A cold shrink cable terminal mounting quality detection method based on insulation resistance stability

Abstract

The application discloses a kind of based on insulation resistance stability cold shrink cable terminal head installation quality detection method, its steps are: multiple incremental direct-current test voltages are sequentially applied to the conductor between cold shrink cable terminal head and ground shielding layer, multiple charge-discharge measurement cycles are repeatedly executed under each voltage level, and insulation resistance repeated measurement matrix is formed, mean value and variation coefficient are calculated by voltage level after temperature correction, and by mean value stepwise drop amplitude and variation coefficient mutation analysis, identify mutation voltage, combined with pearson correlation coefficient, form installation quality feature vector, determine defect type according to feature vector combination mode and output comprehensive evaluation result.The application can effectively identify interface air gap, insulation layer processing and other installation defects without damaging the structure of the terminal head, improve the accuracy and reliability of the installation quality detection of the cold shrink cable terminal head.

Description

TECHNICAL FIELD

[0001] The invention belongs to the technical field of installation quality inspection of power cable accessories, and specifically relates to a method for detecting the installation quality of cold-shrinkable cable terminals based on insulation resistance stability. BACKGROUND ART

[0002] Cross-linked polyethylene (XLPE) insulated cables have been widely used in urban distribution networks, power supply for industrial and mining enterprises, and power supply and distribution systems for various projects due to their excellent electrical, heat-resistant and mechanical properties. In the laying and operation of XLPE cable lines, cable terminals are key accessories connecting cable conductors and external equipment, and their installation quality is directly related to the safe operation of the entire cable line. Cold-shrinkable cable terminals are widely used in medium and low voltage distribution cable accessories due to their advantages of convenient on-site installation, no need for heating shrinkage, and wide adaptation temperature range, and have gradually become one of the mainstream forms of XLPE insulated cable terminals.

[0003] At present, the quality acceptance after the installation of cold-shrinkable cable terminals mainly relies on a megohmmeter for a single insulation resistance test, that is, measuring the insulation resistance value between the conductor and the grounding shielding layer under a single fixed voltage, and comparing it with the minimum qualified threshold specified in national standards or industry standards. If the insulation resistance value is higher than the threshold, the installation is judged as qualified. In some engineering practices, absorption ratio or polarization index is also used as an auxiliary criterion, but it still belongs to a single or few measurement method under a single voltage condition.

[0004] In actual use, the following problems have occurred: the above single measurement method can only reflect the instantaneous absolute level of insulation resistance, and cannot reveal the stability characteristics of insulation resistance under repeated voltage application. Common installation defects of cold-shrinkable terminals, such as residual air gaps at the interface, uneven contact pressure between the cold-shrinkable sleeve and the cable insulation layer, insufficient surface treatment of the insulation layer, etc., often do not cause a significant decrease in insulation resistance under low-voltage single measurement conditions. However, under higher voltage or repeated voltage application conditions, these minor defects will cause intensified fluctuation or progressive attenuation of insulation resistance, and existing detection methods lack effective recognition ability for such hidden defects. SUMMARY OF THE INVENTION

[0005] The existing installation quality inspection methods for cold-shrinkable cable terminals only rely on single insulation resistance measurement under a single voltage, cannot identify the stability difference of insulation resistance under repeated voltage application at multiple voltages, and are difficult to distinguish different types of installation defects. Based on this, the present invention is proposed as follows: A method for detecting installation quality of cold-shrinkable cable terminals based on insulation resistance stability, comprising the following steps: Step S01, collecting the ambient temperature at the installation site of the cold-shrinkable cable terminal , applying voltage sequentially between the conductor and the grounding shielding layer of the cold-shrinkable cable terminal Increasing DC test voltage At each voltage level, apply voltage for a preset charging time until the insulation resistance reading stabilizes and record it. Then, disconnect the voltage and wait for the insulation to recover for a preset discharge time to complete one measurement cycle. Repeat this process at each voltage level. Each measurement cycle records all steady-state insulation resistance values, forming an insulation resistance repeatability measurement matrix. , , This step involves repeated measurements at multiple voltage levels to obtain the complete response characteristics of the insulation resistance, providing a data foundation for subsequent stability analysis.

[0006] Step S02, based on ambient temperature With preset reference temperature The difference is used to correct the temperature of all elements in the repeated insulation resistance measurement matrix using a preset temperature correction coefficient, resulting in a corrected insulation resistance matrix. This step eliminates the influence of ambient temperature on the insulation resistance measurement value, making the measurement results under different temperature conditions comparable.

