A withstand voltage test method and system for transformer production

Abstract

The present application relates to the technical field of direct current converter transformer, and particularly relates to a withstand voltage test method and system for transformer production, comprising: obtaining a breakdown transient current signal and air pressure in a power frequency withstand voltage test process; dividing the breakdown transient current signal into positive signals and negative signals, and respectively decomposing the positive signals and the negative signals to obtain all node signals of the positive signals and the negative signals, further obtaining a polarity difference degree of the positive signals and the negative signals on each node signal, combining energy and energy release rate of the positive signals and the negative signals on each node signal to obtain a real possibility of air gap breakdown, so as to judge whether breakdown occurring in the power frequency withstand voltage test is air gap breakdown. The present application analyzes current change of the transformer in a local time range at a breakdown moment, so as to accurately distinguish air breakdown and insulator breakdown.

Description

Technical Field

[0001] This invention relates to the field of DC converter transformer technology, and more specifically to a withstand voltage test method and system for transformer production. Background Technology

[0002] A transformer is mainly composed of two or more sets of coils and an iron core. Its purpose is to step up or down the voltage of alternating current, change the impedance, and separate circuits. To ensure that the insulation performance of the produced transformer meets the design requirements, a withstand voltage test must be performed on the transformer during the production stage to avoid accidents caused by insulation breakdown during the use of the transformer.

[0003] The traditional power frequency withstand voltage test method for transformer withstand voltage testing involves applying a voltage 1.5 to 2 times higher than the rated voltage to the transformer and maintaining it for one minute to obtain the insulation performance between different windings and between the windings and ground, thereby conducting a withstand voltage test on the transformer. However, during the power frequency withstand voltage test, air gap breakdown may occur, and air gap breakdown can easily be confused with actual insulation defects. Therefore, the traditional power frequency withstand voltage test method for transformer withstand voltage testing may lead to misjudgments. Summary of the Invention

[0004] This invention provides a withstand voltage test method and system for transformer production to solve the existing problem that traditional power frequency withstand voltage test methods cannot accurately detect transformer quality.

[0005] The present invention provides a withstand voltage testing method and system for transformer production, which adopts the following technical solution:

[0006] One embodiment of the present invention provides a withstand voltage test method for transformer production, the method comprising the following steps:

[0007] Acquire the breakdown transient current signal and the air pressure during the power frequency withstand voltage test;

[0008] The breakdown transient current signal is divided into positive and negative signals, and each signal is further decomposed to obtain several node signals. Based on the differences in amplitude and index value of the data points in the node signals, the matching distance between the data points of the positive and negative signals is obtained, thus obtaining several matching pairs of the positive and negative signals in the node signals. Based on the phase difference and matching distance between the data points in all matching pairs of the positive and negative signals in the node signals, the degree of polarity difference between the positive and negative signals in the node signals is obtained.

[0009] Based on the amplitude and corresponding frequency range of the positive and negative signals on the same labeled node signals, and combined with the degree of polarity difference between the positive and negative signals on the node signals, the true polarity difference between the positive and negative signals on each node signal is obtained; the breakdown transient current signal is divided into rising segment and falling segment, and the energy release rate of the breakdown transient current signal is obtained based on the rising segment and falling segment. Combined with the true polarity difference between the positive and negative signals on each node signal, the air gap breakdown factor is obtained.

[0010] Based on the air pressure and air gap breakdown factor during the power frequency withstand voltage test, the true probability of air gap breakdown is obtained, and it is determined whether the breakdown that occurs in the power frequency withstand voltage test is air gap breakdown.

[0011] Preferably, the specific method for dividing the breakdown transient current signal into positive and negative signals, and decomposing the positive and negative signals respectively to obtain several node signals of the positive and negative signals is as follows:

[0012] The portion of the breakdown transient current signal with an amplitude greater than or equal to 0 is classified as a positive signal, and the portion of the breakdown transient current signal with an amplitude less than 0 is classified as a negative signal.

[0013] Wavelet packet decomposition is performed on the positive and negative signals respectively to obtain all node signals of the positive and negative signals respectively. The number of decomposition levels of the wavelet packet decomposition is: The This is the preset number of decomposition layers.

[0014] Preferably, the specific method for obtaining the matching distance between data points of the positive and negative signals in the node signals based on the difference in amplitude and index value of the data points of the positive and negative signals in the node signals, and thus obtaining several matching pairs of the positive and negative signals in the node signals, includes:

[0015] Based on the frequency ranges corresponding to all node signals of both the positive and negative signals, sort all node signals of both signals in ascending order and assign an index label to each node signal of both signals sequentially; for the positive signal... Within the signal of the first node The data point and the negative signal Within the signal of the first node The data point; based on the positive signal's first data point; Within the signal of the first node The data point and the negative signal Within the signal of the first node The amplitude and index of the nth data point are used to obtain the nth positive signal. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points;

[0016] The DTW algorithm is used to analyze the first positive signal. Each data point within the nth node signal is related to the first negative signal. Within each node signal, each data point is matched based on the matching distance between them to obtain the positive and negative signals at the [number]th node. Several matching pairs of node signals.

