Transformer oil temperature change process dielectric loss prediction method, system, device and medium

Abstract

The application provides a transformer oil temperature change process dielectric loss prediction method, system, equipment and medium, and the method comprises the steps of: obtaining transformer oil samples with different aging degrees through oil-paper insulation accelerated thermal aging test; obtaining oil sample ultrasonic signals of each transformer oil sample according to ultrasonic technology, and obtaining corresponding oil sample ultrasonic pulse factors according to each oil sample ultrasonic signal; obtaining oil sample dielectric loss values of each transformer oil sample according to a dielectric loss tester; establishing a dielectric loss prediction model according to the oil sample ultrasonic pulse factors and the oil sample dielectric loss values; obtaining a to-be-tested ultrasonic pulse factor of the transformer oil to be tested in real time, inputting the to-be-tested ultrasonic pulse factor into the dielectric loss prediction model for dielectric loss prediction, and obtaining a first dielectric loss prediction value. The application establishes a dielectric loss prediction model by taking the ultrasonic signal pulse factor as a characteristic representing the dielectric loss of the transformer oil, realizes real-time online prediction of the dielectric loss, is simple and convenient to operate, and can effectively improve the prediction accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transformer detection, in particular to a transformer oil temperature change process dielectric loss prediction method and system based on ultrasonic signal pulse factors, a computer device and a storage medium. BACKGROUND

[0002] The service life of a transformer depends on the insulation state, and most transformer failures are caused by damage to the insulation system. Transformer oil is an important component of the insulation system in the transformer, and accurate quantitative detection of the state of the transformer oil is of great significance to ensure the safe and stable operation of the transformer, and can effectively improve the service life of the transformer and the reliability of power supply. However, the transformer oil will deteriorate during long-term operation under the influence of external environment and internal physicochemical factors, and in this process, the water content and impurities in the transformer oil will continue to increase, resulting in increased conductivity of the transformer oil and rising dielectric loss factor, ultimately causing a serious decline in the insulation performance of the transformer oil. The process is irreversible, which can significantly reduce the safety performance of the transformer and shorten the service life of the transformer, and in severe cases, can cause transformer equipment to burn and other major accidents, endangering the stability of the power system. Therefore, timely detection of the dielectric loss of the transformer oil to ensure that the dielectric loss factor is controlled within the qualified range is a necessary condition to ensure the safe operation of the transformer and prevent transformer accidents.

[0003] The existing dielectric loss prediction method of transformer oil mainly analyzes through an offline oil dielectric loss automatic tester. The measurement system is complex and the operation is cumbersome, which is not conducive to realizing real-time online monitoring of the transformer oil. Moreover, the influence of temperature factors is not considered in the dielectric loss prediction process, resulting in insufficient accuracy of the dielectric loss prediction, and the application requirement of real-time and accurate monitoring of the transformer oil in actual engineering cannot be effectively met. SUMMARY

[0004] The purpose of the present application is to provide a transformer oil temperature change process dielectric loss prediction method. By taking the ultrasonic signal pulse factor as a characteristic representing the dielectric loss of the transformer oil, a linear model between the pulse factor and the dielectric loss is established to realize simple and reliable prediction of the dielectric loss, and a temperature correction dielectric loss prediction model based on the ultrasonic signal pulse factor is established to further correct the dielectric loss prediction result. The application defects of the existing transformer oil dielectric loss prediction method are effectively solved, which is convenient for realizing real-time online monitoring of the transformer oil, effectively improving the accuracy of the dielectric loss prediction in the real-time change process of the transformer oil temperature, and effectively ensuring the safe and reliable operation of the power transformer, which has very important engineering significance.

[0005] In order to achieve the above-mentioned purpose, it is necessary to provide a transformer oil temperature change process dielectric loss prediction method, system, computer device and storage medium in view of the above technical problems.

[0006] In a first aspect, the embodiments of the present application provide a transformer oil temperature change process dielectric loss prediction method, the method comprising the following steps:

[0007] Through an oil-paper insulation accelerated thermal aging test, transformer oil samples of different aging degrees are obtained;

[0008] According to ultrasonic technology, an oil sample ultrasonic signal of each transformer oil sample is obtained, and a corresponding oil sample ultrasonic pulse factor is obtained according to each oil sample ultrasonic signal;

[0009] According to a dielectric loss tester, an oil sample dielectric loss value of each transformer oil sample is obtained;

[0010] According to the oil sample ultrasonic pulse factor and the oil sample dielectric loss value, a dielectric loss prediction model is established;

[0011] A to-be-detected ultrasonic pulse factor of a to-be-detected transformer oil is obtained in real time, and the to-be-detected ultrasonic pulse factor is input into the dielectric loss prediction model for dielectric loss prediction to obtain a first dielectric loss prediction value.

