Insulation state detection system adaptive to adaptive resonance type high-voltage terminal

Abstract

The invention discloses an insulation state detection system adaptive to a self-adaptive resonance type high-voltage terminal, which belongs to the technical field of terminal equipment insulation detection and specifically comprises an insulation detection platform, a parameter acquisition unit, a multi-parameter parallel analysis unit, an insulation state problem diagnosis unit and a graded early warning unit. According to the method, multiple independent insulation detection target channels are arranged for the to-be-detected resonant high-voltage terminal, characteristic parameter data are acquired under the specific resonant frequency based on the multiple independent insulation detection target channels, the characteristic parameter data acquisition accuracy is improved, multiple characteristic parameters are continuously tracked, and a real-time parameter trend baseline is established; calculating the forward deviation degree value of each parameter based on a real-time parameter trend baseline and a health parameter trend baseline, performing parallel analysis on the multi-parameter forward deviation degree value by combining a multi-parameter forward deviation correlation model to identify the insulation problem type, and performing cooperative judgment through multi-parameter correlation and cross validation. And the accuracy and reliability of insulation problem type identification are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of terminal equipment insulation detection, more particularly to an insulation state detection system suitable for adaptive resonance type high voltage terminals. BACKGROUND

[0002] The adaptive resonance type high voltage terminal generally refers to an intelligent high voltage terminal device used in resonance test systems, cable or GIS tests and other occasions. For the insulation state detection of resonance type high voltage terminals, it is not only a routine preventive test, but also a core link to ensure the safe, reliable and economic operation of the entire high voltage test or power supply system. The insulation detection needs to consider not only the conventional method of general high voltage equipment, but also the particularity of the resonance circuit.

[0003] During the manufacturing and operation of the high voltage terminal, various factors can cause the formation of insulation defects, and the identification and classification are quite complex. According to the search, patent No. CN118050606A discloses a real-time insulation state detection system, which evaluates the insulation state by collecting various electrical parameters (current, voltage, temperature, humidity, pressure) and insulation surface images. The patent collects operating condition parameters, and the changes of these parameters are usually the result or accompanying phenomenon of insulation deterioration, rather than the cause or early direct representation. The accuracy of diagnosis completely depends on the correlation of surface phenomena, rather than the essential analysis of insulation failure mechanism. For example, the system can only sense the temperature rise, but cannot diagnose the root cause of high temperature, which is insulation damp, internal discharge or material aging. This leads to strong early warning lag and high false alarm rate, and cannot detect and locate local insulation defects.

[0004] Therefore, in view of the above problems, the present application provides an insulation state detection system suitable for adaptive resonance type high voltage terminals. SUMMARY

[0005] The present application aims to solve the existing technical problems, and provides an insulation state detection system suitable for adaptive resonance type high voltage terminals compared with the prior art.

[0006] The purpose of the present application can be achieved by the following technical scheme: an insulation state detection system suitable for adaptive resonance type high voltage terminals, comprising an insulation detection platform, a parameter acquisition unit, a multi-parameter parallel analysis unit, an insulation state problem diagnosis unit and a hierarchical early warning unit. The insulation detection platform calibrates the insulation detection target channel for the resonance type high voltage terminal to be tested;

[0007] The parameter acquisition unit is used to acquire the characteristic parameter data based on the insulation detection target channel collected and processed at a specific resonance frequency, including electrical characteristic parameters, partial discharge characteristic parameters and temperature characteristic parameters. The characteristic parameter data is sent to the multi-parameter parallel analysis unit;

[0008] The multi-parameter parallel analysis unit is used to obtain characteristic parameter data and a preset health parameter trend baseline, to establish a corresponding real-time parameter trend baseline according to the real-time obtained characteristic parameter data, to calculate a positive deviation degree value of each parameter based on the real-time parameter trend baseline and the health parameter trend baseline, and to send the positive deviation degree value to the insulation state problem diagnosis unit;

[0009] The insulation state problem diagnosis unit obtains a preset multi-parameter positive deviation correlation model, performs parallel analysis on the multi-parameter positive deviation degree value in combination with the multi-parameter positive deviation correlation model, identifies an insulation problem type, and sends the insulation problem type to the hierarchical early warning unit. The hierarchical early warning unit generates an insulation damp signal, an insulation aging signal and an insulation discharge signal according to the insulation problem type.

