison matrix R, as shown in Equation 7. This matrix has the following property: if two indicators are of equal importance, the corresponding element value is 0.5.

[0087] (7)

[0088] In the formula, r 11 =r 12 =…=r 33 =0.5 indicates that the importance of each feature is equal.

[0089] The aforementioned third-order comparison matrix R is a consistency matrix with all elements equal to 0.5, and its largest eigenvalue is... =3, the corresponding normalized feature vector is [1 / 3, 1 / 3, 1 / 3], which yields the subjective weight assignment result W for each feature. a =[1 / 3, 1 / 3, 1 / 3].

[0090] In information theory, entropy is a measure of uncertainty and can be used to determine the degree of dispersion of an indicator. This embodiment uses the entropy weighting method to assign objective weights. Since this method relies only on the dispersion of the data itself, the assignment results are objective. By combining it with the subjective weighting method described above, a comprehensive subjective and objective weighting can be achieved, improving the accuracy of the evaluation results. The process of assigning objective weights is as follows:

[0091] Cable samples with different aging levels were selected as sample i, and the values ​​of three characteristic quantities were calculated respectively. These characteristic quantities were then used as index j. Equation 8 was used to analyze the original data x. ijStandardize the data to obtain the standardized data x. ij '.

[0092] (8)

[0093] In the formula, max(x) j ) and min(x j ) represent the maximum and minimum values ​​of the j-th indicator, respectively. This step aims to limit the data to between 0 and 1, reducing the impact of maxima on the data.

[0094] Calculate the weight p of the i-th sample value in the j-th indicator. ij As shown in Equation 9.

[0095] (9)

[0096] According to the definition of information entropy, the information entropy E of the j-th indicator is... j It is calculated using Equation 10.

[0097] (10)

[0098] In the formula, n is the number of cable samples with different aging degrees, and in this embodiment, n=6; m is the number of dielectric characteristic quantities, and in this embodiment, m=3.

[0099] Information entropy E of each indicator j After the calculation is completed, the entropy weight w of each indicator is calculated using Equation 11. bj .

[0100] (11)

[0101] Will Figure 3-5 The objective weights of the three feature quantities D, A, and T are calculated using the method described above, and the final objective weights W of the three feature quantities D, A, and T are obtained. b The values ​​are 0.3471, 0.3293, and 0.3236, respectively.

[0102] As a further preferred embodiment, this example selects 6 groups of XLPE cable samples with different degrees of thermal aging as samples for entropy weight calculation. The aging time of the samples, the corresponding elongation at break retention rate H, and the measured values ​​of each characteristic quantity are shown in Table 2 below:

[0103] Table 2. Raw data of core characteristic quantities of samples at different aging levels

[0104]

[0105] Based on the above raw data, the entropy weight method was used to calculate the values ​​according to Equations 8-11. The objective weights of the three feature quantities D, A, and T were obtained as 0.2554, 0.3944, and 0.3502, respectively. The calculation process is as follows:

[0106] The above raw data are standardized according to Formula 8 to eliminate the influence of dimensions.

[0107] Calculate the proportion p of the sample values ​​for each indicator according to Formula 9. ij ;

[0108] Calculate the information entropy E of each indicator according to formula 10. j The information entropies of D, A, and T are 0.785202, 0.668298, and 0.705423, respectively.

[0109] The entropy weights of each indicator are calculated according to Equation 11, and the final objective weight result is W_b=[0.2554, 0.3944, 0.3502].

[0110] As a preferred embodiment, this method uses ordered binary comparison quantification as the subjective weighting method and entropy value method as the objective weighting method. To avoid obscuring particularly important information, the importance of the indicators is comprehensively considered, and therefore, subjective and objective weights are combined for comprehensive analysis. Let the subjective weight be W. a =[w a1 , w a2 , w a3 The objective weight is W. b =[w b1 , w b2 , w b3 The combined weight W = [w1, w2, w3] is obtained by adjusting the objective weight coefficient through subjective weight, and the calculation formula is shown in Equation 12.

[0111] (12)

[0112] The final combined weights of each feature are shown in Table 3.

[0113] Table 3 Weights of each feature quantity

[0114]

[0115] The test results show that as the degree of aging increases, the values ​​of the three characteristic quantities all tend to increase, and the larger the value of the characteristic quantity, the worse the performance of the electrical equipment. The normalization formula is shown in Equation 13 below.