[0007] Step S03: Calculate the mean value of the corrected insulation resistance matrix for each voltage level. and coefficient of variation The mean sequence is differentially analyzed at each level to identify the voltage at which the voltage drop between adjacent voltage levels first exceeds a preset abrupt change threshold, which is denoted as the mean abrupt change voltage. The coefficient of variation sequence is analyzed step by step to identify the voltage at which the coefficient of variation first exceeds a preset stability threshold, which is denoted as the stability mutation voltage. This step simultaneously characterizes the change in insulation state with voltage from two dimensions: mean decay trend and repeated measurement volatility.

[0008] Step S04: Calculate the Pearson correlation coefficient between the mean-decreasing sequence and the coefficient of variation sequence. , the mean sudden voltage Stability sudden voltage Pearson correlation coefficient and the corrected mean of the highest voltage level This step constructs an installation quality feature vector, compressing multidimensional statistical indicators into a feature vector to achieve a structured representation of the installation quality status.

[0009] Step S05: Determine the type of installation defect based on the combination pattern of each component of the installation quality feature vector, calculate the comprehensive installation quality assessment index, and output the defect type and comprehensive assessment result. This step maps the feature vector to specific defect diagnosis conclusions, providing quantitative quality feedback to installers.

[0010] Furthermore, in step S01, the preset charging time is not less than 60 seconds, the preset discharging time is not less than the preset charging time, and the interval between two adjacent measurement cycles is not less than the preset discharging time. Sufficient charging time ensures that the absorbed current fully decays to a steady state, and sufficient discharging time and cycle intervals ensure that the polarization of the insulating medium is completely released, thereby guaranteeing the independence and data reliability of each repeated measurement.

[0011] Further, in step S02, the temperature correction adopts an exponential correction model, and the correction formula is as follows: ,in For each The insulation resistance correction factor corresponding to temperature changes. The exponential correction model conforms to the physical law that the resistivity of polymer insulating materials changes exponentially with temperature, and can provide higher correction accuracy over a wider temperature range compared to linear correction.

[0012] Further, in step S05, the defect type determination includes the following rule: when the stability sudden voltage... Less than the mean sudden voltage And Pearson correlation coefficient When the voltage exceeds a preset threshold, it is determined to be an interface air gap defect; when the stability change voltage... Not less than the mean sudden voltage And Pearson correlation coefficient When the value exceeds a preset threshold, it is determined to be a high-voltage activated interface air gap defect. When it is determined to be an interface air gap defect or a high-voltage activated interface air gap defect and the corrected mean of the highest voltage level is lower than the preset qualified threshold, it is superimposed as a composite defect of air gap and insulation layer treatment. When the coefficient of variation of all voltage levels is lower than the preset stability threshold and the corrected mean of the highest voltage level is lower than the preset qualified threshold, it is determined to be an insulation layer treatment defect. When there is a coefficient of variation of at least one voltage level that is not lower than the preset stability threshold and the Pearson correlation coefficient does not exceed the preset relevant threshold, it is determined to be an atypical defect. In the above rules, interface air gap defects and insulation layer treatment defects are mutually exclusive in the coefficient of variation dimension—the former requires the coefficient of variation of some voltage levels to increase, and the latter requires all coefficients of variation to be lower than the threshold—therefore, the two will not directly overlap. The composite defect determination identifies poor insulation layer treatment masked by air gap symptoms by superimposing the corrected mean condition on the basis of the air gap determination. The atypical defect determination covers the case where the coefficient of variation increases but the decrease and fluctuation lack statistical coupling, ensuring that there are no judgment gaps in the classification logic.

[0013] Further, in step S05, the comprehensive installation quality assessment index The weighted comprehensive evaluation index is calculated by weighting the normalized value of the sudden voltage, the absolute value of the Pearson correlation coefficient, and the normalized value of the corrected mean, with the sum of all weighting coefficients being 1. This index maps multidimensional feature information into a single quality score, facilitating rapid determination of installation quality levels on-site.

[0014] Further, in step S01, the The lowest voltage of an incremental DC test voltage Not less than 0.5 times the rated voltage of the cable, maximum voltage The voltage increment should not exceed five times the rated voltage of the cable, and the voltage increment between adjacent voltage levels should be set in a geometric or arithmetic manner. This voltage range covers the range from the normal operating voltage to the high voltage range where weak points in the insulation may be exposed, and the geometric or arithmetic increment setting ensures that the spacing between each voltage level is reasonable.

[0015] Furthermore, in step S03, the coefficient of variation is calculated as the ratio of the standard deviation to the mean, and the gradual decrease in the mean is calculated using a relative decrease rate. Using a relative decrease rate instead of an absolute decrease eliminates the impact of differences in insulation resistance base caused by different cable specifications and lengths on abrupt change identification.

[0016] Further, in step S02, the preset reference temperature The temperature is 20°C. The correction factor is 10°C. The value range is 1.3 to 1.5. This parameter setting conforms to the empirical values ​​of the temperature characteristics of XLPE insulation materials and is suitable for common medium and low voltage cross-linked polyethylene insulated cable cold shrink termination testing scenarios.