[0012] Further, the step of obtaining, according to ultrasonic technology, an ultrasonic signal of each transformer oil sample comprises:

[0013] A transformer oil ultrasonic detection platform is constructed in advance; the transformer oil ultrasonic detection platform comprises an ultrasonic signal emitting device, a temperature control device, an oil container, and an ultrasonic signal detection device; the ultrasonic signal emitting device comprises a signal generator, a power amplifier, and a first ultrasonic sensor connected in sequence; the temperature control device comprises a temperature controller, and a temperature sensor and a heating rod connected with the temperature controller; the ultrasonic signal detection device comprises a second ultrasonic sensor, a digital oscilloscope, and a computing terminal connected in sequence; the first ultrasonic sensor, the temperature sensor, the heating rod, and the second ultrasonic sensor are all immersed in transformer oil in the oil container;

[0014] According to the transformer oil ultrasonic detection platform, ultrasonic detection is performed on each transformer oil sample to obtain a corresponding ultrasonic signal.

[0015] Further, the dielectric loss prediction model is represented as:

[0016] Loss D =k1*I f +b1

[0017] Wherein, Loss D represents dielectric loss; I f represents an ultrasonic pulse factor; k1 and b1 represent constants.

[0018] Further, the method further comprises:

[0019] According to the ultrasonic technology, an ultrasonic pulse factor of each transformer oil sample at a plurality of preset temperatures is obtained, and a corresponding pulse factor prediction model is established according to the ultrasonic pulse factor of each transformer oil sample at the plurality of preset temperatures;

[0020] According to the pulse factor prediction model, the dielectric loss prediction model is corrected to obtain a corresponding temperature-corrected dielectric loss prediction model;

[0021] The measured ultrasonic pulse factor is input into the temperature-corrected dielectric loss prediction model to predict the dielectric loss, and a second dielectric loss prediction value is obtained.

[0022] Further, the step of correcting the dielectric loss prediction model according to the pulse factor prediction model to obtain a corresponding temperature-corrected dielectric loss prediction model comprises:

[0023] The real-time oil temperature of the transformer oil to be tested is obtained, and the real-time oil temperature is input into the pulse factor prediction model to estimate the pulse factor, and a pulse factor estimation value is obtained;

[0024] According to the pulse factor estimation value and the corresponding measured ultrasonic pulse factor, a temperature correction factor is obtained; the temperature correction factor is represented as:

[0025]

[0026] Wherein, β represents the temperature correction factor; and respectively represent the measured ultrasonic pulse factor and the pulse factor estimation value;

[0027] According to the temperature correction factor, the dielectric loss prediction model is corrected to obtain the temperature-corrected dielectric loss prediction model.

[0028] Further, the temperature-corrected dielectric loss prediction model is represented as:

[0029]

[0030] Wherein, represents the dielectric loss based on temperature correction; I f represents the ultrasonic pulse factor; β represents the temperature correction factor; k1 and b1 represent constants.

[0031] In a second aspect, an embodiment of the present application provides a transformer oil temperature change process dielectric loss prediction system, the system comprising:

[0032] An oil sample acquisition module is configured to obtain transformer oil samples with different aging degrees through an oil-paper insulation accelerated thermal aging test;

[0033] The oil sample feature extraction module is configured to acquire oil sample ultrasonic signals of each transformer oil sample according to ultrasonic technology, and obtain corresponding oil sample ultrasonic pulse factors according to the oil sample ultrasonic signals.

[0034] The oil sample dielectric loss detection module is configured to obtain oil sample dielectric loss values of each transformer oil sample according to a dielectric loss tester.

[0035] The first model construction module is configured to establish a dielectric loss prediction model according to the oil sample ultrasonic pulse factors and the oil sample dielectric loss values.

[0036] The first dielectric loss prediction module is configured to acquire a to-be-tested ultrasonic pulse factor of transformer oil to be tested in real time, and input the to-be-tested ultrasonic pulse factor into the dielectric loss prediction model to perform dielectric loss prediction, and obtain a first dielectric loss prediction value.

[0037] Further, the system further comprises:

[0038] The second model construction module is configured to acquire ultrasonic pulse factors of each transformer oil sample at a plurality of preset temperatures according to ultrasonic technology, and establish a corresponding pulse factor prediction model according to the ultrasonic pulse factors of each transformer oil sample at the plurality of preset temperatures.

[0039] The prediction model correction module is configured to correct the dielectric loss prediction model according to the pulse factor prediction model, and obtain a corresponding temperature-corrected dielectric loss prediction model.

[0040] The second dielectric loss prediction module is configured to input the to-be-tested ultrasonic pulse factor into the temperature-corrected dielectric loss prediction model to perform dielectric loss prediction, and obtain a second dielectric loss prediction value.

[0041] In a third aspect, an embodiment of the present application further provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0042] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement the steps of the above method.