[0017] Preferably, the first step in acquiring the positive signal... Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points includes the following specific methods:

[0018] The first positive signal Within the signal of the first node The index value of the nth data point, minus the index value of the negative signal. Within the signal of the first node The square of the difference between the index values ​​of the data points is denoted as the horizontal distance; the positive signal's first... Within the signal of the first node The amplitude of the nth data point, plus the nth negative signal Within the signal of the first node The square of the sum of the amplitudes of the data points is taken as the vertical distance; the square root of the sum of the horizontal and vertical distances is taken as the positive signal's first value. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points.

[0019] Preferably, the method for obtaining the degree of polarity difference between the positive and negative signals on the node signal based on the phase difference and matching distance between data points in all matching pairs of the positive and negative signals on the node signal includes:

[0020] For any data point in any node signal of a positive signal, a local data is preset. The data point within the node signal of the positive signal that is closest to the data point within the node signal of the positive signal. A number of data points are used as local data points of the data points in the node signal of the positive signal. Hilbert transform calculation is performed on the local data points of the data points in the node signal of the positive signal to obtain the phase of the data points in the node signal of the positive signal.

[0021] Obtain the phase of all data points in all node signals for both positive and negative signals;

[0022] For positive and negative signals, the first The node signal will be compared with the positive signal at the 1st node. The absolute value of the phase difference between two data points in any matched pair of the n node signals is used as the positive signal and the negative signal in the nth node. The weight of the matching distance between two data points in the matching pair of the node signals; for positive and negative signals at the The weighted average of the matching distances between the two data points in all matching pairs of the nth node signal is used to obtain the positive and negative signals at the nth node. The degree of polarity difference on the signals of each node.

[0023] Preferably, the specific method for obtaining the true polarity difference between the positive and negative signals on each node signal based on the amplitude and corresponding frequency range of the positive and negative signals on the same labeled node signal, combined with the degree of polarity difference between the positive and negative signals on the node signal, includes:

[0024] For the positive signal The node signal will be the positive signal of the first node. The sum of the squares of the amplitudes of all data points in the nth node signal is taken as the nth node signal. The energy of the node signal; the positive signal of the first node. The energy of the nth node signal is divided by the sum of the energies of all node signals of the positive signal. This ratio is taken as the energy of the nth node signal of the positive signal. The energy factor of the node signal; the positive signal's first node signal The index label of the nth node signal is compared with the largest index label of all node signals of the positive signal. The resulting ratio is taken as the index label of the nth node signal of the positive signal. The index factor of the nth node signal will be the positive signal's nth node signal. The product of the energy factor and the index factor of the nth node signal is used as the first positive signal. The single-pole high-frequency factor of the signal at each node;

[0025] The first step in obtaining the negative signal The single-pole high-frequency factor of the nth node signal will be the positive signal's nth node signal. The single-pole high-frequency factor of the nth node signal and the negative signal The product of the single-pole high-frequency factors of the n node signals is used as the first positive signal and the second negative signal. The high-frequency factor of the node signal; the positive signal and the negative signal of the first node. The high-frequency factors of the signal at each node and the positive and negative signals at the 1st node. The product of the polarity differences at each node signal is used as the positive and negative signals at the 1st node. The degree of true polarity difference on the signals at each node.

[0026] Preferably, the specific method for dividing the breakdown transient current signal into a rising segment and a falling segment, and obtaining the energy release rate of the breakdown transient current signal based on the rising segment and the falling segment, includes:

[0027] The data point corresponding to the breakdown moment of the breakdown transient current signal is used as the segmentation point. The breakdown transient current signal is divided into rising segment and falling segment. The rising segment and falling segment are integrated respectively to obtain the integration results of the rising segment and the falling segment. The ratio between the integration result of the rising segment and the integration result of the falling segment is used as the energy release rate of the breakdown transient current signal.

[0028] Preferably, the specific method for obtaining the air gap breakdown factor is as follows:

[0029] The product of the true polarity difference between the positive and negative signals at each node signal and the energy release rate of the breakdown transient current signal is used as the air gap breakdown characteristic of the positive and negative signals at each node signal. The sum of the air gap breakdown characteristics of the positive and negative signals at each node signal is used as the air gap breakdown factor.

[0030] Preferably, the specific method for obtaining the true probability of air gap breakdown based on the air pressure and air gap breakdown factor during the power frequency withstand voltage test includes:

[0031] Obtain the worst-case air pressure, and record the absolute value of the difference between the air pressure during the power frequency withstand voltage test and the worst-case air pressure as the air pressure influence coefficient. Normalize the ratio of the air gap breakdown factor to the air pressure influence coefficient, and take the normalized result as the true probability of air gap breakdown.

[0032] Another embodiment of the present invention provides a withstand voltage testing system for transformer production, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any of the above-described withstand voltage testing methods for transformer production.