[0017] Further, in step S04, when the decrease in voltage between all adjacent voltage levels in the mean sequence does not exceed a preset abrupt change threshold, the mean abrupt change voltage... Assign the value as the highest test voltage When the values ​​of all voltage levels in the coefficient of variation sequence do not exceed the preset stability threshold, the stability mutation voltage is... Assign the value as the highest test voltage This assignment strategy ensures that the installation quality feature vector retains its complete structure even in scenarios without sudden changes, avoiding abnormalities in subsequent defect diagnosis logic due to missing components.

[0018] Further, in step S01, for each voltage level After each measurement cycle is completed, outliers that deviate from the temporary arithmetic mean of that voltage level by more than 3 times the temporary standard deviation are removed, and additional measurement cycles are performed to maintain the voltage level. Each measurement is valid. The outlier removal mechanism effectively eliminates biased data caused by external interference or instrument transient failures, improving the data quality of the insulation resistance repetitive measurement matrix.

[0019] Compared with existing technologies, this invention has the following advantages: First, by constructing an insulation resistance matrix through repeated measurements at multiple voltage levels, it fully explores the information on the changes in insulation state with voltage and repeated pressure application, overcoming the limitation of traditional single measurement methods that only obtain instantaneous absolute values. Second, by introducing the coefficient of variation as a stability index for repeated measurements, it can sensitively reflect the insulation resistance fluctuations caused by minor installation defects, improving the detection rate of latent defects. Third, by using the Pearson correlation coefficient between the mean reduction and the coefficient of variation to reveal the coupling degree of the two degradation modes, it provides a statistical basis for defect classification. Fourth, the defect classification rules distinguish between air gap defects and insulation layer treatment defects based on the mutual exclusion relationship in the coefficient of variation dimension, and identify composite defects masked by air gap symptoms by superimposing and correcting the mean conditions. At the same time, it sets atypical defect categories to cover abnormal fluctuations but low coupling conditions, forming a complete judgment system with no overlap and no omissions in the parameter space, realizing the transformation from a binary judgment of "qualified / unqualified" to refined defect diagnosis. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the overall process of the cold-shrink cable termination installation quality inspection method of the present invention. Figure 2 This is a sub-flowchart of the insulation resistance statistical analysis and mutation identification in step S03 of the present invention; Figure 3 This is a sub-flowchart of defect classification, diagnosis and comprehensive evaluation in step S05 of the present invention. Detailed Implementation

[0021] The following detailed description, in conjunction with the accompanying drawings, illustrates specific embodiments of the present invention to enable those skilled in the art to more clearly understand the invention. This embodiment uses a cold-shrink outdoor termination of a 10kV XLPE insulated cable as the testing object to describe the complete implementation process of the method of the present invention. The present invention proposes a method for testing the installation quality of cold-shrink cable terminations based on insulation resistance stability, the steps of which are as follows: Step S01: Collect ambient temperature and repeatedly perform charge-discharge measurement cycles at multiple increasing voltage levels to construct an insulation resistance repetitive measurement matrix. After the cold-shrink cable termination is installed and before it is put into operation, a temperature sensor is first used to collect the ambient temperature at the installation site. This temperature will serve as the basis for subsequent temperature correction. The temperature measurement location should be selected close to the air environment or the surface of the cable outer sheath to reflect the temperature environment of the terminal insulation as closely as possible. Then, connect the high-voltage end of the insulation resistance tester to the conductor of the cold-shrink cable terminal, and the low-voltage end (grounding end) to the grounded shielding layer to establish a measurement circuit. Before measurement, ensure that the terminal is completely isolated from the power grid and other equipment, and that the conductor and shielding layer are fully discharged to avoid interference from residual charge on the initial measurement value.

[0022] The test voltage is set using a stepped increase method. The purpose of this step-increase is to gradually increase the electrical stress on the insulation from low to high voltage, so that defects of different severity are activated and exposed within their respective sensitive voltage ranges. (Pre-determined...) DC test voltage , ... ,satisfy Minimum test voltage Typically not lower than the cable's rated voltage 0.5 times that, maximum test voltage Normally does not exceed the rated voltage of the cable. Five times. The voltage increment between adjacent voltage levels can be set using an arithmetic progression (e.g., increasing by 500V per level) or a geometric progression (e.g., multiplying each level by a fixed factor) to cover the range from conventional operating voltage to higher voltages where weak points in the insulation may be exposed. In practical engineering, for 10kV XLPE cable cold-shrink terminations, a typical voltage setting might be... =2500V =5000V =7500V =10000V, that is =4, voltage increment is 2500V.