[0043] The present application provides a method, system, computer equipment, and storage medium for predicting dielectric loss during transformer oil temperature changes. The method achieves the following: obtaining transformer oil samples with different aging degrees through accelerated thermal aging tests of oil-paper insulation; acquiring ultrasonic signals of each transformer oil sample using ultrasonic technology; obtaining corresponding ultrasonic pulse factors based on the ultrasonic signals of each oil sample; obtaining the dielectric loss value of each transformer oil sample using a dielectric loss tester; establishing a dielectric loss prediction model based on the ultrasonic pulse factors and dielectric loss values ​​of the oil samples; inputting the real-time acquired ultrasonic pulse factors of the transformer oil under test into the dielectric loss prediction model to predict dielectric loss, obtaining a first dielectric loss prediction value; and further refining the dielectric loss prediction result by establishing a temperature-corrected dielectric loss prediction model based on the ultrasonic signal pulse factors to obtain a second dielectric loss prediction value. Compared with existing technologies, this method for predicting dielectric losses during transformer oil temperature changes establishes a dielectric loss prediction model by using the ultrasonic signal pulse factor as a characteristic of transformer oil dielectric loss, and further establishes a temperature-corrected dielectric loss prediction model based on the sensitivity of the pulse factor to temperature. This enables real-time online prediction of dielectric losses during transformer oil temperature changes. It is not only simple and convenient to operate, but also effectively improves the accuracy of dielectric loss prediction during real-time changes in transformer oil temperature, thereby effectively ensuring the safe and reliable operation of power transformers, which has significant engineering implications. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of dielectric loss values ​​corresponding to different transformer oil sample types in an embodiment of the present invention;

[0045] Figure 2 Based on Figure 1 A schematic diagram showing the correlation coefficients between the ultrasonic signals of transformer oil samples and the dielectric values;

[0046] Figure 3 This is a flowchart illustrating the method for predicting dielectric losses during transformer oil temperature changes in this embodiment of the invention.

[0047] Figure 4 This is a schematic diagram of the structure of the ultrasonic testing platform for transformer oil in an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram illustrating the application scenario of the ultrasonic testing platform for transformer oil in this embodiment of the invention;

[0049] Figure 6 This is a schematic diagram of the ultrasonic waveform of a transformer oil sample at different temperatures in an embodiment of the present invention;

[0050] Figure 7 This is a schematic diagram of the pulse factor prediction model corresponding to transformer oil sample 1 in this embodiment of the invention;

[0051] Figure 8 is a structural schematic diagram of a transformer oil temperature change process dielectric loss prediction system in an embodiment of the present application;

[0052] Figure 9 is an internal structure diagram of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purposes, technical solutions and beneficial effects of the present application clearer, the present application is further described in detail below in combination with the drawings and embodiments. Obviously, the following described embodiments are only a part of the embodiments of the present application, and are only used to illustrate the present application, but not to limit the scope of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0054] The transformer oil temperature change process dielectric loss prediction method provided by the present application is a transformer oil temperature change process dielectric loss prediction method based on an ultrasonic signal pulse factor, which is realized based on an ultrasonic detection technology. The ultrasonic detection technology is a nondestructive detection with advantages of fast detection, high measurement precision, convenient operation and strong anti-electromagnetic interference ability, which can provide reliable guarantee for the accuracy of the dielectric loss prediction based on the ultrasonic signal in the present application. When the ultrasonic technology is actually used to detect the transformer oil, the relevant statistical features shown in Table 1 such as a signal propagation time C1, an absolute mean value C2, a root mean square value C3, a peak-to-peak value C4, a root amplitude value C5, a standard deviation C6, a skewness C7, a kurtosis C8, a waveform factor C9 and a pulse factor C10 can be extracted based on the collected ultrasonic signal of the transformer oil. In principle, these features are used to describe the change between different oil samples, i.e. can be used for dielectric loss prediction of the transformer oil, but in order to ensure the simplicity and reliability of the prediction method, the present application first analyzes and screens the 10 ultrasonic signal statistical features shown in Table 1, and obtains the statistical feature pulse factor with high correlation with the dielectric loss as the feature representing the dielectric loss of the transformer oil for establishing a dielectric loss prediction model, so as to realize simple and reliable prediction of the dielectric loss.

[0055] Table 1: 10 statistical feature calculation expressions of oil sample ultrasonic signal x(t)

[0056]

[0057] Note: TE represents the time when the ultrasonic signal is first received by the ultrasonic sensor; T represents the number of time point data of the acquired signal from the time point after reception;

[0058] The screening and verification process of determining the ultrasonic signal pulse factor as the feature representing the dielectric loss of the transformer oil is as follows:

[0059] Step 1: Obtain 5 different aging degrees (aging 0 days, aging 6 days, aging 15 days, aging 20 days, and aging 28 days) of transformer oil samples through an oil-paper insulation accelerated thermal aging test, and mark them as oil sample 1, oil sample 2, oil sample 3, oil sample 4, and oil sample 5 respectively, and each type of sample is 20, a total of 100 samples; at the same time, use a dielectric loss tester to complete the dielectric loss test of all oil sample samples, and obtain the test results as shown in Figure 1 .

[0060] Step 2: Use ultrasonic technology to complete the ultrasonic signal measurement of all oil sample samples, and extract 10 statistical characteristics of each oil sample ultrasonic signal as shown in Table 1;

[0061] Step 3: Calculate the correlation coefficient between each statistical characteristic and the dielectric loss value of the oil sample by using the following Pearson correlation coefficient formula, and obtain the calculation results as shown in Figure 3 .