[0010] Further, the insulation detection target channel includes a voltage reference measurement channel, a ground current measurement channel, a partial discharge measurement channel and a sensing physical measurement channel.

[0011] Further, the process of obtaining the characteristic parameter data includes:

[0012] The electrical characteristic parameters are obtained based on the voltage reference measurement channel and the ground current measurement channel. The electrical characteristic parameters specifically include a dielectric loss value and a capacitance value.

[0013] The partial discharge characteristic parameters are obtained based on the partial discharge measurement channel. The partial discharge characteristic parameters specifically include a discharge increase value, wherein the discharge increase value is the product value obtained by data normalization processing of the part of the partial discharge amount exceeding a preset partial discharge amount threshold and the discharge time duration during which the partial discharge amount continuously exceeds the preset partial discharge amount threshold.

[0014] The temperature characteristic parameters are obtained based on the sensing physical measurement channel. The temperature characteristic parameters specifically include a temperature gradient value, and the temperature gradient value is a temperature rise value of a real-time temperature exceeding a preset standard temperature.

[0015] Further, the process of establishing the health parameter trend baseline includes:

[0016] The multi-parameter parallel analysis unit collects original health characteristic parameter data as a health data set based on the insulation detection target channel, establishes a health parameter trend baseline for a single health data set, and stores the health parameter trend baseline in the insulation detection platform. The health parameter trend baseline specifically includes a health dielectric loss value trend baseline, a health capacitance value trend baseline, a health discharge increase value trend baseline and a health temperature gradient value trend baseline.

[0017] Further, the process of calculating the positive deviation degree value of each parameter based on the real-time parameter trend baseline and the health parameter trend baseline includes:

[0018] The real-time acquired characteristic parameter data is taken as a real-time data set, a real-time parameter trend baseline is established for a single real-time data set, and the real-time parameter trend baseline specifically includes a real-time dielectric loss value trend baseline, a real-time capacitance trend baseline, a real-time discharge increment value trend baseline and a real-time temperature gradient value trend baseline;

[0019] The real-time dielectric loss value trend baseline and the healthy dielectric loss value trend baseline are plotted in the same rectangular coordinate system, vertical auxiliary lines are drawn from two edge points of the real-time dielectric loss value trend baseline to the healthy dielectric loss value trend baseline, the coverage area of the dielectric loss value trend baseline above the healthy dielectric loss value trend baseline is calculated, and a dielectric loss positive deviation degree value is marked;

[0020] Similarly, a capacitance positive deviation degree value, a discharge increment positive deviation degree value and a temperature gradient positive deviation degree value are obtained.

[0021] Further, the process of parallel analysis of the multi-parameter positive deviation degree values to identify the insulation problem type includes:

[0022] The insulation state problem diagnosis unit inputs the obtained dielectric loss positive deviation degree value, the capacitance positive deviation degree value, the discharge increment positive deviation degree value and the temperature gradient positive deviation degree value into a multi-parameter positive deviation correlation model, respectively compares them with corresponding preset upper limit thresholds, judges whether each parameter positive deviation degree value exceeds the deviation upper limit, when the deviation upper limit is exceeded, the characteristic parameter is marked as a significant deviation parameter, otherwise the characteristic parameter is marked as a slight deviation parameter, and the insulation problem type is identified according to the deviation parameter judgment result.

[0023] Further, when the dielectric loss value and the capacitance are marked as significant deviation parameters, and the discharge increment value and the temperature gradient value are marked as slight deviation parameters, the insulation problem type is determined as a moisture problem type;

[0024] When the dielectric loss value is marked as a significant deviation parameter, and the capacitance, the discharge increment value and the temperature gradient value are marked as slight deviation parameters, the insulation problem type is determined as an aging problem type;

[0025] When the discharge increment value and the temperature gradient value are marked as significant deviation parameters, and the dielectric loss value and the capacitance are marked as slight deviation parameters, the insulation problem type is determined as a partial discharge problem type.