[0023] At each voltage level The following charge / discharge measurement cycle is then executed: Apply DC test voltage. Maintain preset charging time (For no less than 60 seconds), wait for the absorbed current to fully decay and for the insulation resistance reading to stabilize before recording the steady-state insulation resistance value. Then disconnect the voltage source and short-circuit the conductor to the grounded shielding layer to discharge, maintaining the preset discharge time. (not less than) This allows the polarization charge accumulated in the insulating medium to be fully released. For example... Figure 1 As shown, the above measurement cycle is repeated at each voltage level. Second-rate( Typically, 3 to 5 cycles are performed, with the interval between two adjacent cycles not less than the discharge duration, to ensure consistent initial conditions for each measurement. The discharge duration is set based on the fact that a slow polarization process exists in the XLPE insulating medium. If the discharge is insufficient, the residual polarization charge will accumulate during the next charging process, leading to a systematically higher subsequent measurement value and reducing the comparability of repeated measurement results.

[0024] Each voltage level After each measurement cycle is completed, the temporary arithmetic mean and temporary standard deviation of all measurements at that voltage level are calculated. Measurements deviating from the temporary mean by more than three times the temporary standard deviation are identified as outliers and removed. Outliers may be caused by transient interference from the measuring instrument, sudden changes in contact resistance due to loose wiring, or other non-insulation condition changes. After removing outliers, the corresponding number of measurement cycles are performed to maintain the voltage level. One valid measurement value. Finally, in At each voltage level, obtain A matrix of repeated insulation resistance measurements is formed from the steady-state insulation resistance values. ,in Indicates the voltage level sequence number. This indicates the measurement cycle number within the same voltage level.

[0025] Step S02: Apply a temperature correction factor to the insulation resistance repetitive measurement matrix to eliminate the influence of ambient temperature. The volume resistivity of polymeric insulating materials such as XLPE decreases with increasing temperature; therefore, insulation resistance values ​​measured at different ambient temperatures are not directly comparable. This step is based on ambient temperature. With preset reference temperature The difference is related to the repeated measurement matrix of insulation resistance. Temperature correction is applied to all elements in the matrix to obtain the corrected insulation resistance matrix. .

[0026] The temperature correction uses an exponential correction model, and its correction formula is as follows:

[0027] in, The preset reference temperature is usually set to 20℃; This is the temperature step size, typically taken as 10℃; For each For XLPE insulation material, the insulation resistance correction factor corresponding to temperature change is as follows: The empirical value range is 1.3 to 1.5. The physical basis of this exponential model is that the thermal excitation process of charge carriers in polymer insulating materials follows the Arrhenius relation, and the resistivity changes approximately exponentially with temperature. When the ambient temperature... Higher than reference temperature hour, If the value is negative, the correction factor is... If the value is less than 1, the lower measured value at high temperature will be corrected upwards to the equivalent value at the reference temperature; when Below When the correction factor is greater than 1, the excessively high measurement value at low temperatures will be corrected downwards. This is based on commonly used engineering parameters. =1.4、 Taking 10℃ as an example, if the temperature at the installation site is... =30℃, reference temperature =20℃, then the correction factor is The equivalent value at 20℃ is obtained by multiplying the measured insulation resistance by 0.714. This correction range is consistent with engineering experience, verifying the rationality of the model parameter selection.

[0028] After temperature correction, the matrix All elements in the sample are normalized to the reference temperature. The equivalent insulation resistance value under the given conditions eliminates the interference of ambient temperature differences on subsequent statistical analysis.

[0029] It should be noted that the exponential correction model in and It exhibits good accuracy within a temperature range not exceeding 30℃. When the temperature difference is too large, it is advisable to consult the temperature-resistivity curve provided by the insulation material manufacturer for segmented correction. In typical engineering environments for 10kV XLPE cable cold-shrink terminations, the installation site temperature is typically... ℃ to Between ℃, take The maximum temperature difference is 30℃ when the temperature is 20℃, and the above exponential model can meet the correction accuracy requirements.

[0030] Step S03: Calculate the mean and coefficient of variation of the corrected insulation resistance matrix for each voltage level and identify sudden voltage changes. For the modified insulation resistance matrix Statistical analysis was performed row by row according to voltage level. For example... Figure 2 As shown, first calculate each voltage level Down Arithmetic mean of the corrected insulation resistance values:

[0031] mean This reflects the central tendency of insulation resistance at this voltage level. Under normal insulation conditions, the average insulation resistance typically decreases slowly but remains at a high level as the test voltage increases. This slow decrease is mainly due to the nonlinear conductivity of XLPE insulation material under higher electric field strengths—as the electric field strength increases, space charge injection and carrier migration in the insulator are enhanced, leading to a slight decrease in the equivalent volume resistivity. For well-installed cold-shrink terminals, the average resistance reduction between voltage levels typically does not exceed [a certain percentage]. The voltage level decreases gradually; however, if there are installation defects, the voltage level may experience a significant acceleration in the decrease, with the drop rate increasing sharply.