[0062]

[0063] In the formula, n is the total number of oil sample samples, n = 100; C is each statistical characteristic of the ultrasonic signal; and Y is the dielectric loss value of the oil sample.

[0064] As shown in Figure 2 , among the 10 time domain statistical characteristics of the oil sample ultrasonic signal, the correlation coefficient between the pulse factor and the dielectric loss value of the oil sample is the highest, and reaches 0.907, which proves that the pulse factor of the ultrasonic signal has a high correlation with the dielectric loss, and it has high reliability and high effectiveness to use the pulse factor of the ultrasonic signal as a characteristic representing the dielectric loss of the transformer oil to establish a dielectric loss prediction model, which can provide reliable guarantee for accurate, reliable, and rapid non-destructive detection of the dielectric loss of the transformer oil based on the dielectric loss prediction model. The following embodiments will describe the dielectric loss prediction method of the transformer oil in the temperature change process in detail.

[0065] In one embodiment, as shown in Figure 3 , a dielectric loss prediction method in a temperature change process of transformer oil is provided, comprising the following steps:

[0066] S11, obtain transformer oil samples with different aging degrees through an oil-paper insulation accelerated thermal aging test; wherein the different aging degrees of the transformer oil samples can be selected according to actual application requirements, for example, the specific different aging degree types and the number of oil sample samples of different aging degree types can be set according to requirements, which are not limited here;

[0067] S12, obtain oil sample ultrasonic signals of each transformer oil sample according to ultrasonic technology, and obtain corresponding oil sample ultrasonic pulse factors according to each oil sample ultrasonic signal;

[0068] Specifically, the step of acquiring the ultrasonic signal of each transformer oil sample according to the ultrasonic technology comprises:

[0069] A transformer oil ultrasonic detection platform is constructed in advance; in principle, the existing related ultrasonic detection device can be used, but the influence of temperature on the ultrasonic signal is also considered when the dielectric loss is predicted, in order to facilitate the collection of effective ultrasonic signals at different oil temperatures and improve the accuracy of dielectric loss prediction, the embodiment preferably provides Figure 4 The transformer oil ultrasonic detection platform shown in the figure comprises an ultrasonic signal emitting device P1, a temperature regulating device P2, an oil container P3 and an ultrasonic signal detection device P4, and as shown in Figure 5 The ultrasonic signal emitting device P1 comprises a signal generator P11, a power amplifier P12 and a first ultrasonic sensor P13 connected in sequence; the temperature regulating device P2 comprises a temperature controller P21, and a temperature sensor P22 and a heating rod P23 connected with the temperature controller P21; the ultrasonic signal detection device P4 comprises a second ultrasonic sensor P41, a digital oscilloscope P42 and a computing terminal P43 connected in sequence; the first ultrasonic sensor P13, the temperature sensor P22, the heating rod P23 and the second ultrasonic sensor P41 are all immersed in the transformer oil in the oil container P3;

[0070] According to the transformer oil ultrasonic detection platform, ultrasonic detection is performed on each transformer oil sample to obtain the corresponding ultrasonic signal; wherein the specific detection method of the ultrasonic signal comprises:

[0071] The frequency, waveform and amplitude of the signal generator are set; the amplification gain of the power amplifier, and the center frequency of the first ultrasonic sensor and the second ultrasonic sensor are set; at the same time, the measurement environment temperature of the oil sample is set; the transformer oil ultrasonic detection platform is started for ultrasonic detection, the ultrasonic signal of each detection is acquired through the computing terminal, and in order to reduce the detection error as much as possible, the embodiment preferably adopts the method of repeatedly measuring the ultrasonic signal of each oil sample multiple times and then taking the average value to obtain the final used ultrasonic signal;

[0072] It should be noted that the selection of the above-mentioned transformer oil ultrasonic detection platform related equipment parameters can be determined according to the actual situation, but it should be noted that in order to ensure that the performance of the first ultrasonic sensor and the second ultrasonic sensor reaches the best, the frequency of the signal generator needs to be consistent with the center frequency of the ultrasonic sensor, and the amplitude of the signal generator and the related parameters of the power amplifier can be adjusted according to the actual measurement situation; for example, the frequency of the signal generator is set to 2MHz, the waveform is sine wave, and the amplitude is 200mV; the amplification gain of the power amplifier is 20dB, and the center frequency of the two ultrasonic sensors is 2MHz; the measurement environment of the oil sample is kept at 30℃, and the ultrasonic signal of each sample is measured 5 times and then averaged to obtain the final ultrasonic signal.

