A study on the effect of temperature rise on the effectiveness of detecting typical defects in rubber-impregnated paper sleeves

Epoxy rubber-impregnated paper casing has the advantages of stable electrical performance, explosion-proof function, small volume, lightweight, etc., and has been widely used in transformer casing and through-wall casing. However, its manufacturing process is complex, which makes it easy to produce defects in the production process, and it is difficult to detect certain defects in the test at room temperature. To study the partial discharge and dielectric loss characteristics of typical defects of rubber-impregnated paper bushings after temperature rise, we designed and prepared defective bushings with moisture, metal particles, and core cracks, and measured the partial discharge pattern and dielectric loss factor of the bushings after they were heated to a certain temperature. The results show that the core crack defective casing discharge increases more after heating, followed by moisture defective casing, and metal particles defective casing has the smallest impact, indicating that for the temperature rise test in the thermal state of the core crack, moisture defective casing detection effect is better. The results of the study can provide certain references and guidance for the factory inspection and troubleshooting of rubber-impregnated paper casing.


Introduction
Gum-impregnated paper bushings are a kind of electrical equipment used for transformers and throughwall bushings, whose main insulation is made of multiple layers of insulating paper and aluminum foil alternately wound on a conduit and then impregnated with resin and cured in a vacuum environment [1] .The gum-impregnated paper casing does not have problems such as oil leakage and gas production and has a high heat-resistant insulation grade, but has limited current-carrying capacity and overload capacity [2][3] .The rubber-impregnated paper casing will not cause explosion failure, because it does not regard oil as a combustible material.There is no gas as an explosive medium and rubber-impregnated paper casing explosion-related cases do not occur.For the current of 1, 600 A to 3, 150 A of converter transformer network side current, rubber-impregnated paper casing also has excellent applicability and has carried out relevant research and pilot applications [4] .Rubber-impregnated paper casing is gradually becoming a viable alternative to oil-paper casing programs.
Due to its complex manufacturing process, the rubber-impregnated paper casing is prone to defects in the production process, such as metal particles, air gaps, and moisture, which can lead to partial discharges, thus affecting the insulating performance of the casing and the safety of operation [5] .The factory test of rubber-impregnated paper casing mainly refers to the test items stipulated in standards such as IEC 60137-2017 [6] , including the frequency dry withstand voltage test, measurement of dielectric loss angle tangent and capacitance value, and insulation resistance measurement.However, the casing will produce high temperatures under the influence of operating working voltage, current and frequency, etc.The high temperature will lead to insulation changes inside the casing and reduce its dielectric strength and partial discharge margin.At the same time, the high temperature will also change the electric field distribution inside the casing, making the electric field strength in some areas increase, which will trigger or aggravate the partial discharge.Therefore, it is difficult to detect certain defects under normal temperature conditions.
Currently, relevant studies have been carried out on the changing law of insulation performance of rubber-impregnated paper (RIP) casing under different temperature conditions [7] , initially investigating the influence law of temperature on the IF dielectric loss of UHV RIP casing [8] , which tested the insulation characteristic parameters such as insulation resistance, IF dielectric loss, IF capacitance, and other insulation characteristics of RIP casing under different temperatures.The theoretical model of the temperature distribution within the adiabatic body and the calculation methods are established, and the allowable load of RIP casing is evaluated based on the maximum thermal voltage [9][10] .However, these studies seldom involve the study of the influence law of temperature increase on partial discharge, dielectric loss, and electric capacity of the typical defective rubber-impregnated paper casing, and the experimental validation of the detection effectiveness of typical defects at room temperature and after temperature increase is insufficient.
To study the typical defects within the rubber-impregnated paper casing in the temperature rise on the partial discharge, dielectric loss, and capacitance of the influence of the law, the author produced several typical defects casing samples to carry out the temperature rise test and temperature stabilization to carry out the dielectric loss, capacitance, and partial discharge experiments.We explore the temperature rise on the rubber-impregnated paper casing typical defects in the detection mode for the rubber-impregnated paper casing factory inspection and fault diagnosis to provide certain references and guidance.