[0032] Then, calculations were performed at each voltage level. The standard deviation of the second correction insulation resistance value and coefficient of variation:

[0033] coefficient of variation The coefficient of variation (CV) measures the relative dispersion of repeated measurements at the same voltage level and can be considered an indicator of the repeatability stability of insulation resistance at that voltage level. A smaller CV indicates higher consistency in insulation resistance readings after repeated voltage applications, meaning better repeatability stability. Under good insulation conditions, repeated insulation resistance measurements should be highly consistent with a very small CV (typically below 0.05). If defects such as interface air gaps exist, intermittent discharges may occur in these gaps at specific voltage levels, leading to significant fluctuations in insulation resistance values ​​across measurements, a marked increase in the CV, and consequently, a deterioration in repeatability stability. Therefore, the trend of the CV with voltage level reveals the response characteristics of insulation defects under different voltage excitations and is an important basis for distinguishing different types of installation defects.

[0034] In obtaining the mean sequence Then, a step-by-step differential analysis is performed. The relative descent rate between adjacent voltage levels is calculated:

[0035] Check step by step along the direction of voltage increase ,when First time exceeding the preset mutation threshold (For example, 0.15, meaning a drop exceeding 15%), the corresponding voltage will be... denoted as mean change voltage This indicates that the average insulation resistance has significantly decreased at that voltage. If all... If none of them exceed the threshold, then Marked as no mutation.

[0036] Similarly, for the coefficient of variation sequence Perform step-by-step analysis, checking each step along the voltage increase direction, when... First time exceeding the preset stability threshold (e.g., 0.08) will correspond to the voltage. This is denoted as the stability change voltage. This indicates that the stability of repeated measurements deteriorated significantly at this voltage.

[0037] The above mutation threshold and stability threshold The selection criteria are as follows: for a well-installed cold shrink termination, the relative decrease in the average insulation resistance between adjacent voltage levels typically does not exceed [a certain value]. ,Will Setting it to 0.15 can eliminate fluctuations in normal voltage drops and capture abnormal sudden drops; similarly, the coefficient of variation for good insulation in repeated measurements at a single voltage level is typically below 0.03. Setting it to 0.08 balances detection sensitivity and false alarm rate. In actual engineering, the above threshold can be calibrated and optimized based on historical statistical data of the same batch of terminal heads.

[0038] When the decrease between all adjacent voltage levels in the mean sequence does not exceed When the mean voltage change occurs, it indicates that no significant decrease in insulation resistance was observed across the entire test voltage range. Assign the value as the highest test voltage This signifies that the terminal head has a high voltage withstand capability. Similarly, when the coefficient of variation for all voltage levels does not exceed [a certain value], [the following applies]. When the stability change voltage is reached... Assigned value This assignment strategy ensures that the installation quality feature vector has a complete four-dimensional structure in all measurement scenarios, eliminating the need for branching in subsequent classification logic.

[0039] Step S04: Calculate the Pearson correlation coefficient and construct the installation quality feature vector. This step aims to quantify the statistical association between the mean decay trend and changes in repeated measures stability. The mean decay sequence obtained in the previous step is then used... ( ) and coefficient of variation sequence ( Taking the first N-1 values ​​to maintain consistent sequence lengths, we use these as two paired data sets and calculate their Pearson correlation coefficients.

[0040] in The mean of the rate of decline sequence, This represents the mean of the coefficient of variation series. Pearson correlation coefficient. The range of values ​​is .when When the absolute value of the value approaches 1, it indicates that the mean decay and increased volatility of insulation resistance are highly synchronized in the voltage dimension. This correlation physically means that the defect causing the resistance reduction is also the root cause of the measurement volatility; both degradation characteristics are driven by the same defect mechanism. Positive correlation ( A value close to 1) indicates a voltage level with a large mean drop and a high coefficient of variation, corresponding to a typical air gap discharge mechanism; when the absolute value is small (e.g., below 0.3), it indicates that the two lack a regular correlation in the voltage dimension, and the insulation degradation may be caused by uniformity deterioration rather than partial discharge.

[0041] Finally, the following four indicators will be used to construct the installation quality feature vector:

[0042] This eigenvector characterizes the installation quality status of cold-shrink cable terminations from four complementary dimensions: the mean abrupt change voltage reflects the abrupt change in the absolute level of insulation resistance; the stability abrupt change voltage reflects the abrupt change in repeated measurement fluctuations; the Pearson correlation coefficient reflects the degree of coupling between the two degradation modes; and the corrected mean at the highest voltage level reflects the insulation's ability to withstand extreme conditions. The four components are clearly physically independent—the first two are positional scalars in the voltage domain, the third is a statistical correlation strength scalar, and the fourth is an amplitude scalar in the resistance domain—there is no mathematical linear dependence between them. Therefore, the eigenvector can effectively distinguish different installation quality states in a low-dimensional space.