[0033] The beneficial effects of the technical solution of this invention are as follows: Since the physical process of air gap breakdown is highly dependent on electrode polarity, while insulator breakdown is caused by defects inside the insulator material, and electrode polarity has no significant effect on the direction of the discharge path, the breakdown transient current signal can be divided into positive and negative signals based on this. Furthermore, since the moment of breakdown during the power frequency withstand voltage test is an external manifestation of the superposition of multiple physical processes, in order to map the superimposed physical processes to the mathematical feature space, it is necessary to decompose the breakdown transient current signal to obtain several node signals of positive and negative signals. By quantifying the polarity difference between positive and negative signals at the node signals, high-frequency detail features are avoided from being masked by low-frequency signals. At the same time, since the rapid migration of positive signal electrons and the slow migration of negative signal ions during air gap breakdown cause significant phase misalignment, in order to make the subsequent judgment on whether it is air gap breakdown by the degree of polarity difference between positive and negative signals at the same node signal more accurate, it is necessary to use the phase difference between the two data points in each matching degree as the weight, so as to accurately obtain the degree of polarity difference between positive and negative signals at each node signal.

[0034] Furthermore, since the high-frequency component accounts for a large proportion of the transient current change generated during air gap breakdown, while the high-frequency component accounts for a small proportion of the transient current change generated during insulator breakdown in a transformer, when determining whether the transient current signal of breakdown is the current signal generated during air gap breakdown by the degree of polarity difference between positive and negative signals at all node signals, the frequency range and energy corresponding to each node signal should be considered to accurately distinguish between air gap breakdown and insulator breakdown. In addition, since the test environment of the power frequency withstand voltage test also affects air gap breakdown, this application analyzes the change of current in the local time range of the transformer at the moment of breakdown by further combining the analysis of the power frequency withstand voltage test environment, so as to accurately distinguish between air breakdown and insulator breakdown. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart illustrating the steps of a withstand voltage test method for transformer production according to the present invention. Detailed Implementation

[0037] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a withstand voltage testing method and system for transformer production based on the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0039] The following description, in conjunction with the accompanying drawings, details a specific scheme for a withstand voltage testing method and system for transformer production provided by the present invention.

[0040] Please see Figure 1 The diagram illustrates a flowchart of a withstand voltage test method for transformer production according to an embodiment of the present invention. The method includes the following steps:

[0041] Step S001: Obtain the breakdown transient current signal and the air pressure during the power frequency withstand voltage test.

[0042] It should be noted that during the traditional power frequency withstand voltage test for transformers, air gap breakdown may occur. Air gap breakdown can easily be confused with actual insulation faults in the test. Therefore, the traditional power frequency withstand voltage test method for transformers may lead to misjudgment. Therefore, this embodiment proposes a withstand voltage test method for transformer production to accurately distinguish between air gap breakdown and insulation breakdown in the power frequency withstand voltage test.

[0043] Specifically, a sampling frequency is preset. With transient time range The and The specific value can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, it is used as... hertz, Taking milliseconds as an example, during the withstand voltage test of the transformer through the power frequency withstand voltage test, the high-frequency current transformer is connected to the transformer grounding circuit, causing the high-frequency current transformer to... The sampling frequency is used to collect the current signal. When a breakdown occurs during the power frequency withstand voltage test, the power supply is immediately cut off, and the moment of breakdown is recorded as the breakdown time. The distance from the breakdown time is calculated with the breakdown time as the center. The current signal collected within the range is used as the breakdown transient current signal, and the air pressure during the power frequency withstand voltage test is collected by the air pressure sensor. Since the power frequency withstand voltage test is a well-known existing technology, it will not be described in detail in this embodiment.

[0044] Thus, the breakdown transient current signal and the air pressure during the power frequency withstand voltage test were obtained.

[0045] Step S002: Divide the breakdown transient current signal into a positive signal and a negative signal, and decompose the positive signal and the negative signal respectively to obtain several node signals of the positive signal and the negative signal; according to the difference in amplitude and index value of the data points of the positive signal and the negative signal in the node signals, obtain the matching distance between each data point in the node signals of the positive signal and the negative signal, and then obtain several matching pairs of the positive signal and the negative signal in the node signals; according to the phase difference and matching distance between the data points in all matching pairs of the positive signal and the negative signal in the node signals, obtain the degree of polarity difference of the positive signal and the negative signal in the node signals.

[0046] It should be noted that the physical process of air gap breakdown is highly dependent on electrode polarity. For the positive electrode, under high voltage, electrons are triggered by collisions with the gas in the air gap, resulting in avalanche ionization and the formation of electron avalanches. The high electric field strength of the positive electrode causes a large number of positive ions to accumulate at the head of the electron avalanches, which strengthens the electric field and causes the streamer to extend rapidly towards the negative electrode. The root of the streamer is heated by the Joule current to form a high-temperature plasma channel (leader), which further guides the discharge towards the negative electrode. For the negative electrode, although electrons from the negative electrode migrate towards the positive electrode, the tail of the electron avalanches leaves behind a cloud of positive ions. This cloud of positive ions forms a reverse electric field at the negative electrode, thereby suppressing the subsequent ionization process. In other words, the physical process of air gap breakdown is highly dependent on electrode polarity. In contrast, insulator breakdown is caused by defects inside the insulator material. Electrode polarity has no significant effect on the direction of the discharge path. Therefore, this can be used as a basis to divide the transient current signal of breakdown into positive and negative signals. The difference between the positive and negative signals can be used to quantify the difference in polarity between the positive and negative electrodes, thus distinguishing between air gap breakdown and insulator breakdown.