[0026] Compared with the prior art, the advantages of the present application are:

[0027] 1. The scheme is for the to-be-measured resonant high-voltage terminal, multiple independent insulation detection target channels are set, feature parameter data is collected at a specific resonant frequency based on the multiple independent insulation detection target channels, multiple feature parameters are continuously tracked and a real-time parameter trend baseline is established, the positive deviation degree value of each parameter is calculated based on the real-time parameter trend baseline and the healthy parameter trend baseline, the multiple parameter positive deviation degree values are analyzed in parallel to identify the insulation problem type in combination with the multiple parameter positive deviation correlation model, and the accuracy and reliability of the insulation problem type identification are improved through the collaborative judgment of the multiple parameter correlation and cross verification, breaking the limitations of a single parameter.

[0028] 2. The insulation detection target channel of the to-be-measured resonant high-voltage terminal is calibrated, and the calibration target is to synchronously and accurately obtain original signals from different sensors, improve the accuracy of feature parameter data acquisition, and the method of multiple parameter correlation analysis with multiple key feature parameters is a correct and advanced technical path for building an intelligent high-voltage equipment insulation state diagnosis system. It upgrades from simple threshold judgment of "single parameter operation" to deep pattern recognition of "multiple parameter cooperation", which can more accurately and reliably reveal the real state of the insulation, and provide strong technical support for the safe and stable operation of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0029] Fig. 1 The system principle block diagram of the present application is shown in the figure;

[0030] Fig. 2 The method flowchart of the present application is shown in the figure. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the accompanying drawings of the embodiments of the present application; obviously, the described embodiments are only part of the embodiments of the present application, but not all the embodiments; based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor are within the scope of protection of the present application.

[0032] Embodiment one: the present application discloses an insulation state detection system suitable for adaptive resonant high-voltage terminal, please refer to Figs. 1-2 , which comprises an insulation detection platform, a parameter acquisition unit, a multiple parameter parallel analysis unit, an insulation state problem diagnosis unit and a hierarchical early warning unit, the insulation detection platform calibrates the insulation detection target channel of the to-be-measured resonant high-voltage terminal, the insulation detection target channel comprises a voltage reference measurement channel, a ground current measurement channel and a partial discharge measurement channel, and a physical measurement channel.

[0033] Among them, the voltage reference measurement channel is calibrated to accurately measure the voltage division ratio and phase shift of the built-in voltage divider; the ground current measurement channel is calibrated to accurately calibrate the transformation ratio and frequency response characteristics of the built-in high-frequency current transformer; the partial discharge measurement channel is calibrated to optimize for high-frequency pulses (such as 1MHz-30MHz) using an ultra-high frequency sensor; although the partial discharge monitoring and dielectric loss / capacitance channels are shared, a dedicated partial discharge measurement channel is set up, which has higher sensitivity and uses a completely independent UHF sensor to detect electromagnetic waves of 300MHz-3GHz, with extremely strong anti-interference capability; the sensor physics measurement channel is calibrated to monitor temperature distribution and gradient using fiber optic thermometry or infrared thermal imagers.

[0034] The calibration measurement channel system ensures that the voltage and current signals acquired from these built-in sensors are accurate in both amplitude and phase.

[0035] The parameter acquisition unit is used to acquire characteristic parameter data based on the insulation detection target channel at a specific resonant frequency and after processing and calculation. The characteristic parameter data specifically includes electrical characteristic parameters, partial discharge characteristic parameters, and temperature characteristic parameters.

[0036] The specific process for obtaining the feature parameter data includes:

[0037] Electrical characteristic parameters are obtained by the coordinated use of voltage reference measurement channel and grounding current measurement channel. The electrical characteristic parameters specifically include dielectric loss value and capacitance.

[0038] Partial discharge characteristic parameters are obtained based on the partial discharge measurement channel. Specifically, the partial discharge characteristic parameter is the discharge increment value. The discharge increment value is the product of the portion of partial discharge exceeding the preset partial discharge threshold and the discharge duration during which the partial discharge exceeds the preset partial discharge threshold, after data normalization.

[0039] Temperature characteristic parameters are obtained based on the physical measurement channel of the sensor. Specifically, the temperature characteristic parameter is the temperature gradient value, which is the temperature rise value when the real-time temperature exceeds the preset standard temperature.