[0043] In actual calculations, if no abrupt changes are detected in either the mean sequence or the coefficient of variation sequence, and All ,at this time The absolute value is usually small, indicating a uniform and stable insulation state with no obvious defect characteristics. If only the coefficient of variation changes abruptly while the mean does not, then... < = This points to the characteristic patterns of early interface air gaps. This combinational logic enables the feature vectors to have clear distinguishing capabilities under different defect severity levels.

[0044] Step S05: Determine the defect type based on the installation quality feature vector and calculate the comprehensive installation quality assessment index. like Figure 3 As shown, based on the values ​​of each component of the installation quality feature vector and their combination relationships, the defect type is determined according to the following rules. The core criterion for defect classification consists of two sets of parameters: the first set is... and Size relationship and Does it exceed the preset relevant threshold? (e.g., 0.7), used to identify air gap defects; the second group is the coefficient of variation. Are all below the preset stability threshold? and corrected mean Is it below the preset qualified threshold? This is used to identify defects in the insulation layer treatment. The two sets of criteria are inherently related in terms of the coefficient of variation: air gap defects inevitably lead to an increase in the coefficient of variation at some voltage levels (exceeding...). The insulation layer treatment defect requires all coefficients of variation to be lower than 100%. This mutual exclusion condition prevents the two types of defects from being triggered independently and simultaneously. However, physically, the same terminal head may simultaneously have both air gaps and poor insulation layer treatment. In this case, the increased coefficient of variation caused by the air gap can mask the independent determination of the insulation layer treatment defect, requiring composite identification through superimposed correction mean conditions. Furthermore, when the coefficient of variation increases but there is a lack of statistical coupling between the decrease and fluctuation, the existing air gap rules cannot match, necessitating the addition of an atypical defect category to eliminate the judgment gap. The specific rules are as follows: 1. Determination of interfacial air gap defects When stability change voltage Less than the mean change voltage And Pearson correlation coefficient Exceed At that time, it was determined to be an interface air gap defect. The physical meaning of this mode is: there is a residual air gap between the cold shrink tubing and the cable insulation layer, and intermittent partial discharge can occur in the air gap at a low voltage, causing the stability of repeated measurements to deteriorate first. Lower); while a significant decrease in the average insulation resistance requires a higher voltage to trigger ( (Higher) Exceed This indicates that voltage levels with large mean drops also have high coefficients of variation, and both degradation characteristics are driven by the same air gap discharge mechanism. The abrupt change in the coefficient of variation preceding the abrupt change in the mean is the core feature of this defect.

[0045] 2. Determination of interfacial air gap defects under high pressure activation When stability change voltage Not less than the mean change voltage ,and Exceed At that time, it was determined to be a high-voltage activated interface air gap defect. In this type of defect, the air gap size is small or located in a region with weak electric field strength, and it is only activated to produce discharge under higher voltage. Therefore, the mean abrupt change and the stability abrupt change occur almost simultaneously or the mean abrupt change occurs first. ). A higher value indicates that the two degradation modes are still highly coupled, and the defect mechanism is the same as in Rule 1, the only difference being the higher activation voltage of the air gap. Rules 1 and 2... and The size relationship is complementary (corresponding to respectively) and Together they covered All scenarios when the threshold is exceeded.

[0046] 3. Determination of composite defects in the treatment of air gap-associated insulation layer When rule 1 or rule 2 has been triggered (i.e., an air gap defect has been identified), and the corrected average value of the highest voltage level... Below the preset qualified threshold When these defects are combined, the defect is classified as a composite defect of air gap and insulation layer treatment. The reason for this rule is that when air gap defects and insulation layer treatment defects coexist, the increased coefficient of variation caused by the air gap makes the independent determination criteria for insulation layer treatment defects (all) more difficult to apply. The previous conclusion is no longer valid; the insulation treatment issue was masked by the air gap symptoms. To avoid omissions, additional checks should be performed on top of the air gap assessment. Is it lower than This parameter reflects the absolute withstand capability of the insulation under the highest voltage. If it is significantly lower, it indicates that in addition to the air gap, there is also a problem of reduced overall insulation resistance. This rule is a supplement to, rather than a replacement for, rules 1 and 2. The final output result simultaneously indicates both the air gap type and the poor insulation treatment.

[0047] 4. Determination of Insulation Layer Treatment Defects When the coefficient of variation of all voltage levels All below ,and Below When this occurs, it is determined to be an insulation layer treatment defect. This pattern manifests as a low absolute value of insulation resistance but good stability in repeated measurements. The reason is insufficient cleaning of the cable insulation surface or incomplete peeling of the semi-conductive layer, leading to a uniform decrease in the overall insulation resistance, but without the presence of localized air gaps that cause intermittent discharge. The triggering condition for this rule is "all". "Triggering conditions for rules 1 and 2" Exceed (Implicit part) The "increase" rule is mutually exclusive in the coefficient of variation dimension, so this rule is only triggered in independent scenarios where there are no air gap defects.