[0047] Preferably, in a specific embodiment of the present invention, the portion of the breakdown transient current signal with an amplitude greater than or equal to 0 is classified as a positive signal, and the portion of the breakdown transient current signal with an amplitude less than 0 is classified as a negative signal.

[0048] It should be noted that the moment of breakdown during the power frequency withstand voltage test is an external manifestation of the superposition of multiple physical processes. Among them, air gap breakdown is an external manifestation of the superposition of millisecond-level arc, microsecond-level ionization, and nanosecond-level streamer; while insulator breakdown is an external manifestation of the superposition of defect charge release and carbonization channel expansion. Therefore, in order to map the superimposed physical processes to the mathematical feature space, it is necessary to decompose the breakdown transient current signal to avoid high-frequency details being masked by low-frequency signals.

[0049] Preferably, in a specific embodiment of the present invention, wavelet packet decomposition is performed on the positive signal and the negative signal respectively to obtain all node signals of the positive signal and the negative signal respectively, wherein the number of decomposition layers of the wavelet packet decomposition is: The The preset number of decomposition layers, The specific value can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, it is used as... Taking db10 as an example, this embodiment does not impose strict requirements on the selection of the small fundamental wavelet in wavelet packet decomposition. Since wavelet packet decomposition is a well-known prior art, it will not be described in detail in this embodiment.

[0050] Furthermore, based on the frequency ranges corresponding to all node signals of both the positive and negative signals, the node signals of both signals are sorted in ascending order, and an index label is assigned to each node signal of both signals sequentially; for the positive signal... Within the signal of the first node The data point and the negative signal Within the signal of the first node The data point; based on the positive signal's first data point; Within the signal of the first node The data point and the negative signal Within the signal of the first node The amplitude and index of the nth data point are used to obtain the nth positive signal. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points;

[0051] The first step in obtaining the positive signal Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points includes the following specific methods:

[0052] The first positive signal Within the signal of the first node The index value of the nth data point, minus the index value of the negative signal. Within the signal of the first node The square of the difference between the index values ​​of the data points is denoted as the horizontal distance; the positive signal's first... Within the signal of the first node The amplitude of the nth data point, plus the nth negative signal Within the signal of the first node The square of the sum of the amplitudes of the data points is taken as the vertical distance; the square root of the sum of the horizontal and vertical distances is taken as the positive signal's first value. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points;

[0053] As an example, obtaining the first positive signal Within the signal of the first node The data point and the negative signal Within the signal of the first node The specific formula for calculating the matching distance between data points is as follows:

[0054]

[0055] In the formula, The first positive signal Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points; The first positive signal Within the signal of the first node The index value of each data point; The first negative signal Within the signal of the first node The index value of each data point; The first positive signal Within the signal of the first node The magnitude of each data point; The first negative signal Within the signal of the first node The amplitude of each data point.

[0056] Furthermore, the DTW algorithm is used to analyze the first positive signal. Each data point within the nth node signal is related to the first negative signal. Within each node signal, each data point is matched based on the matching distance between them to obtain the positive and negative signals at the [number]th node. The matching pairs of the node signals are described in detail in this embodiment, as the DTW algorithm is a well-known existing technology.

[0057] It should be noted that when positive and negative signals are matched in the same node signal, the smaller the matching distance between the two data points, the smaller the polarity difference between the positive and negative signals. Therefore, the degree of polarity difference between positive and negative signals in the same node signal can be quantified by the matching distance between the two data points in all matched pairs of positive and negative signals in the same node signal.

[0058] It should be further explained that, due to the significant phase misalignment caused by the rapid migration (leading phase) of positive signal electrons and the slow migration (lagging phase) of negative signal ions when air gap breakdown occurs, the phase difference between the positive and negative signals in air gap breakdown is large. In order to more accurately distinguish between air gap breakdown and insulator breakdown, when quantifying the degree of polarity difference between positive and negative signals at the same node signal through all matched pairs of positive and negative signals, the phase difference between data points within the matched pair should be used as the weight. This makes the subsequent judgment on whether it is air gap breakdown based on the degree of polarity difference between positive and negative signals at the same node signal more accurate.

[0059] Preferably, in a specific embodiment of the present invention, for any data point in any node signal of a positive signal, a local data is preset. The The specific value can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, it is used as... Taking this as an example, the data point within the node signal of the positive signal that is closest to the node signal of the positive signal is... A number of data points are used as local data points of the data points in the node signal of the positive signal. Hilbert transform is performed on the local data points of the data points in the node signal of the positive signal to obtain the phase of the data points in the node signal of the positive signal. Since the Hilbert transform is a well-known prior art, it will not be described in detail in this embodiment.