Test article production
According to the statistics, among the fault cases caused by insulation discharge in the gum-impregnated paper casing, cracking of capacitor core, metal particles, and moisture are the most important types of defects.Three types of defective models of rubber-impregnated paper casing with a rated voltage of 72.5 kV and rated current of 630 A were designed and produced.
To analyze and compare the change in casing insulation performance under various defective (faulty) conditions, the faulty casing model adopts the same design parameters, and different conditions are simulated during or after the completion of casing manufacturing to customize the faulty casing with different defects.The types of defects include microcracks, metal particles, and moisture, in addition to the production of a normal casing as a reference for the test.The casing model specimen information is shown in Table 1.

Model 3-core crack defective casing.
Between the first and second layers of capacitive screens, near the position of the end screen tap of the casing, some grooves are carved along the axial direction, and this area is purely epoxy after casting.The production schematic is shown in Figure 3.

Test platform and test method
The temperature rise test of the casing is carried out by using a step-up transformer, increasing the current step by step to 1.2 times the rated current.Platinum resistance temperature sensors are installed on the surface of the casing and guide rods, and the temperature of the guide rods is controlled to be within 120℃ by applying the current.The temperature is stabilized by keeping the steady state condition for 1 h, and then the experimental study of dielectric loss, capacitance, and partial discharge is carried out.The partial discharge test is carried out on the casing by the high-voltage test power supply, which is shown in Figure 5 below.The partial discharge test circuit is shown in Figure 5.

Partial discharge test results for casing with metal particle defects
In the room temperature state before the temperature rise test, JFD-4000 and Techimp partial discharge detector are used to carry out the partial discharge test on the metal particles' defective scaled casing, obtain the partial discharge test data, and compare and analyze with the test data in the hot state after the temperature rise test.Before the partial discharge test, the background noise of the test environment was collected after closing the gate, and the partial discharge background level was 7.9 PC, as shown in Figure 6.Partial discharge tests were carried out on the metal particle defective casing in the cold state before the temperature rise test and in the hot state after the temperature rise test, and the test results are shown in Figure 7.In Figure 7, (a) and (b) show the results of partial discharge detection before and after the temperature rise of the scaled casing for metal particle defects using JFD-4000 partial discharge detector; (c) and (d) show the results of partial discharge detection before and after the temperature rise of the scaled casing for metal particle defects using Techimp partial discharge detector, respectively.A voltage of 55 kV was applied to the metal particle defective casing at cold room temperature prior to the ramp-up test, with a maximum discharge of approximately 380 PC.After the warming test, the partial discharge test in the hot state was carried out again, the maximum apparent discharge was also about 380 PC, and the discharge characteristics were unchanged.Hot and cold environments have essentially no effect on metal particle defect casing partial discharges.
As shown in Figure 8, we continue to apply the voltage of 65 kV to the defective casing of metal particles.The defect excites a maximum apparent discharge of 955 PC, and elevating the applied voltage has a large effect on the partial discharge of metal particles, but it is not sensitive to temperature.

Partial discharge test results of defective casing for core crack defective casing
In the room temperature state before the temperature rise test, JFD-4000 and Techimp partial discharge test equipment are used to collect data to carry out the partial discharge test of the core crack defective casing, obtain the partial discharge test data, and compare and analyze with the test data in the hot state after the temperature rise test.Before the partial discharge test, the background noise of the test environment was collected after closing the gate, and the partial discharge background level was 12.4 PC, as shown in Figure 9. Partial discharge tests were carried out on the core crack defective casing in the cold state before the temperature rise test and in the hot state after the temperature rise test, and the test results are shown in Figure 10.In Figure 10, (a) and (b) show the results of partial discharge detection before and after temperature rise of core crack defect scaled casing by JFD-4000 partial discharge detector; (c) and (d) show the results of partial discharge detection before and after temperature rise of core crack defect scaled casing by Techimp partial discharge detector, respectively.A voltage of 55 kV was applied to the core crack defective casing at cold room temperature prior to the ramp-up test, with a maximum discharge of approximately 285 PC.After the warming test, the partial discharge test in the hot state was carried out again, and the maximum apparent discharge was about 600 PC, with a very large increase in the discharge intensity, and the discharge characteristic spectrogram showed an obvious development trend.Core crack defects are more likely to be excited by defects in the hot state after the temperature rise test.