[0040] The specific process for obtaining the dielectric loss value and capacitance is as follows: Based on the voltage reference measurement channel and the grounding current measurement channel, the power frequency fundamental component is extracted from the synchronous voltage signal and grounding current signal. The phase difference between the current and voltage is accurately calculated through fast Fourier transform or sine wave parameter fitting algorithm. The dielectric loss value and capacitance are obtained by parametric calculation through the phase difference. It should be emphasized that environmental data (temperature and humidity data) is needed to normalize the dielectric loss value to eliminate environmental influences.

[0041] The specific acquisition process of the partial discharge quantity is as follows: since the signal captured by the UHF sensor is very weak (microvolt level), it is first amplified by a low-noise, high-bandwidth preamplifier, then filtered by a band-pass filter (such as 300MHz-1.5GHz) to further filter out out-of-band noise, and the amplified and filtered analog signal is sent to a high-speed ADC (analog-to-digital converter) for sampling to complete the identification and extraction of the partial discharge pulse signal and obtain the partial discharge quantity;

[0042] The parameter acquisition unit sends the characteristic parameter data to the multi-parameter parallel analysis unit, and the method of performing multi-parameter correlation analysis on multiple key characteristic parameters is a correct and advanced technical path for constructing an intelligent high-voltage equipment insulation state diagnosis system.

[0043] The multi-parameter parallel analysis unit is used to acquire characteristic parameter data and a preset health parameter trend baseline, the health parameter trend baseline is pre-stored in the insulation detection platform, a corresponding real-time parameter trend baseline is established according to the real-time acquired characteristic parameter data, and a positive deviation degree value of each parameter is calculated based on the real-time parameter trend baseline and the health parameter trend baseline;

[0044] The establishment process of the health parameter trend baseline includes:

[0045] The multi-parameter parallel analysis unit collects original health characteristic parameter data as a health data set based on the insulation detection target channel, establishes a health parameter trend baseline for a single health data set, and stores the health parameter trend baseline in the insulation detection platform. The health parameter trend baseline specifically includes a health dielectric loss value trend baseline, a health capacitance trend baseline, a health discharge increase trend baseline, and a health temperature gradient value trend baseline.

[0046] The process of calculating the positive deviation degree value of each parameter based on the real-time parameter trend baseline and the health parameter trend baseline includes:

[0047] The real-time acquired characteristic parameter data is taken as a real-time data set, a real-time parameter trend baseline is established for a single real-time data set, and the real-time parameter trend baseline specifically includes a real-time dielectric loss value trend baseline, a real-time capacitance trend baseline, a real-time discharge increase trend baseline, and a real-time temperature gradient value trend baseline.

[0048] The real-time dielectric loss value trend baseline and the health dielectric loss value trend baseline are plotted in the same rectangular coordinate system, a vertical auxiliary line is drawn from the two edge points of the real-time dielectric loss value trend baseline to the health dielectric loss value trend baseline, the coverage area of the dielectric loss value trend baseline above the health dielectric loss value trend baseline is calculated, and the positive deviation degree value of the dielectric loss is marked.

[0049] Similarly, the capacitance positive deviation degree value, the discharge increase positive deviation degree value, and the temperature gradient positive deviation degree value are obtained.

[0050] The positive deviation values ​​of dielectric loss, capacitance, discharge increase, and temperature gradient are all sent to the insulation condition problem diagnosis unit.

[0051] The insulation condition problem diagnosis unit acquires a preset multi-parameter positive deviation correlation model. The multi-parameter positive deviation correlation model is pre-stored in the insulation detection platform. The multi-parameter positive deviation correlation model is a machine learning model trained based on the multi-parameter positive deviation degree values ​​under healthy conditions. During the training process, the model continuously learns different parameter positive deviation degree values ​​to accurately assess the insulation condition of the new high-voltage terminal. Combined with the multi-parameter positive deviation correlation model, the multi-parameter positive deviation degree values ​​are analyzed in parallel to identify the insulation problem type.