[0116] (13)

[0117] In the formula, x maxx is the maximum critical value of the cable insulation characteristic. min The values ​​are the measured values ​​of the cable insulation characteristics before aging (the maximum value is selected as the extreme value during aging, Dmax=11.58, Dmin=8.47, and so on). x represents the measured value of the cable insulation characteristics. i This represents the normalized result for each feature quantity.

[0118] As shown in the above formula, after normalization, the closer the value of each characteristic quantity is to 1, the better the insulation performance. Let p i =1-x i After assigning weights to each characteristic quantity, the obtained cable insulation health status score S is shown in Equation 14.

[0119] (14)

[0120] Based on the insulation state boundary values ​​of each characteristic quantity in Table 1, the corresponding relationship between the calculated health state score S and the insulation aging state is shown in Table 4.

[0121] Table 4. Correspondence between health status score and insulation aging status

[0122]

[0123] The process of evaluating the insulation state of high-voltage cables based on high-voltage frequency domain dielectric response characteristics in this embodiment of the invention is as follows: Figure 6 As shown in the diagram. First, the frequency domain dielectric response curve of the cable is measured using a high-voltage ultra-low frequency dielectric measurement device. Three or more voltages are selected (ensuring the cable insulation electric field strength is between 0.5kV / mm and 1kV / mm), and the test frequency is between 0.01 Hz and 0.1 Hz. After obtaining the frequency domain dielectric response curve spectrum, the dispersion D of the dielectric loss tangent, the integral value A of the high-voltage frequency domain dielectric curve, and the value T of the high-voltage dielectric loss tangent are calculated. Finally, the cable insulation condition is evaluated based on the combined weights, and the aging state of the cable insulation is determined after calculating the health score S.

[0124] 5. On-site case analysis

[0125] To verify the effectiveness of the above-mentioned high-voltage cable insulation condition assessment method, this embodiment will assess and analyze the insulation condition of the high-voltage cable in the field. A 300-meter section of a 110kV high-voltage cable line was selected for high-voltage frequency domain dielectric performance testing. This cable section has been in operation for a long time and is currently undergoing power outage maintenance testing in an environment such as... Figure 7 As shown.

[0126] The testing steps are as follows:

[0127] (1) After confirming that all connections at both ends of the cable to be tested are disconnected, peel off the outer sheath, armor and outer semiconducting layer for 20 cm length at both ends of the cable to expose the main insulation of the cable. Wipe the surface of the main insulation of the cable with alcohol to remove the influence of impurities.

[0128] (2) The high-voltage test terminal of the ultra-low frequency dielectric measurement device is connected to the cable core, and the device grounding point and the cable aluminum sheath grounding point are reliably grounded. The signal control terminal of the ultra-low frequency device is connected to the PC via USB, and the data receiving device is connected to the PC via optical fiber to achieve high and low voltage isolation requirements.

[0129] (3) Set the signal frequency to 0.1, 0.05, 0.02 and 0.01 Hz, and the output voltage to 14kV, 21kV and 28kV, outputting them sequentially from low to high. After the test data stabilizes, record the data and take the average of 5 sets of data as the test result. Record the test dielectric loss tangent value at each voltage level. The test results are shown in Table 5.

[0130] Table 5. Field cable test results

[0131]

[0132] After obtaining the test results, Equations 1 and 2 were used to calculate the test data. The resulting dispersion D of the dielectric loss tangent and the integral value A of the high-voltage frequency domain dielectric curve were 10.31 and 0.002, respectively. Simultaneously, the high-voltage dielectric loss tangent T, directly read from the measurement results, was 0.033. Substituting the calculated characteristic quantities into Equation 12 for combined weight matching and normalization via Equation 13, the final health score S of the tested high-voltage cable was calculated using Equation 14, which was 0.4884. Based on the correspondence between the health score and insulation aging status in Table 4, it can be determined that the insulation status of the tested high-voltage cable is in a critical state, and the maintenance cycle should be shortened.

[0133] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0134] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

[0135] This invention is not limited to the preferred embodiment described above. Anyone inspired by this invention can derive various other forms of cable insulation aging assessment methods based on high-voltage frequency domain dielectric response characteristics. All equivalent variations and modifications made within the scope of the claims of this invention shall fall within the scope of this invention.

Claims

1. A method for evaluating cable insulation aging based on high-voltage frequency domain dielectric response characteristics, characterized in that: At at least three different voltages, apply an ultra-low frequency sweep test signal to the cable under test and obtain the frequency domain response curve of the dielectric loss tangent value corresponding to each voltage. At least two dielectric characteristics are calculated based on the frequency domain response curve. The dielectric characteristics include a first characteristic that characterizes the dispersion of the dielectric loss tangent under different voltages, and a second characteristic that characterizes the frequency domain integral characteristics of the dielectric loss tangent under the highest test voltage. Based on a pre-set correlation evaluation model that uses the aging index of cable insulation mechanical properties as a benchmark, the insulation health status score of the tested cable is calculated in combination with the dielectric characteristic quantity, and the insulation aging status is determined based on the health status score.

2. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics as described in claim 1, characterized in that: The frequency range of the ultra-low frequency sweep test signal is 0.01Hz to 0.1Hz, and the test frequency points cover at least 4 frequency points within this band; the at least three different voltages must make the electric field strength of the insulation between 0.5kV / mm and 1kV / mm. During the test, the cable under test is first pre-pressurized until the dielectric loss tangent value stabilizes, and then the test is repeated multiple times for each test frequency point. The average value is taken as the dielectric loss tangent value of that frequency point, and the acquisition time is not less than 3 complete test signal cycles.

3. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics as described in claim 1, characterized in that: The first characteristic quantity is the dispersion of the dielectric loss tangent value, and the calculation formula is: In the formula, D is the dispersion of the dielectric loss tangent, and N is the total number of test frequency points. For testing voltage and , For the nth test frequency point, This represents the dielectric loss tangent measured at voltage U and frequency f. The second characteristic quantity is the integral value of the high-voltage frequency domain dielectric curve, and the calculation formula is as follows: In the formula, A is the integral value of the high-voltage frequency domain dielectric curve. Here, f represents the dielectric loss tangent at the highest test voltage, and f is the test frequency. to This refers to the ultra-low frequency test band range.

4. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics according to claim 3, characterized in that: The dielectric characteristic also includes a third characteristic, which is the dielectric loss tangent measured at the highest test voltage and the lowest test frequency.

5. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics according to claim 1, characterized in that: The aging index of the cable insulation mechanical properties is the elongation at break retention rate of the cable insulation material, which is the percentage of the current elongation at break to the initial elongation at break of the cable insulation material; the correlation evaluation model is a polynomial fitting model obtained by fitting multiple sets of cable sample test data with different aging degrees.

6. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics according to claim 1, characterized in that: When calculating the insulation health status score, a combined weighting method integrating subjective and objective factors is used to determine the weight of each dielectric characteristic quantity. The combined weighting method is as follows: the subjective weight of each dielectric characteristic quantity is obtained by using the ordered binary comparison quantization method, the objective weight of each dielectric characteristic quantity is obtained by using the entropy weight method, and the final combined weight of each dielectric characteristic quantity is obtained by multiplication synthesis and normalization.

7. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics as described in claim 6, characterized in that: Before calculating the insulation health status score, the measured values ​​of each dielectric characteristic quantity are first reverse normalized, and then the insulation health status score is calculated using the following formula: In the formula, S is the insulation health status score, and M is the total number of dielectric characteristics. The combined weights of the corresponding dielectric characteristics, , This is the value after inverse normalization of the corresponding dielectric characteristic.

8. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics according to claim 1, characterized in that: Based on the insulation health status score, cable insulation aging is divided into three states: abnormal state, warning state, and normal state. Different levels correspond to different operation and maintenance strategies.

9. The cable insulation aging assessment method based on high-voltage frequency domain dielectric response characteristics according to claim 1, characterized in that: The cable under test is a cross-linked polyethylene (XLPE) insulated high-voltage power cable. Before testing, both ends of the cable under test are disconnected from other electrical equipment, the cable ends are exposed to expose the main insulation layer and the surface is cleaned. During testing, the high-voltage end of the testing device is connected to the cable core, the grounding end is reliably grounded to the cable's metal sheath, and the signal acquisition end is connected to the host computer through optical fiber to achieve high and low voltage isolation.

10. A cable insulation aging assessment system, characterized in that, The method for performing any one of claims 1-9 includes an ultra-low frequency testing module, a characteristic quantity calculation module, an aging assessment module, and a data storage module; The ultra-low frequency test module is used to apply ultra-low frequency sweep test signals of multiple voltage levels to the cable under test and acquire the frequency domain response curve of the dielectric loss tangent value corresponding to each voltage level. The characteristic quantity calculation module is used to calculate the corresponding dielectric characteristic quantity based on the frequency domain response curve; The aging assessment module is used to calculate the insulation health status score and output the insulation aging status assessment result based on the preset correlation assessment model and combined weights. The data storage module is used to store the preset correlation evaluation model and combined weight parameters.

Images