Partial discharge test results for casing with defective core dampness
In the room temperature state before the temperature rise test, JFD-4000 and Techimp partial discharge test equipment are used to collect data and carry out the partial discharge test of the casing with moisture defects in the core, obtain the partial discharge test data, and analyze them in comparison with the test data in the hot state after the temperature rise test.Before the partial discharge test, the background noise of the test environment after closing the gate is collected, and the partial discharge background level is 18.5 PC, as shown in Figure 11.The partial discharge test was carried out on the core moisture defective casing in the cold state before the temperature rise test and in the hot state after the temperature rise test, and the test results are shown in Figure 12.In Figure 12, (a) and (b) show the results of partial discharge detection of core moisture defect scaled casing before and after temperature rise by JFD-4000 partial discharge detector; (c) and (d) show the results of partial discharge detection of core moisture defect scaled casing before and after temperature rise by Techimp partial discharge detector, respectively.A voltage of 55 kV was applied to the core moisture defective casing at cold room temperature prior to the ramp-up test, with a maximum discharge of approximately 95 PC.After the warming test, the partial discharge test in the hot state was carried out again and the apparent discharge reached 190 PC.Core moisture defects changed in the type of defects excited in the thermal state after the temperature rise test.We analyze the reason for the moisture casing in the thermal state after the temperature rise test, and the existence of a water vapor diffusion state may form bubbles and increase defective weak links.

Dielectric loss and capacitance test results for defective casing
AI-6000 dielectric loss tester was utilized to conduct dielectric loss test and capacitance test on typical defective shrinkage casing in a cold state at room temperature before a heating test and in a hot state after heating test, and the test results are shown in Table 2 below.From the table, the dielectric loss and capacitance of the three kinds of defective casing increase before and after the warming test, in which the core crack defective dielectric loss increases the most, with an increase of 83.7%.This is followed by an increase of 37% in dielectric loss magnitude for metal particle defects, and temperature also has some effect on all three defective capacitances, affecting the magnitude increase by about 2%.

Conclusion
This paper focuses on the local discharge and dielectric loss experiments on the casing with metal particles, core cracks, and moisture defects after warming, and mainly draws the following conclusions: (1) Warming test thermal conditions on the partial discharge effect of metal particles defective casing is small, which will not affect the same voltage under the action of the discharge characteristics but will enhance the dielectric loss and capacitance amplitude, especially on the dielectric loss.Further increasing the voltage level from 55 kV to 65 kV, the partial discharge of metal particle defective casing increases more than twice, which shows that the voltage has more influence on the partial discharge of metal particle defective casing.
(2) Warming test thermal conditions on the partial discharge of core crack defective casing has a greater impact, a substantial excitation of the local discharge characteristics, and the same voltage.The warming test before and after the increase in the amount of discharge is larger, and the discharge characteristics of the spectrum show the development trend and the warming test thermal state of a greater impact on the dielectric loss.Through the test, we can see that the warming of the thermal state of the value of the dielectric loss enhances the value of 83%, so the warming test thermal state of the core cracks in the detection of the effect is better.
(3) Warming test thermal conditions on the core moisture defective casing partial discharge has a certain effect on the discharge amplitude increases, while the dielectric loss value also has a positive increase in the impact.The metal particle defects and core crack defects have less impact, and the three defects of capacitance of the three defects show a positive trend of increase, but the increase is not large.

2. 1 . 1 .
Model 1-moisture defective casing model.The sealing cover of the casing end screen tap measurement terminals was left open for 72 hours to create a moisture defect.The fabrication is shown schematically in Figure 1.

Figure 3 .
Figure 3. Preparation of core crack defect casing.

Figure 7 .
Figure 7.The difference in partial discharge characteristics before and after warming of metal particle defective casing at 55kV.

Figure 8 .
Figure 8. Characteristics of partial discharge with voltage increase in the thermal state of metal particle.

Figure 10 .
Figure 10.The difference in partial discharge characteristics before and after warming of core crack defective casing at 55kV.

Figure 12 .
Figure 12.The difference in partial discharge characteristics before and after warming of core moisture defective casing at 55kV.

Table 1 .
List of casing model specimens.
3Casing design or fabrication methods for various defect models are described below.

Table 2 .
Typical defective casing dielectric loss and capacitance test results before and after temperature rise.