[0048] 5. Determination of Atypical Defects When there is a coefficient of variation at at least one voltage level Not less than ,and Not exceeding At that time, it was determined to be an atypical defect. Physically, this situation corresponds to a state where "repeated measurements at certain voltage levels show fluctuations, but there is a lack of statistical coupling between the fluctuations and the average value decrease," which may be caused by multiple factors such as external electromagnetic interference, poor wiring contact, or the superposition of various minor defects. Because... The level is low, and neither rules 1 nor 2 are triggered; also because of some The condition is elevated, and Rule 4 is also not satisfied. The atypical defect determination fills the above logical gap. It is recommended that the output be marked "Atypical defect, manual review recommended" so that technical personnel can make further diagnoses based on the on-site situation.

[0049] 6. Judgment of no obvious defects When the coefficient of variation of all voltage levels All below The decrease in the mean sequence at each level did not exceed ,and Not less than At this point, the installation quality is deemed acceptable, with no obvious defects. and All equal to , Typically, the values ​​are small, and all components of the feature vector are within the normal range.

[0050] The above six rules constitute a complete judgment system, which can be summarized into the following logical structure: First, based on Does it exceed For a first-level branch—if it exceeds the limit, it enters the air gap category judgment (rule 1 or 2) and is superimposed. Determine if a compound defect exists (Rule 3); if not, check if all coefficients of variation are below a certain threshold. For secondary branches—if all are lower than, then according to Distinguish between insulation layer treatment defects (Rule 4) and qualified ones (Rule 6). If there is an increase, it is classified as an atypical defect (Rule 5). Among them, each branch has no overlap or omission in the parameter space.

[0051] After determining the defect type, calculate the comprehensive installation quality assessment index. This index is based on a weighted sum of the deviations of each component of the eigenvector from a preset benchmark:

[0052] in This is a normalized function of the sudden voltage with respect to the rated voltage, and its specific form is: , To test the upper limit of voltage multiplier (in this embodiment) =5), the higher the sudden voltage, the larger the normalized value, indicating that the insulation remains stable at higher voltages; To correct the mean with respect to a preset pass threshold, the normalization function is defined as follows: When the corrected mean reaches or exceeds the qualified threshold, the normalization value is 1; when it is below the threshold, it is reduced proportionally.

[0053] , , , Preset weighting coefficients and satisfying The principle of weight allocation is: mean change voltage. and stability change voltage The critical voltage that directly reflects defect activation has a significant weight in evaluating installation quality; the Pearson correlation coefficient... As an auxiliary indicator for defect type identification, its weight is appropriate; the highest voltage level is used to correct the mean. Reflecting insulation margin under extreme conditions, it has a relatively low weight. In engineering practice, a typical set of weights is set as follows: =0.30、 =0.30、 =0.20、 =0.20. A higher value indicates better installation quality. If the quality falls below a preset quality threshold (e.g., 0.6), it indicates that the installation quality is substandard and the installation process needs to be checked or reworked.

[0054] In summary, The index compresses the multidimensional information of the feature vector into a single value, making it easier for engineers to quickly determine whether the installation is up to standard. The results of each component of the feature vector and the defect type assessment provide detailed diagnostic information to guide targeted rework measures. For example, when the diagnosis is an interface air gap defect, installers should focus on checking the interface adhesion between the cold shrink tubing and the cable insulation layer, and if necessary, clean the interface again and reinstall the cold shrink tubing. When the diagnosis is an insulation layer treatment defect, the surface treatment process of the cable insulation layer should be checked, with particular attention to the peeling quality of the semiconductive layer and the cleanliness of the insulation surface.

[0055] In practical engineering applications, the entire testing process can be completed independently by one person on-site. The testing equipment only requires a digital high-voltage insulation resistance tester and a temperature sensor, both of which are conventional power testing equipment. =4、 Taking voltage level 5 as an example, each voltage level undergoes 5 cycles, with each cycle lasting approximately 3 minutes (60 seconds of charging + 120 seconds of discharging). For all 4 voltage levels, this takes approximately 60 minutes (including cycle intervals). Adding data processing time of about 5 minutes, the overall testing time is less than 70 minutes. While this increases testing time compared to traditional single-measurement methods, it significantly improves the amount of information obtained, effectively identifying hidden installation defects before the terminal head is put into operation, thus avoiding power outages and safety risks caused by later malfunctions. The final output includes the defect type and comprehensive evaluation results, providing clear quality feedback to installers.