[0073] S13, according to the dielectric loss tester, the dielectric loss value of each transformer oil sample is obtained; wherein the dielectric loss tester only needs to meet the transformer oil dielectric loss measurement requirement, which is not limited here;

[0074] S14, according to the oil sample ultrasonic pulse factor and the oil sample dielectric loss value, a dielectric loss prediction model is established; wherein the type of the dielectric loss prediction model can also be selected according to actual needs in principle, but considering the simple and efficient demand of meeting real-time online prediction, and the strong correlation between the pulse factor and the dielectric loss value has been proved in the foregoing pulse factor screening process, the embodiment preferably adopts a linear relationship model to represent the correlation between the pulse factor and the dielectric loss value, that is, the oil sample ultrasonic pulse factor and the oil sample dielectric loss value are linearly fitted to obtain the corresponding dielectric loss prediction model, which is represented as:

[0075] Loss D =k1*I f +b1

[0076] Wherein, Loss D represents the dielectric loss; I f represents the ultrasonic pulse factor; k1 and b1 represent constants, which can be determined by fitting the actual sampled oil sample ultrasonic pulse factor and oil sample dielectric loss value, for example, based on the transformer oil sample shown in Figure 1 The dielectric loss prediction model obtained by linear fitting is: Loss D =0.00323*I f -0.00657; it should be noted that the specific coefficients given here are only exemplary descriptions, and are not specific limitations on the dielectric loss prediction model;

[0077] S15, real-time acquisition of the to-be-tested transformer oil, and input the to-be-tested ultrasonic pulse factor into the dielectric loss prediction model to predict the dielectric loss, and obtain a first dielectric loss prediction value; wherein, the acquisition method of the to-be-tested ultrasonic pulse factor can adopt the acquisition method of the oil sample ultrasonic pulse factor, which is not repeated here; the obtained to-be-tested ultrasonic pulse factor is input into the dielectric loss prediction model to obtain Loss D which is the first dielectric loss prediction value corresponding to the to-be-tested transformer oil.

[0078] In principle, the above-mentioned method of real-time prediction of dielectric loss of the to-be-tested transformer oil based on the dielectric loss prediction model has met the real-time online prediction, simple and convenient operation, and the use requirement of improving the accuracy of dielectric loss prediction in actual engineering application, but considering the influence of the temperature of the transformer oil in actual working condition on the ultrasonic signal, the embodiment also studies the change rule of the ultrasonic signal feature with the temperature of the transformer oil, and establishes a corresponding temperature correction dielectric loss prediction model, to further improve the accuracy of the dielectric loss prediction in the temperature change process of the transformer oil based on the ultrasonic signal feature.

[0079] Specifically, in one embodiment, a transformer oil temperature change process dielectric loss prediction method is provided, and the method further comprises:

[0080] According to the ultrasonic technology, the ultrasonic pulse factor of each transformer oil sample at a plurality of preset temperatures is acquired, and a corresponding pulse factor prediction model is established according to the ultrasonic pulse factor of each transformer oil sample at a plurality of preset temperatures; wherein, the pulse factor prediction model construction process comprises:

[0081] The transformer ultrasonic detection platform is used again to measure the ultrasonic signal of each transformer oil sample at a plurality of different temperature values in the preset temperature range (the temperature of the oil liquid is about 35-55°C when the actual working condition transformer is normal oil sampling), and after the corresponding ultrasonic pulse factor is obtained, it is divided into experimental set samples and validation set samples; in this embodiment, the ultrasonic signal of each transformer oil sample at 35°C, 40°C, 45°C, 50°C and 55°C is preferably acquired, and the corresponding sample division is shown in Table 2, and taking a sample in the oil sample as an example, the ultrasonic signal waveform shown in the following table can be obtained. Figure 6

[0082] Table 2 Transformer oil sample acquisition and division at different temperatures

[0083]

[0084] Based on the experimental set samples of the oil sample shown in Table 2, the change relationship linear equation between the ultrasonic signal pulse factor and the oil sample temperature of the oil liquid sample of the five different dielectric losses shown in Table 3 can be established, that is, the corresponding pulse factor prediction model is obtained.​ represents the pulse factor size at temperature T, wherein the pulse factor prediction model corresponding to oil sample 1 is as shown in Figure 7 .

[0085] Table 3 Relationship equation (equation slope and intercept) between ultrasonic signal pulse factor and temperature of five different oil samples

[0086] Oil sample type Slope k2 intercept b2 Equation fit R 2 ]]> Oil sample 1 0.1178 -1.4469 0.9908 Oil sample 2 0.1238 -1.5817 0.9893 Oil sample 3 0.1289 -1.6521 0.9903 Oil sample 4 0.1332 -1.7189 0.9892 Oil sample 5 0.1376 -1.7921 0.9887

[0087] According to the pulse factor prediction model, the dielectric loss prediction model is corrected to obtain a corresponding temperature-corrected dielectric loss prediction model; wherein the step of correcting the dielectric loss prediction model according to the pulse factor prediction model to obtain a corresponding temperature-corrected dielectric loss prediction model comprises:

[0088] Obtaining the real-time oil temperature of the transformer oil to be detected, and inputting the real-time oil temperature into the pulse factor prediction model to estimate the pulse factor, to obtain a pulse factor estimate value;

[0089] According to the pulse factor estimate value and the corresponding to-be-measured ultrasonic pulse factor, a temperature correction factor is obtained; the temperature correction factor is represented as:

[0090]