[0060] Similarly, obtain the phase of all data points in all node signals of both positive and negative signals;

[0061] Furthermore, for the positive and negative signals... The node signal will be compared with the positive signal at the 1st node. The absolute value of the phase difference between the two data points in each matched pair of the node signals is used as the positive signal and the negative signal in the th node. The weight of the matching distance between the two data points in each matching pair of the node signals; for positive and negative signals in the th node... The weighted average of the matching distances between the two data points in all matching pairs of the nth node signal is used to obtain the positive and negative signals at the nth node. The degree of polarity difference on the signals of each node.

[0062] As an example, positive and negative signals in the first... The specific formula for calculating the degree of polarity difference on the signals of each node is as follows:

[0063]

[0064] In the formula, This indicates that the positive and negative signals are at the 1st... The degree of polarity difference on the signals of each node; This indicates that the positive and negative signals are at the 1st... The number of all matched pairs of the node signal; This indicates that the positive and negative signals are at the 1st... The first node signal The matching distance between two data points in a matching pair; This indicates that the positive and negative signals are at the 1st... The first node signal The phase of the first data point in each matching pair; This indicates that the positive and negative signals are at the 1st... The first node signal The phase of the second data point in each matching pair; This represents the function that takes the absolute value.

[0065] It should be noted that, This represents the matching distance between data points within a matching pair. The larger the matching distance between data points within all matching pairs, the stronger the positive signal. The first node signal and the negative signal The worse the symmetry between the signals at each node, that is, the more pronounced the difference between the positive and negative signals at the 1st node... The greater the polarity difference on the signal at each node, and the more significant the phase misalignment caused by the rapid migration of positive signal electrons and the slow migration of negative signal ions during air gap breakdown, the more accurate the subsequent judgment on whether it is air gap breakdown is based on the degree of polarity difference between positive and negative signals on the same node signal. Therefore, the phase difference between the two data points in each matching degree should be used as the weight to accurately obtain the degree of polarity difference between positive and negative signals on each node signal.

[0066] Thus, the degree of polarity difference between positive and negative signals at each node signal is obtained.

[0067] Step S003: Based on the amplitude and corresponding frequency range of the positive and negative signals on the same labeled node signals, and combined with the degree of polarity difference between the positive and negative signals on the node signals, obtain the true polarity difference between the positive and negative signals on each node signal; divide the breakdown transient current signal into rising segment and falling segment, obtain the energy release rate of the breakdown transient current signal based on the rising segment and falling segment, and combine with the true polarity difference between the positive and negative signals on each node signal to obtain the air gap breakdown factor.

[0068] It should be noted that when conducting withstand voltage tests on transformers based on power frequency withstand voltage tests, whether it is air gap breakdown or insulation breakdown in the transformer, transient changes in current will occur during breakdown. Since air gap breakdown is usually a gas discharge involving rapid electrical discharge and arc formation, this process will produce a steep current rise, resulting in high-frequency oscillations. In contrast, insulation breakdown in the transformer involves the slow formation of a conductive path. Therefore, under normal circumstances, the high-frequency component accounts for a large proportion of the transient current change generated during air gap breakdown, while the high-frequency component accounts for a small proportion of the transient current change generated during insulation breakdown. Therefore, when determining whether the transient current signal of breakdown is the current signal generated during air gap breakdown by the degree of polarity difference between positive and negative signals at all node signals, the frequency range and energy of each node signal should be considered to accurately distinguish between air gap breakdown and insulation breakdown.

[0069] Preferably, in a specific embodiment of the present invention, for the first positive signal... The node signal will be the positive signal of the first node. The sum of the squares of the amplitudes of all data points in the nth node signal is taken as the nth node signal. The energy of the node signal; the positive signal of the first node. The energy of the nth node signal is divided by the sum of the energies of all node signals of the positive signal. This ratio is taken as the energy of the nth node signal of the positive signal. The energy factor of the node signal; the positive signal's first node signal The index label of the nth node signal is compared with the largest index label of all node signals of the positive signal. The resulting ratio is taken as the index label of the nth node signal of the positive signal. The index factor of the nth node signal will be the positive signal's nth node signal. The product of the energy factor and the index factor of the nth node signal is used as the first positive signal. The single-pole high-frequency factor of the signal at each node;

[0070] Similarly, to obtain the negative signal... The single-pole high-frequency factor of the nth node signal will be the positive signal's nth node signal. The single-pole high-frequency factor of the nth node signal and the negative signal The product of the single-pole high-frequency factors of the n node signals is used as the first positive signal and the second negative signal. The high-frequency factor of the node signal; the positive signal and the negative signal of the first node. The high-frequency factors of the signal at each node and the positive and negative signals at the 1st node. The product of the polarity differences at each node signal is used as the positive and negative signals at the 1st node. The degree of true polarity difference on the signals at each node.

