Discharge stage division method based on amplitude characteristics of electro-magnetic-acoustic multi-parameter signals

Abstract

The present invention belongs to the field of electrical equipment measurement technology, and particularly relates to a discharge stage division method based on the amplitude characteristics of electromagnetic-magnetic-acoustic multi-parameter signals, comprising: collecting parameter signals of a transformer, obtaining original data by recording the amplitude of the parameter signals during a surface discharge test; inputting the original data into a multi-parameter amplitude normalization model, and obtaining a multi-parameter amplitude normalization result M. n ; Obtain a normalized result data set, and construct a normalized value M based on the normalized result data set n Trend graph of changes over time; calculate normalized value M n The slope of each sampling point in the time-varying trend graph uses the slope division method to realize the stage division of the transformer surface discharge development process. The discharge stage division method can eliminate the influence of factors such as spatial position, measurement principle, and measurement sensitivity on the accuracy of a single measurement method; it fully reveals the essence of the fault development process and improves the reliability and rationality of the stage division.

Description

Technical Field

[0001] The invention belongs to the technical field of electrical equipment measurement, and in particular relates to a discharge stage division method based on the amplitude characteristics of an electric-magnetic-acoustic multi-parameter signal. Background Art

[0002] As an irreplaceable key link in the new power system, the health status of the transformer greatly affects the safe and reliable operation of the entire power grid. The internal insulation structure of engineering transformers is mainly oil-paper composite insulation. Due to moisture erosion, impurity intrusion, insulation material aging, component deformation and other reasons, the performance of the composite insulation material is degraded, the local electric field is distorted, and the discharge develops along the oil-paper interface. Under long-term operating conditions, the oil-paper composite insulation gradually deteriorates until it breaks down, causing the transformer box to rupture or even explode, causing huge economic losses and directly leading to the shutdown of part of the power grid. Therefore, clarifying the evolution law of the surface discharge development process of transformer oil-paper insulation, realizing reliable monitoring of characteristic parameters and accurate judgment of fault severity are key issues that need to be solved in the field of transformer online monitoring.

[0003] In the prior art, the research on the development process of surface discharge mainly focuses on the influence of different test conditions on the performance changes of insulating materials, breakdown voltage, etc., and the judgment of fault severity through a single characteristic parameter. There are relatively few studies on the stage division based on the law of multi-parameter changes. Since complex multi-physical fields will be generated during the development of insulation structure discharge faults, the perception method of a single parameter is affected by factors such as spatial position, measurement principle, and measurement sensitivity. It may not be able to respond to discharges at different stages in a timely and accurate manner, and thus cannot fully reveal the essence of the fault development process.

[0004] In summary, there is an urgent need for a discharge stage division method based on the amplitude characteristics of the electromagnetic-magnetic-acoustic multi-parameter signal. For the development process of oil-paper insulation discharge under AC voltage, it can eliminate the influence of factors such as spatial position, measurement principle, and measurement sensitivity on the accuracy of a single measurement method; it can fully reveal the essence of the fault development process, provide a reliable diagnostic basis for the development degree of transformer surface discharge faults, and improve the reliability and rationality of stage division. Summary of the invention

[0005] The object of the present invention is to provide a method for dividing discharge stages based on the amplitude characteristics of an electric-magnetic-acoustic multi-parameter signal, which is characterized by comprising the following steps:

[0006] Step S1: collecting parameter signals of the transformer, and obtaining original data by recording the amplitude of the parameter signals during the surface discharge test, wherein the parameter signals include: conventional pulse current signals, ultrasonic signals, high-frequency current signals, and ultra-high-frequency signals;

[0007] Step S2: Input the original data in step S1 into the multi-parameter amplitude normalization model to obtain the multi-parameter amplitude normalization result M n ;

[0008] Step S3: repeating steps S1 and S2 to obtain a normalized result data set, and constructing a multi-parameter amplitude fusion normalized curve based on the normalized result data set;

[0009] Step S4: Calculate the slope of each sampling point of the multi-parameter amplitude fusion normalized curve in step S3, and use the slope division method to realize the stage division of the development process of transformer surface discharge, wherein the discharge stage includes: small energy discharge stage, low energy discharge stage, high energy discharge stage, and near breakdown stage.

[0010] The multi-parameter amplitude normalization model in step S2 is:

[0011]

[0012] In the formula, a 1 , b 1 , c 1 ,q 1 are the initial amplitudes of UHF signal, ultrasonic signal, high frequency current signal and conventional pulse current signal respectively; a n , b n , c n ,q n They are respectively the parameter amplitudes of the ultra-high frequency signal, ultrasonic signal, high-frequency current signal and conventional pulse current signal of the nth data point.