[0052] The process of identifying insulation problem types through parallel analysis of the positive deviation values ​​of multiple parameters includes:

[0053] The insulation condition problem diagnosis unit acquires the positive deviation values ​​of dielectric loss, capacitance, discharge increase, and temperature gradient. These values ​​are then input into a multi-parameter positive deviation correlation model. Each parameter's positive deviation value is compared with its corresponding preset upper limit threshold to determine whether it exceeds the upper limit. If it does, the parameter is marked as a significantly deviated parameter; otherwise, it is marked as a slightly deviated parameter. The insulation problem type is identified based on the deviation parameter judgment results.

[0054] When the dielectric loss value and capacitance value are marked as significantly deviating from the parameters, and the discharge increase value and temperature gradient value are marked as slightly deviating from the parameters, the insulation problem type is marked as a moisture problem type.

[0055] When the dielectric loss value is marked as significantly deviating from the parameter, and the capacitance, discharge increase value, and temperature gradient value are marked as slightly deviating from the parameter, the insulation problem type is marked as an aging problem type;

[0056] When the discharge increase value and temperature gradient value are marked as significantly deviating from the parameters, and the dielectric loss value and capacitance value are marked as slightly deviating from the parameters, the insulation problem type is marked as a partial discharge problem type;

[0057] Based on the real-time parameter trend baseline and the health parameter trend baseline, the positive deviation value of each parameter is calculated. Combined with the multi-parameter positive deviation correlation model, the positive deviation value of the multi-parameters is analyzed in parallel to identify the type of insulation problem. Using the positive deviation value as an input feature is a very effective method in multi-parameter correlation analysis. This can eliminate the influence of absolute dimensions, focus on relative change patterns, and make collaborative judgments through multi-parameter correlation and cross-validation, breaking the limitations of a single parameter and improving the accuracy and reliability of insulation problem type identification.

[0058] The insulation condition problem diagnosis unit sends various insulation problem types to the graded early warning unit.

[0059] The graded early warning unit generates insulation moisture signal, insulation aging signal, and insulation discharge signal based on the type of insulation problem. Specifically, when a moisture problem is received, an insulation moisture signal is generated; when an aging problem is received, an insulation aging signal is generated; and when a partial discharge problem is received, an insulation discharge signal is generated.

[0060] It should be added that the article involves multiple types of threshold comparison analysis. Thresholds, preset values, preset ranges, etc. are set for result comparison analysis in order to determine good or bad. The magnitude of these values ​​is set and stored based on a combination of large-scale model analysis of sample data and human experience. They can also be appropriately adjusted based on seasonal or common-sense influence conditions.

[0061] In summary, this invention sets up multiple independent insulation detection target channels for the adaptive resonant high-voltage terminal under test. The calibration of the targets is to synchronously and with high fidelity acquire raw signals from different sensors, thereby improving the accuracy of characteristic parameter data acquisition. Based on the multiple independent insulation detection target channels, characteristic parameter data is collected at a specific resonant frequency. Multiple characteristic parameters are continuously tracked and a real-time parameter trend baseline is established. Based on the real-time parameter trend baseline and the healthy parameter trend baseline, the positive deviation value of each parameter is calculated. Combined with the multi-parameter positive deviation correlation model, the positive deviation values ​​of multiple parameters are analyzed in parallel to identify the insulation problem type. Through multi-parameter correlation and cross-validation, collaborative judgment is performed, breaking the limitations of a single parameter and improving the accuracy and reliability of insulation problem type identification.

[0062] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto; any equivalent substitutions or modifications made by those skilled in the art within the technical scope disclosed in the present invention, based on the technical solution and its improved concept, should be covered within the scope of protection of the present invention.

Claims

1. An insulation condition detection system adapted to an adaptive resonant high voltage termination, characterized by: The insulation detection platform comprises an insulation detection platform, a parameter acquisition unit, a multi-parameter parallel analysis unit, an insulation state problem diagnosis unit, and a hierarchical early warning unit. The parameter acquisition unit is used to acquire characteristic parameter data based on the insulation detection target channel under a specific resonance frequency, including electrical characteristic parameters, partial discharge characteristic parameters, and temperature characteristic parameters. The multi-parameter parallel analysis unit is used to acquire characteristic parameter data and a preset health parameter trend baseline, establish a corresponding real-time parameter trend baseline based on real-time acquired characteristic parameter data, calculate a positive deviation degree value of each parameter based on the real-time parameter trend baseline and the health parameter trend baseline, and send it to the insulation state problem diagnosis unit. The insulation state problem diagnosis unit acquires a preset multi-parameter positive deviation correlation model, combines the multi-parameter positive deviation correlation model to perform parallel analysis on the multi-parameter positive deviation degree value, identifies the insulation problem type, and sends each type of insulation problem type to the hierarchical early warning unit.

2. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 1, characterized in that: The insulation detection target channel comprises a voltage reference measurement channel, a ground current measurement channel, a partial discharge measurement channel, and a sensing physical measurement channel.

3. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 2, characterized in that: The acquisition process of the characteristic parameter data comprises: The electrical characteristic parameters are acquired based on the voltage reference measurement channel and the ground current measurement channel, and the electrical characteristic parameters specifically include dielectric loss and capacitance. The partial discharge characteristic parameters are acquired based on the partial discharge measurement channel, and the partial discharge characteristic parameters specifically include a discharge increase value, wherein the discharge increase value is the product of the part of the partial discharge amount exceeding the preset partial discharge amount threshold and the discharge duration of the partial discharge amount continuously exceeding the preset partial discharge amount threshold after data normalization processing. The temperature characteristic parameters are acquired based on the sensing physical measurement channel, and the temperature characteristic parameters specifically include a temperature gradient value, which is the temperature rise value of the real-time temperature exceeding the preset standard temperature.

4. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 3, characterized in that: The establishment process of the health parameter trend baseline comprises: The multi-parameter parallel analysis unit acquires original health characteristic parameter data as a health data set based on the insulation detection target channel, establishes a health parameter trend baseline for each health data set, and stores it in the insulation detection platform.

5. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 4, characterized in that: The process of calculating the positive deviation degree value of each parameter based on the real-time parameter trend baseline and the health parameter trend baseline comprises: The real-time acquired characteristic parameter data is used as a real-time data set, a real-time parameter trend baseline is established for each real-time data set, and the real-time parameter trend baseline specifically includes a real-time dielectric loss trend baseline, a real-time capacitance trend baseline, a real-time discharge increase value trend baseline, and a real-time temperature gradient value trend baseline. The real-time dielectric loss value trend baseline and the healthy dielectric loss value trend baseline are plotted in the same rectangular coordinate system, two edge points of the real-time dielectric loss value trend baseline are vertically assisted to the healthy dielectric loss value trend baseline, the coverage area of the dielectric loss value trend baseline above the healthy dielectric loss value trend baseline is calculated, and the dielectric loss positive deviation degree value is marked; Similarly, the capacitance positive deviation degree value, the discharge increase positive deviation degree value, and the temperature gradient positive deviation degree value are obtained.

6. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 5, characterized in that: The process of parallel analysis of the multi-parameter positive deviation degree value to identify the insulation problem type includes: The insulation state problem diagnosis unit inputs the obtained dielectric loss positive deviation degree value, the capacitance positive deviation degree value, the discharge increase positive deviation degree value, and the temperature gradient positive deviation degree value into the multi-parameter positive deviation correlation model, respectively compares them with the corresponding preset upper limit threshold, judges whether each parameter positive deviation degree value exceeds the deviation upper limit, when it exceeds the deviation upper limit, marks the characteristic parameter as a significant deviation parameter, otherwise marks the characteristic parameter as a slight deviation parameter, and identifies the insulation problem type according to the deviation parameter judgment result.

7. The insulation condition detection system adapted to the adaptive resonance type high voltage terminal according to claim 6, characterized in that: When the dielectric loss value and the capacitance are marked as significant deviation parameters, and the discharge increase value and the temperature gradient value are marked as slight deviation parameters, the insulation problem type is determined as a damp problem type; When the dielectric loss value is marked as a significant deviation parameter, and the capacitance, the discharge increase value, and the temperature gradient value are marked as slight deviation parameters, the insulation problem type is determined as an aging problem type; When the discharge increase value and the temperature gradient value are marked as significant deviation parameters, and the dielectric loss value and the capacitance are marked as slight deviation parameters, the insulation problem type is determined as a partial discharge problem type.

Images