Claims

1. A method for inspecting the installation quality of cold-shrink cable terminations based on insulation resistance stability, characterized in that, Includes the following steps: Step S01: Collect the ambient temperature at the installation site of the cold-shrink cable termination. Apply sequentially between the conductor and the grounding shield of the cold-shrink cable termination. Increasing DC test voltage At each voltage level, apply voltage for a preset charging time until the insulation resistance reading stabilizes and record it. Then, disconnect the voltage and wait for the insulation to recover for a preset discharge time to complete one measurement cycle. Repeat this process at each voltage level. Each measurement cycle records all steady-state insulation resistance values, forming an insulation resistance repeatability measurement matrix. , , ; Step S02, based on ambient temperature With preset reference temperature The difference is used to correct the temperature of all elements in the repeated insulation resistance measurement matrix using a preset temperature correction coefficient, resulting in a corrected insulation resistance matrix. ; Step S03: Calculate the mean value of the corrected insulation resistance matrix for each voltage level. and coefficient of variation The mean sequence is differentially analyzed at each level to identify the voltage at which the voltage drop between adjacent voltage levels first exceeds a preset abrupt change threshold, which is denoted as the mean abrupt change voltage. The coefficient of variation sequence is analyzed step by step to identify the voltage at which the coefficient of variation first exceeds a preset stability threshold, which is denoted as the stability mutation voltage. ; Step S04: Calculate the Pearson correlation coefficient between the mean-decreasing sequence and the coefficient of variation sequence. , the mean sudden voltage Stability sudden voltage Pearson correlation coefficient and the corrected mean of the highest voltage level This constitutes the installation quality feature vector; Step S05: Determine the type of installation defect based on the combination pattern of each component of the installation quality feature vector, calculate the comprehensive installation quality evaluation index, and output the defect type and comprehensive evaluation result.

2. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S01, the preset charging time is not less than 60 seconds, and the preset discharging time is not less than the preset charging time; The interval between two adjacent measurement cycles shall not be less than the preset discharge duration.

3. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S02, the temperature correction adopts an exponential correction model, and the correction formula is as follows: ,in For each The insulation resistance correction factor corresponding to temperature changes.

4. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S05, the defect type determination includes the following rules: when the stability sudden voltage... Less than the mean sudden voltage And Pearson correlation coefficient When the voltage exceeds a preset threshold, it is determined to be an interface air gap defect; when the stability change voltage... Not less than the mean change voltage And Pearson correlation coefficient When the value exceeds a preset threshold, it is determined to be a high-voltage activated interface air gap defect; when it is determined to be an interface air gap defect or a high-voltage activated interface air gap defect and the highest voltage level is corrected average value When the values ​​are below the preset acceptable threshold, the defects are superimposed and judged as composite defects in the air gap and associated insulation layer; when the coefficient of variation for all voltage levels is... All are below the preset stability threshold and the corrected average value of the highest voltage level If the value is below the preset acceptable threshold, it is determined to be a defect in the insulation layer treatment; when there is a coefficient of variation of at least one voltage level... Not lower than the preset stability threshold and Pearson correlation coefficient If the threshold value is not exceeded, it is determined to be an atypical defect.

5. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 4, characterized in that: In step S05, the comprehensive installation quality assessment index The calculation formula is ,in This is a normalized function of the sudden voltage with respect to the rated voltage. To correct the normalization function of the mean with respect to the preset qualified threshold, , , , Preset weighting coefficients and satisfying .

6. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S01, the The lowest voltage of an incremental DC test voltage Not less than 0.5 times the rated voltage of the cable, maximum voltage The voltage increment between adjacent voltage levels shall not exceed five times the rated voltage of the cable, and the voltage increment shall be set in a geometric or arithmetic manner.

7. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S03, the coefficient of variation The calculation formula is ,in For the first voltage level The standard deviation of the corrected insulation resistance value; The gradual decrease in the mean is expressed as a relative decrease rate. Perform the calculation.

8. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 3, characterized in that: In step S02, the preset reference temperature The temperature is 20°C. The correction factor is 10°C. The value range is from 1.3 to 1.

5.

9. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S04, when the decrease between all adjacent voltage levels in the mean sequence does not exceed a preset abrupt change threshold, the mean abrupt change voltage... Assign the value as the highest test voltage When the values ​​of all voltage levels in the coefficient of variation sequence do not exceed the preset stability threshold, the stability mutation voltage is... Assign the value as the highest test voltage .

10. The method for inspecting the installation quality of cold-shrink cable terminations according to claim 1, characterized in that: In step S01, for each voltage level After each measurement cycle is completed, outliers that deviate from the temporary arithmetic mean of that voltage level by more than 3 times the temporary standard deviation are removed, and additional measurement cycles are performed to maintain the voltage level. One valid measurement value.

Images