[0071] It should be noted that, This represents the first positive signal and the second negative signal. The high-frequency component values ​​of the signal at each node. The larger the value, the higher the positive signal's relative strength compared to the negative signal. The higher the frequency range and energy of the node signal, the more the energy of the transient current signal generated by air gap breakdown is concentrated in the high-frequency components. Moreover, the polarity difference of the signal is large during air gap breakdown. That is, when air gap breakdown occurs, the child nodes with high frequency range and high energy are more affected by the air gap breakdown. Therefore, a larger weight should be assigned to the child nodes with high frequency range and high energy to evaluate the degree of polarity difference between the positive and negative signals on the node signal, so as to accurately distinguish between air gap breakdown and insulator breakdown.

[0072] It should be further explained that during the withstand voltage test of the transformer based on the power frequency withstand voltage test, when air gap breakdown occurs, the gas ionization is completed instantaneously, and the energy is released in a concentrated manner within a very short time. In contrast, when the insulator breaks down, it is formed by the gradual carbonization of the insulating material or the gradual expansion of the conductive path within the insulating material, and its energy release rate is slow. In order to further accurately distinguish between insulator breakdown and air gap breakdown, it is also necessary to obtain the energy release rate of the breakdown transient current signal. Based on the degree of difference in the true polarity of the positive and negative signals at all node signals, combined with the energy release rate of the breakdown transient current signal, the air gap breakdown factor can be obtained.

[0073] Preferably, in a specific embodiment of the present invention, the data point corresponding to the breakdown moment of the breakdown transient current signal is used as the segmentation point, and the breakdown transient current signal is divided into a rising segment and a falling segment. The rising segment and the falling segment are integrated respectively to obtain the integration results of the rising segment and the falling segment. The ratio between the integration result of the rising segment and the integration result of the falling segment is used as the energy release rate of the breakdown transient current signal.

[0074] It should be noted that the integral results of the rising and falling segments represent the energy of the rising current signal before breakdown and the energy of the falling current signal after breakdown, respectively. Since the energy is released in a concentrated manner during the breakdown of the air gap, the energy of the falling current signal after the breakdown of the air gap is much smaller than the energy of the falling current signal after the breakdown of the insulator. Therefore, the ratio between the integral results of the rising and falling segments can be used as the energy release rate of the transient current signal during breakdown.

[0075] Specifically, the product of the true polarity difference between the positive and negative signals at each node signal and the energy release rate of the breakdown transient current signal is used as the air gap breakdown characteristic of the positive and negative signals at each node signal, and the sum of the air gap breakdown characteristics of the positive and negative signals at each node signal is used as the air gap breakdown factor.

[0076] As an example, the specific formula for calculating the air gap breakdown factor is:

[0077]

[0078] In the formula, Indicates the air gap breakdown factor; This indicates that the positive and negative signals are at the 1st... The degree of true polarity difference on the signals of each node; This indicates the energy release rate of the breakdown transient current signal; Indicates the number of node signals.

[0079] It should be noted that, This indicates that the positive and negative signals are at the 1st digit. Air gap breakdown characteristics on all node signals; since the physical process of air gap breakdown is highly dependent on electrode polarity, while insulator breakdown is caused by defects inside the insulator material, electrode polarity has no significant effect on the direction of the discharge path. Therefore, the greater the difference in the true polarity of the positive and negative signals on all node signals, the more likely it is to be air gap breakdown. Since gas ionization is completed instantaneously during air gap breakdown, energy is released in a concentrated manner within a very short time. In contrast, insulator breakdown is formed by the gradual carbonization of the insulating material or the gradual expansion of the conductive path within the insulating material, and its energy release rate is slow. Therefore, the higher the energy release rate of the breakdown transient current signal, the more likely it is to be air gap breakdown.

[0080] Thus, the air gap breakdown factor is obtained.

[0081] Step S004: Based on the air pressure and air gap breakdown factor during the power frequency withstand voltage test, obtain the true probability of air gap breakdown and determine whether the breakdown that occurred during the power frequency withstand voltage test was air gap breakdown.

[0082] It should be noted that the air gap breakdown factor obtained in step S003 is only obtained by analyzing the transient current signal of breakdown during the power frequency withstand voltage test. However, when the transformer is actually subjected to withstand voltage test by power frequency withstand voltage test, the test environment of the power frequency withstand voltage test will also affect the air gap breakdown. Under normal circumstances, air is most easily broken down when the air pressure is 7.5 kPa. That is, the closer the ambient air pressure is to 7.5 kPa during the power frequency withstand voltage test, the easier it is for air gap breakdown to occur. At this time, when obtaining the true probability of air gap breakdown through the air gap breakdown factor obtained in step S003, it is necessary to appropriately amplify the probability of air gap breakdown.

[0083] Preferably, in a specific embodiment of the present invention, 7.5 kPa is taken as the worst air pressure, and the absolute value of the difference between the air pressure during the power frequency withstand voltage test and the worst air pressure is recorded as the air pressure influence coefficient. The ratio of the air gap breakdown factor to the air pressure influence coefficient is normalized (normalization can be performed using the sigmoid function), and the normalized result is taken as the true probability of air gap breakdown.