[0013] The slope division method in step S4 includes:

[0014] When the slope is greater than or equal to 0 and the slope is less than or equal to 12.50, it is determined to be a small energy discharge stage;

[0015] When the slope is greater than or equal to 40.29 and the slope is less than or equal to 119.68, it is determined to be a low-energy discharge stage;

[0016] When the slope is greater than or equal to 260.90 and the slope is less than or equal to 484.49, it is determined to be a high-energy discharge stage;

[0017] When the slope is greater than or equal to 486.70, it is determined to be close to the breakdown stage.

[0018] The beneficial effects of the present invention are:

[0019] The present invention discloses a discharge stage division method based on the amplitude characteristics of an electric-magnetic-acoustic multi-parameter signal. For the development process of oil-paper insulation discharge under an alternating voltage, the method can eliminate the influence of factors such as spatial position, measurement principle, and measurement sensitivity on the accuracy of a single measurement method. The method can fully reveal the essence of the fault development process, provide a basis for stage division according to the law of multi-parameter changes, and provide a reliable diagnosis basis for the development degree of transformer surface discharge faults. Through the fusion judgment of the multi-parameter signal amplitude characteristics, the method enriches the information characterizing the discharge degree, avoids the disadvantage that a single information measurement may not be able to clearly characterize a certain stage of the discharge, and improves the reliability and rationality of the stage division. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of the method for dividing discharge stages based on the amplitude characteristics of the electric-magnetic-acoustic multi-parameter signal of the present invention;

[0021] Figure 2 It is a schematic diagram of the change curve of each parameter signal during the discharge process;

[0022] Figure 3 Schematic diagram of multi-parameter amplitude fusion normalization curve. DETAILED DESCRIPTION

[0023] The present invention provides a method for dividing discharge stages based on the amplitude characteristics of an electric-magnetic-acoustic multi-parameter signal. The present invention is further described in detail below in conjunction with the accompanying drawings.

[0024] like Figure 1 The embodiment of the present invention shown discloses a method for dividing discharge stages based on the amplitude characteristics of an electro-magnetic-acoustic multi-parameter signal, comprising the following steps:

[0025] Step S1: collecting parameter signals of the transformer, and obtaining original data by recording the amplitude of the parameter signals during the surface discharge test, wherein the parameter signals include: conventional pulse current signals, ultrasonic signals, high-frequency current signals, and ultra-high-frequency signals;

[0026] In this embodiment, the variation curves of each parameter signal during the discharge process are as follows: Figure 2 As shown, the parameter signals used are conventional pulse current signal, ultrasonic signal, high frequency current signal and ultra-high frequency signal. The original data are obtained by recording the amplitudes of the four signals measured during the surface discharge test.

[0027] Step S2: Input the original data in step S1 into the multi-parameter amplitude normalization model to obtain the multi-parameter amplitude normalization result M n ;

[0028] In this embodiment, the initial values ​​of the sensors collecting the four signals are obtained as reference values, the four parameter amplitudes of the data points (1, 2, 3…n) are taken as the actual values ​​representing the discharge, and the multi-parameter amplitude normalization result M is calculated by the multi-parameter amplitude normalization model. n .

[0029] The multi-parameter amplitude normalization model in step S2 is:

[0030]

[0031] In the formula, a 1 , b 1 , c 1 ,q 1 are the initial amplitudes of UHF signal, ultrasonic signal, high frequency current signal and conventional pulse current signal respectively; a n , b n , c n ,q n They are respectively the parameter amplitudes of the ultra-high frequency signal, ultrasonic signal, high-frequency current signal and conventional pulse current signal of the nth data point.

[0032] Step S3: repeating steps S1 and S2 to obtain a normalized result data set, and constructing a multi-parameter amplitude fusion normalized curve based on the normalized result data set;

[0033] In this embodiment, the discharge phase is divided according to the multi-parameter amplitude fusion normalization curve to form a normalized value M n Trend graphs over time;

[0034] Step S4: Calculate the normalized value M in step S3 n The slope of each sampling point in the time-varying trend graph uses the slope division method to realize the stage division of the development process of transformer surface discharge. The discharge stages include: small energy discharge stage, low energy discharge stage, high energy discharge stage, and near breakdown stage.

[0035] In this embodiment, the multi-parameter amplitude fusion normalization curve is as follows: Figure 3 As shown, the multi-parameter amplitude fusion normalized curve shows the normalized value M n The trend of change over time, analyzing the characteristics of the image, found that the slope has obvious changes, according to M n There are trends of slow increase, rapid increase, inflection point and surge, and the development process of transformer surface discharge is divided into small energy discharge stage, low energy discharge stage, high energy discharge stage and near breakdown stage.