[0091] wherein β represents the temperature correction factor; and respectively represent the to-be-measured ultrasonic pulse factor and the pulse factor estimate value;

[0092] According to the temperature correction factor, the dielectric loss prediction model is corrected to obtain the temperature-corrected dielectric loss prediction model; wherein the temperature-corrected dielectric loss prediction model is represented as:

[0093]

[0094] wherein, represents the temperature-corrected dielectric loss; I f represents the ultrasonic pulse factor; β represents the temperature correction factor; k1 and b1 represent constants;

[0095] Specifically, as shown in the dielectric loss prediction model of the transformer oil sample, the temperature correction is performed to obtain a temperature-corrected dielectric loss prediction model as follows: Figure 1 wherein β0 represents the temperature correction factor calculated according to the oil temperature of the corresponding transformer oil sample and the pulse factor prediction model; it should be noted that the specific coefficients of the temperature-corrected dielectric loss prediction model given herein are only exemplary descriptions, and do not limit the specific temperature-corrected dielectric loss prediction model; ​

[0096] inputting the to-be-tested ultrasonic pulse factor into the temperature correction dielectric loss prediction model to perform dielectric loss prediction, to obtain a second dielectric loss prediction value;

[0097] To verify the effectiveness of the temperature correction dielectric loss prediction model obtained by the above method, in this embodiment, the ultrasonic signal pulse factors of all the verification set samples in Table 2 are calculated and the corresponding temperature correction factors, and the temperature correction dielectric loss prediction model is substituted into the temperature correction dielectric loss prediction model, and finally the dielectric loss prediction error shown in Table 4 is obtained.

[0098] Table 4: Dielectric loss prediction error of transformer oil samples at different temperatures

[0099]

[0100] In addition, in this embodiment, the real transformer oil samples (oil sample #1, oil sample #2, oil sample #3 and oil sample #4) of four actual operating conditions of a substation in Guangzhou City are also tested, and the ultrasonic signal of each sample is measured at five different temperatures, and the calculated pulse factor of the oil sample ultrasonic signal is substituted into the temperature correction dielectric loss prediction model between the pulse factor and the dielectric loss to obtain the prediction error of the dielectric loss prediction value and the dielectric loss measured value as shown in Table 5.

[0101] Table 5: Prediction error of dielectric loss of four real transformer oil samples by the dielectric loss prediction method of the present application

[0102]

[0103] From the dielectric loss prediction error results given in Table 4 and Table 5, it can be seen that the pulse factor can achieve good prediction effect on the actual transformer oil samples at different temperatures, and the prediction error can be stably controlled within 25%.

[0104] The embodiment of the present application provides a method for obtaining transformer oil samples with different aging degrees through an oil-paper insulation accelerated thermal aging test, obtaining oil sample ultrasonic signals of the transformer oil samples according to ultrasonic technology, obtaining corresponding oil sample ultrasonic pulse factors according to the oil sample ultrasonic signals, obtaining oil sample dielectric loss values of the transformer oil samples according to a dielectric loss tester, establishing a dielectric loss prediction model according to the oil sample ultrasonic pulse factors and the oil sample dielectric loss values, inputting a to-be-detected ultrasonic pulse factor of to-be-detected transformer oil into the dielectric loss prediction model for dielectric loss prediction to obtain a first dielectric loss prediction value, and further correcting the dielectric loss prediction result by establishing a temperature correction dielectric loss prediction model based on the ultrasonic signal pulse factor to obtain a second dielectric loss prediction value. Compared with the prior art, the method fully considers the high correlation between the pulse factor and the dielectric loss, takes the ultrasonic signal pulse factor as a feature for representing the dielectric loss of the transformer oil, establishes a simple and effective dielectric loss prediction model, and further establishes a temperature correction dielectric loss prediction model based on the sensitivity of the pulse factor to the temperature, so that the dielectric loss change in the temperature change process of the transformer oil is more accurately and reliably predicted in real time online. The method is simple and convenient, can effectively improve the accuracy of dielectric loss prediction in the real-time change process of the transformer oil temperature, and effectively guarantees the safe and reliable operation of the power transformer, and has very important engineering significance.

[0105] It should be noted that although each step in the above flowchart is displayed in sequence according to the arrow, these steps are not necessarily executed in the order indicated by the arrow. Unless otherwise specified herein, the execution of these steps has no strict order limitation, and these steps can be executed in other orders.

[0106] In one embodiment, as shown in Figure 8 A dielectric loss prediction system in a temperature change process of transformer oil is provided, and the system comprises:

[0107] An oil sample acquisition module 1 is configured to obtain transformer oil samples with different aging degrees through an oil-paper insulation accelerated thermal aging test.

[0108] An oil sample feature extraction module 2 is configured to obtain oil sample ultrasonic signals of the transformer oil samples according to ultrasonic technology, and obtain corresponding oil sample ultrasonic pulse factors according to the oil sample ultrasonic signals.

[0109] An oil sample dielectric loss detection module 3 is configured to obtain oil sample dielectric loss values of the transformer oil samples according to a dielectric loss tester.