[0084] As an example, the specific formula for calculating the true probability of air gap breakdown is:

[0085]

[0086] In the formula, This indicates the actual probability of air gap breakdown; Indicates the air gap breakdown factor; This indicates the air pressure during the power frequency withstand voltage test; Indicates the worst air pressure; This represents the sigmoid function, which is used in this embodiment for normalization.

[0087] It should be noted that the closer the air pressure during the power frequency withstand voltage test is to the minimum air pressure of 7.5 kPa, the easier it is for the air to break down. The smaller the value, the more likely air gap breakdown will occur. Therefore, based on the specific air pressure during the power frequency withstand voltage test and the air gap breakdown factor obtained in step S003, the true probability of air gap breakdown can be obtained. After obtaining the true probability of air gap breakdown, it is possible to determine whether the breakdown that occurs in the power frequency withstand voltage test is insulator breakdown or air gap breakdown.

[0088] It should be further explained that, for all fractional operations in this embodiment, in order to avoid the denominator being zero during the fractional operation, 0.01 is added to both the numerator and denominator during the fractional operation.

[0089] Specifically, a threshold for the true probability is preset. The The specific value can be set according to the actual situation. This embodiment does not make a hard requirement. In this embodiment, it is used as... Taking an example, when the actual probability of air gap breakdown is greater than or equal to... If the breakdown that occurs during the power frequency withstand voltage test is an air gap breakdown, then the actual probability of air gap breakdown is less than [a certain value]. If the breakdown occurs during the power frequency withstand voltage test, it is an insulator breakdown.

[0090] It should be further explained that this embodiment analyzes the change in current within a local time range at the moment of transformer breakdown, and combines it with the test environment during the power frequency withstand voltage test to obtain the true probability of air gap breakdown. This is used to distinguish between insulation breakdown and air gap breakdown. When insulation breakdown occurs, the transformer is considered unqualified. When air gap breakdown occurs, the power frequency withstand voltage test is rearranged, and the transformer is retested.

[0091] Another embodiment of the present invention provides a withstand voltage testing system for transformer production, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a withstand voltage testing method for transformer production in steps S001 to S004.

[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A withstand voltage test method for transformer production, characterized in that, The method includes the following steps: Acquire the breakdown transient current signal and the air pressure during the power frequency withstand voltage test; The breakdown transient current signal is divided into positive and negative signals, and each signal is further decomposed to obtain several node signals. Based on the differences in amplitude and index value of the data points in the node signals, the matching distance between the data points of the positive and negative signals is obtained, thus obtaining several matching pairs of the positive and negative signals in the node signals. Based on the phase difference and matching distance between the data points in all matching pairs of the positive and negative signals in the node signals, the degree of polarity difference between the positive and negative signals in the node signals is obtained. Based on the amplitude and corresponding frequency range of the positive and negative signals on the same labeled node signals, and combined with the degree of polarity difference between the positive and negative signals on the node signals, the true polarity difference between the positive and negative signals on each node signal is obtained; the breakdown transient current signal is divided into rising segment and falling segment, and the energy release rate of the breakdown transient current signal is obtained based on the rising segment and falling segment. Combined with the true polarity difference between the positive and negative signals on each node signal, the air gap breakdown factor is obtained. Based on the air pressure and air gap breakdown factor during the power frequency withstand voltage test, the true probability of air gap breakdown is obtained, and it is determined whether the breakdown that occurs in the power frequency withstand voltage test is air gap breakdown.

2. The withstand voltage test method for transformer production according to claim 1, characterized in that, The specific method for dividing the breakdown transient current signal into positive and negative signals, and then decomposing the positive and negative signals to obtain several node signals of the positive and negative signals is as follows: The portion of the breakdown transient current signal with an amplitude greater than or equal to 0 is classified as a positive signal, and the portion of the breakdown transient current signal with an amplitude less than 0 is classified as a negative signal. Wavelet packet decomposition is performed on the positive and negative signals respectively to obtain all node signals of the positive and negative signals respectively. The number of decomposition levels of the wavelet packet decomposition is: The This is the preset number of decomposition layers.

3. The withstand voltage test method for transformer production according to claim 1, characterized in that, The method for obtaining the matching distance between data points of positive and negative signals in the node signals based on the difference in amplitude and index value of the data points of positive and negative signals in the node signals, and thus obtaining several matching pairs of positive and negative signals in the node signals, includes the following specific methods: Based on the frequency ranges corresponding to all node signals of both the positive and negative signals, sort all node signals of both signals in ascending order and assign an index label to each node signal of both signals sequentially; for the positive signal... Within the signal of the first node The data point and the negative signal Within the signal of the first node The data point; based on the positive signal's first data point; Within the signal of the first node The data point and the negative signal Within the signal of the first node The amplitude and index of the nth data point are used to obtain the nth positive signal. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points; The DTW algorithm is used to analyze the first positive signal. Each data point within the nth node signal is related to the first negative signal. Within each node signal, each data point is matched based on the matching distance between them to obtain the positive and negative signals at the [number]th node. Several matching pairs of node signals.