[0036] The slope division method in step S4 includes:

[0037] When the slope is greater than or equal to 0 and the slope is less than or equal to 12.50, it is determined to be a small energy discharge stage;

[0038] When the slope is greater than or equal to 40.29 and the slope is less than or equal to 119.68, it is determined to be a low-energy discharge stage;

[0039] When the slope is greater than or equal to 260.90 and the slope is less than or equal to 484.49, it is determined to be a high-energy discharge stage;

[0040] When the slope is greater than or equal to 486.70, it is determined to be close to the breakdown stage.

[0041] In this embodiment, if Figure 3 As shown,

[0042] When the slope is greater than or equal to 0 and the slope is less than or equal to 12.50, the slope increases slowly and is determined to be a small energy discharge stage;

[0043] When the slope is greater than or equal to 40.29 and less than or equal to 119.68, the slope increases rapidly and is determined to be a low-energy discharge stage;

[0044] When the slope is greater than or equal to 260.90 and less than or equal to 484.49, the slope has an inflection point, which is determined to be a high-energy discharge stage;

[0045] When the slope is greater than or equal to 486.70, the slope increases sharply and is determined to be close to the breakdown stage.

[0046] In this embodiment, the fusion judgment method of multi-parameter signal amplitude characteristics enriches the information characterizing the discharge degree, avoids the disadvantage that a single information measurement may not be able to clearly characterize a certain stage of the discharge, and improves the reliability and rationality of the stage division.

[0047] In summary, the present invention discloses a discharge stage division method based on the amplitude characteristics of the electric-magnetic-acoustic multi-parameter signal, which divides the discharge stages in the development process of the oil-paper insulation surface discharge, collects the time domain waveform of the discharge spectrum using the corresponding sensor, takes the initial value of each sensor as the reference value, and takes the four parameter amplitudes of the data point (1, 2, 3...n) as the actual value representing the discharge; calculates the multi-parameter amplitude normalization result M through the multi-parameter amplitude normalization model n ,like Figure 3 As shown, according to M nThere are trends of slow increase, rapid increase, inflection point and surge. The development process of transformer surface discharge is divided into small energy discharge stage, low energy discharge stage, high energy discharge stage and near breakdown stage. The stage characteristics can be clearly judged according to the slope in the figure; the fusion judgment method of multi-parameter signal amplitude characteristics enriches the information characterizing the degree of discharge, avoids the disadvantage that single information measurement may not be able to clearly characterize a certain stage of discharge, improves the reliability and rationality of stage division, and can analyze the development of oil-paper insulation surface discharge from the perspective of joint detection of multi-parameter signals and diagnose the severity of faults, which makes up for the deficiency of most existing studies that judge the severity of faults by a single characteristic parameter, and provides a basis for stage division according to the law of multi-parameter changes, thereby laying a foundation for large-scale engineering applications of multi-parameter joint detection.

Claims

1. A method for dividing discharge stages based on the amplitude characteristics of electro-magnetic-acoustic multi-parameter signals, characterized in that: The steps include: Step S1: collecting parameter signals of the transformer, and obtaining original data by recording the amplitude of the parameter signals during the surface discharge test, wherein the parameter signals include: conventional pulse current signals, ultrasonic signals, high-frequency current signals, and ultra-high-frequency signals; Step S2: Input the original data in step S1 into the multi-parameter amplitude normalization model to obtain the multi-parameter amplitude normalization result M n ; Step S3: repeating steps S1 and S2 to obtain a normalized result data set, and constructing a multi-parameter amplitude fusion normalized curve based on the normalized result data set; Step S4: Calculate the slope of each sampling point of the multi-parameter amplitude fusion normalized curve in step S3, and use the slope division method to realize the stage division of the development process of transformer surface discharge, wherein the discharge stage includes: small energy discharge stage, low energy discharge stage, high energy discharge stage, and near breakdown stage; The multi-parameter amplitude normalization model in step S2 is: Where a1, b1, c1, q1 are the initial amplitudes of the UHF signal, ultrasonic signal, high-frequency current signal, and conventional pulse current signal respectively; a n , b n , c n ,q n They are respectively the parameter amplitudes of the ultra-high frequency signal, ultrasonic signal, high-frequency current signal and conventional pulse current signal of the nth data point.

2. The method for dividing discharge stages based on the amplitude characteristics of the electro-magnetic-acoustic multi-parameter signal according to claim 1 is characterized in that: The slope division method in step S4 includes: When the slope is greater than or equal to 0 and the slope is less than or equal to 12.50, it is determined to be a small energy discharge stage; When the slope is greater than or equal to 40.29 and the slope is less than or equal to 119.68, it is determined to be a low-energy discharge stage; When the slope is greater than or equal to 260.90 and the slope is less than or equal to 484.49, it is determined to be a high-energy discharge stage; When the slope is greater than or equal to 486.70, it is determined to be close to the breakdown stage.

Images