[0110] A first model construction module 4 is configured to establish a dielectric loss prediction model according to the oil sample ultrasonic pulse factors and the oil sample dielectric loss values.

[0111] The first dielectric loss prediction module 5 is configured to acquire the to-be-tested ultrasonic pulse factor of the transformer oil to be tested in real time, and input the to-be-tested ultrasonic pulse factor into the dielectric loss prediction model to perform dielectric loss prediction, so as to obtain a first dielectric loss prediction value.

[0112] In one embodiment, a transformer oil dielectric loss prediction system during temperature change is provided, and the system further comprises:

[0113] The second model construction module is configured to acquire, according to ultrasonic technology, ultrasonic pulse factors of each transformer oil sample at a plurality of preset temperatures, and establish a corresponding pulse factor prediction model according to the ultrasonic pulse factors of each transformer oil sample at the plurality of preset temperatures.

[0114] The prediction model correction module is configured to correct the dielectric loss prediction model according to the pulse factor prediction model, so as to obtain a temperature-corrected dielectric loss prediction model.

[0115] The second dielectric loss prediction module is configured to input the to-be-tested ultrasonic pulse factor into the temperature-corrected dielectric loss prediction model to perform dielectric loss prediction, so as to obtain a second dielectric loss prediction value.

[0116] For specific limitations of the transformer oil dielectric loss prediction system during temperature change, refer to the limitations of the transformer oil dielectric loss prediction method during temperature change, and the corresponding technical effects can also be obtained equally, which will not be repeated here. Each module in the transformer oil dielectric loss prediction system during temperature change can be realized by software, hardware, and combinations thereof, in whole or in part. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above modules.

[0117] Figure 9 An internal structure diagram of a computer device in one embodiment is shown, which can be a terminal or a server. As shown in FIG. 6, the computer device includes a central processing unit (CPU), a memory, a storage device, a display device, an input device, and a communication interface. Figure 9As shown, the computer device includes a processor, a memory, a network interface, a display, a camera and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a transformer oil temperature change process dielectric loss prediction method. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0118] Those skilled in the art can understand that, Figure 9 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. A specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0119] In one embodiment, a computer device is provided, including a memory, a processor and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the steps of the above method.

[0120] In one embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by the processor to implement the steps of the above method.

[0121] In summary, the transformer oil temperature change process dielectric loss prediction method, system, computer device and storage medium provided by the embodiment of the present application realize that the transformer oil samples of different aging degrees are obtained through the oil-paper insulation accelerated thermal aging test, the oil sample ultrasonic signals of each transformer oil sample are obtained according to the ultrasonic technology, and the corresponding oil sample ultrasonic pulse factors are obtained according to each oil sample ultrasonic signal. After the oil sample dielectric loss value of each transformer oil sample is obtained according to the dielectric loss tester, the dielectric loss prediction model is established according to the oil sample ultrasonic pulse factor and the oil sample dielectric loss value, the measured ultrasonic pulse factor of the transformer oil to be detected is input into the dielectric loss prediction model for dielectric loss prediction to obtain the first dielectric loss prediction value, and the temperature correction dielectric loss prediction model based on the ultrasonic signal pulse factor is established to further correct the dielectric loss prediction result to obtain the second dielectric loss prediction value. The method fully considers the high correlation between the pulse factor and the dielectric loss, takes the ultrasonic signal pulse factor as a characteristic representing the dielectric loss of the transformer oil to establish a simple and effective dielectric loss prediction model, and further establishes a temperature correction dielectric loss prediction model based on the sensitivity of the pulse factor to the temperature to realize more accurate and reliable real-time online prediction of the dielectric loss change in the transformer oil temperature change process. Not only is the operation simple and convenient, but also the accuracy of the dielectric loss prediction in the real-time change process of the transformer oil temperature can be effectively improved, thereby effectively ensuring the safe and reliable operation of the power transformer, and having very important engineering significance.

[0122] Each embodiment in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, the system embodiment is basically similar to the method embodiment, so the description is relatively simple, and the related parts can be referred to the part of the method embodiment. It should be noted that each technical feature of the above embodiments can be combined arbitrarily, and in order to make the description simple, each technical feature of the above embodiments is not described in all possible combinations, but as long as the combination of these technical features does not exist contradictory, it should be considered as the scope of the description.

[0123] The above-described embodiments only express several preferred embodiments of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the patent. It should be noted that for ordinary skilled in the art, without departing from the technical principles of the present application, some improvements and replacements can be made, and these improvements and replacements should be considered as the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the protection scope of the claims.

Claims

1. A method for predicting dielectric loss of transformer oil during temperature change, characterized by, The method comprises the following steps: obtaining transformer oil samples of different aging degrees through an oil-paper insulation accelerated thermal aging test; obtaining oil sample ultrasonic signals of the transformer oil samples according to ultrasonic technology, and obtaining corresponding oil sample ultrasonic pulse factors according to the oil sample ultrasonic signals; obtaining oil sample dielectric loss values of the transformer oil samples according to a dielectric loss tester; establishing a dielectric loss prediction model according to the oil sample ultrasonic pulse factors and the oil sample dielectric loss values; the dielectric loss prediction model is expressed as: wherein, represents the medium loss; represents the ultrasonic pulse factor; and represents a constant; obtaining a first dielectric loss prediction value by inputting a to-be-detected ultrasonic pulse factor of transformer oil to be detected into the dielectric loss prediction model for dielectric loss prediction in real time.