4. The withstand voltage test method for transformer production according to claim 3, characterized in that, The first step in obtaining the positive signal Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points includes the following specific methods: The first positive signal Within the signal of the first node The index value of the nth data point, minus the index value of the negative signal. Within the signal of the first node The square of the difference between the index values ​​of the data points is denoted as the horizontal distance; the positive signal's first... Within the signal of the first node The amplitude of the nth data point, plus the nth negative signal Within the signal of the first node The square of the sum of the amplitudes of the data points is taken as the vertical distance; the square root of the sum of the horizontal and vertical distances is taken as the positive signal's square root. Within the signal of the first node The data point and the negative signal Within the signal of the first node The matching distance between data points.

5. The withstand voltage test method for transformer production according to claim 1, characterized in that, The method for obtaining the degree of polarity difference between positive and negative signals on the node signal based on the phase difference and matching distance between data points in all matched pairs of positive and negative signals on the node signal includes the following specific methods: For any data point in any node signal of a positive signal, a local data is preset. The data point within the node signal of the positive signal that is closest to the data point within the node signal of the positive signal. A number of data points are used as local data points of the data points in the node signal of the positive signal. Hilbert transform calculation is performed on the local data points of the data points in the node signal of the positive signal to obtain the phase of the data points in the node signal of the positive signal. Obtain the phase of all data points in all node signals for both positive and negative signals; For positive and negative signals, the first The node signal will be compared with the positive signal at the 1st node. The absolute value of the phase difference between two data points in any matched pair of the n node signals is used as the positive signal and the negative signal in the nth node. The weight of the matching distance between two data points in the matching pair of the node signals; for positive and negative signals at the The weighted average of the matching distances between the two data points in all matching pairs of the nth node signal is used to obtain the positive and negative signals at the nth node. The degree of polarity difference on the signals of each node.

6. The withstand voltage test method for transformer production according to claim 1, characterized in that, The method for obtaining the true polarity difference between positive and negative signals at each node signal based on the amplitude and corresponding frequency range of the positive and negative signals at the same labeled node signal, combined with the degree of polarity difference between the positive and negative signals at the node signal, includes the following specific methods: For the positive signal The node signal will be the positive signal of the first node. The sum of the squares of the amplitudes of all data points in the nth node signal is taken as the nth node signal. The energy of the node signal; the positive signal of the first node. The energy of the nth node signal is divided by the sum of the energies of all node signals of the positive signal. This ratio is taken as the energy of the nth node signal of the positive signal. The energy factor of the node signal; the positive signal's first node signal The index label of the nth node signal is compared with the largest index label of all node signals of the positive signal. The resulting ratio is taken as the index label of the nth node signal of the positive signal. The index factor of the nth node signal will be the positive signal's nth node signal. The product of the energy factor and the index factor of the nth node signal is used as the first positive signal. The single-pole high-frequency factor of the signal at each node; The first step in obtaining the negative signal The single-pole high-frequency factor of the nth node signal will be the positive signal's nth node signal. The single-pole high-frequency factor of the nth node signal and the negative signal The product of the single-pole high-frequency factors of the n node signals is used as the first positive signal and the second negative signal. The high-frequency factor of the node signal; the positive signal and the negative signal of the first node. The high-frequency factors of the signal at each node and the positive and negative signals at the 1st node. The product of the polarity differences at each node signal is used as the positive and negative signals at the 1st node. The degree of true polarity difference on the signals at each node.

7. The withstand voltage test method for transformer production according to claim 1, characterized in that, The specific method for dividing the breakdown transient current signal into a rising segment and a falling segment, and obtaining the energy release rate of the breakdown transient current signal based on the rising segment and the falling segment, includes the following: The data point corresponding to the breakdown moment of the breakdown transient current signal is used as the segmentation point. The breakdown transient current signal is divided into rising segment and falling segment. The rising segment and falling segment are integrated respectively to obtain the integration results of the rising segment and the falling segment. The ratio between the integration result of the rising segment and the integration result of the falling segment is used as the energy release rate of the breakdown transient current signal.

8. The withstand voltage test method for transformer production according to claim 1, characterized in that, The specific method for obtaining the air gap breakdown factor is as follows: The product of the true polarity difference between the positive and negative signals at each node signal and the energy release rate of the breakdown transient current signal is used as the air gap breakdown characteristic of the positive and negative signals at each node signal. The sum of the air gap breakdown characteristics of the positive and negative signals at each node signal is used as the air gap breakdown factor.

9. The withstand voltage test method for transformer production according to claim 1, characterized in that, The specific method for obtaining the true probability of air gap breakdown based on the air pressure and air gap breakdown factor during the power frequency withstand voltage test includes: Obtain the worst-case air pressure, and record the absolute value of the difference between the air pressure during the power frequency withstand voltage test and the worst-case air pressure as the air pressure influence coefficient. Normalize the ratio of the air gap breakdown factor to the air pressure influence coefficient, and take the normalized result as the true probability of air gap breakdown.

10. A withstand voltage testing system for transformer production, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of a withstand voltage test method for transformer production as described in any one of claims 1-9.

Images