2. The transformer oil temperature change process dielectric loss prediction method of claim 1, wherein, The step of obtaining the ultrasonic signals of the transformer oil samples according to the ultrasonic technology comprises: previously constructing a transformer oil ultrasonic detection platform; the transformer oil ultrasonic detection platform comprises an ultrasonic signal emitting device, a temperature control device, an oil container and an ultrasonic signal detection device; the ultrasonic signal emitting device comprises a signal generator, a power amplifier and a first ultrasonic sensor connected in sequence; the temperature control device comprises a temperature controller, a temperature sensor and a heating rod connected with the temperature controller; the ultrasonic signal detection device comprises a second ultrasonic sensor, a digital oscilloscope and a computing terminal connected in sequence; the first ultrasonic sensor, the temperature sensor, the heating rod and the second ultrasonic sensor are all immersed in transformer oil in the oil container; performing ultrasonic detection on the transformer oil samples according to the transformer oil ultrasonic detection platform to obtain corresponding ultrasonic signals.

3. The transformer oil temperature change process dielectric loss prediction method of claim 1, wherein, The method further comprises: obtaining ultrasonic pulse factors of the transformer oil samples at a plurality of preset temperatures according to ultrasonic technology, and establishing a corresponding pulse factor prediction model according to the ultrasonic pulse factors of the transformer oil samples at the plurality of preset temperatures; modifying the dielectric loss prediction model according to the pulse factor prediction model to obtain a corresponding temperature-modified dielectric loss prediction model; inputting the to-be-detected ultrasonic pulse factor into the temperature-modified dielectric loss prediction model for dielectric loss prediction to obtain a second dielectric loss prediction value.

4. The transformer oil temperature change process dielectric loss prediction method of claim 3, wherein, The step of modifying the dielectric loss prediction model according to the pulse factor prediction model to obtain a corresponding temperature-modified dielectric loss prediction model comprises: obtaining a real-time oil temperature of transformer oil to be detected, and inputting the real-time oil temperature into the pulse factor prediction model for pulse factor estimation to obtain a pulse factor estimation value; obtaining a temperature correction factor according to the pulse factor estimation value and a corresponding to-be-detected ultrasonic pulse factor; the temperature correction factor is expressed as: wherein represents a temperature correction factor; and respectively represent the ultrasound pulse factor to be measured and the pulse factor estimate. modifying the dielectric loss prediction model according to the temperature correction factor to obtain the temperature-modified dielectric loss prediction model.

5. The transformer oil temperature change process dielectric loss prediction method of claim 3, wherein, The temperature-modified dielectric loss prediction model is expressed as: wherein, represents the temperature-corrected medium loss; represents the ultrasonic pulse factor; represents the temperature correction factor; and represents a constant.

6. A transformer oil temperature change process dielectric loss prediction system characterized by, The system comprises: an oil sample obtaining module for obtaining transformer oil samples of different aging degrees through an oil-paper insulation accelerated thermal aging test; an oil sample feature extraction module for obtaining oil sample ultrasonic signals of the transformer oil samples according to ultrasonic technology, and obtaining corresponding oil sample ultrasonic pulse factors according to the oil sample ultrasonic signals; The oil sample dielectric loss detection module is configured to obtain the oil sample dielectric loss values of the transformer oil samples according to the dielectric loss tester. The first model construction module is configured to establish a dielectric loss prediction model according to the oil sample ultrasonic pulse factor and the oil sample dielectric loss values, and the dielectric loss prediction model is expressed as: wherein, represents the medium loss; represents the ultrasonic pulse factor; and represents a constant; The first dielectric loss prediction module is configured to obtain a to-be-detected ultrasonic pulse factor of the transformer oil to be detected in real time, and input the to-be-detected ultrasonic pulse factor into the dielectric loss prediction model to perform dielectric loss prediction, so as to obtain a first dielectric loss prediction value.

7. The transformer oil temperature change process dielectric loss prediction system of claim 6, wherein, The system further comprises: The second model construction module is configured to obtain the ultrasonic pulse factors of the transformer oil samples at a plurality of preset temperatures according to the ultrasonic technology, and establish a corresponding pulse factor prediction model according to the ultrasonic pulse factors of the transformer oil samples at the plurality of preset temperatures. The prediction model correction module is configured to correct the dielectric loss prediction model according to the pulse factor prediction model, so as to obtain a temperature-corrected dielectric loss prediction model. The second dielectric loss prediction module is configured to input the to-be-detected ultrasonic pulse factor into the temperature-corrected dielectric loss prediction model to perform dielectric loss prediction, so as to obtain a second dielectric loss prediction value.

8. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the method in any one of claims 1 to 5.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 5.

Images