CN113848435A - Direct current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability - Google Patents
Direct current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability Download PDFInfo
- Publication number
- CN113848435A CN113848435A CN202111016060.9A CN202111016060A CN113848435A CN 113848435 A CN113848435 A CN 113848435A CN 202111016060 A CN202111016060 A CN 202111016060A CN 113848435 A CN113848435 A CN 113848435A
- Authority
- CN
- China
- Prior art keywords
- partial discharge
- energy distribution
- distribution probability
- voltage
- discharge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 66
- 238000001228 spectrum Methods 0.000 title claims abstract description 40
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000009467 reduction Effects 0.000 claims abstract description 16
- 238000009499 grossing Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000003595 spectral effect Effects 0.000 claims abstract 4
- 230000007547 defect Effects 0.000 claims description 28
- 238000009413 insulation Methods 0.000 claims description 19
- 238000005259 measurement Methods 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 15
- 238000010586 diagram Methods 0.000 claims description 10
- 230000008878 coupling Effects 0.000 claims description 9
- 238000010168 coupling process Methods 0.000 claims description 9
- 238000005859 coupling reaction Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 6
- 238000004088 simulation Methods 0.000 claims description 6
- 238000010183 spectrum analysis Methods 0.000 claims description 5
- 238000011160 research Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 9
- 230000005540 biological transmission Effects 0.000 description 7
- 238000002474 experimental method Methods 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005684 electric field Effects 0.000 description 2
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000011088 calibration curve Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000000752 ionisation method Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/16—Spectrum analysis; Fourier analysis
- G01R23/165—Spectrum analysis; Fourier analysis using filters
- G01R23/167—Spectrum analysis; Fourier analysis using filters with digital filters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
- G01R31/1227—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials
- G01R31/1263—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation
- G01R31/1272—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing of components, parts or materials of solid or fluid materials, e.g. insulation films, bulk material; of semiconductors or LV electronic components or parts; of cable, line or wire insulation of cable, line or wire insulation, e.g. using partial discharge measurements
Abstract
A direct current partial discharge signal classification and identification method based on spectral energy distribution probability sequentially comprises a model building step, a data acquisition step, a smoothing and noise reduction processing step, a frequency spectrum division step and a distribution probability solving step. The method has the beneficial effect that the partial discharge under the needle plate models with different diameters can be identified by drawing the ternary energy probability distribution map. The research aims to lay a foundation for identifying the discharge type of the power equipment under the direct-current voltage and evaluating the discharge hazard.
Description
Technical Field
The invention relates to the field of high-voltage direct-current power transmission, in particular to a direct-current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability.
Background
The high-voltage direct-current transmission has the characteristics of large transmission capacity, long transmission distance, capability of quickly controlling and adjusting the size and direction of transmitted power, small influence on the environment and the like, and is particularly suitable for relieving the current situation of imbalance between the electricity load and the power generation energy distribution in China. At present, the highest voltage grade of high-voltage direct-current transmission in China is +/-800 kV, and the ultrahigh-voltage direct-current transmission project of Jinbei-Jiangsu +/-800 kV which is constructed by starting work in recent years can make the ultrahigh-voltage direct-current transmission project in China more developed.
Under the action of long-term high voltage, the problems of insulation performance reduction, partial discharge and the like of electrical equipment inevitably occur. The partial discharge is not only a result of insulation aging of the electric power equipment but also a significant cause of the insulation aging, which is closely related to defects inside the insulation. As a parameter for effectively monitoring the insulation condition of electrical equipment, the research of partial discharge under the action of direct current voltage is particularly important.
The research on the high-voltage direct-current partial discharge is far behind the alternating current for a long time. Since the end of the 20 th century, the Delft industry university, the netherlands, performed a series of studies on partial discharge under high voltage direct current, starting with physical analysis of partial discharge under direct voltage and applying the partial discharge analysis to the detection of partial discharge of capacitors and high voltage direct current cables. Fromm et al developed a mathematical model of the internal air gap repetitive discharge at dc voltage in 1995, proposing the discharge amplitude, the number of discharge pulses, and the time interval between adjacent discharges as the most basic parameters of partial discharge at dc; p.h.f.morshuis in 2004 summarized the work of Delft industrial university about high voltage dc partial discharge for over a decade, explaining to some extent the generation and development process of dc partial discharge.
The first problem of the direct current partial discharge research is the identification of defects due to the different damage degree of the defects with different properties to the equipment insulation.
Disclosure of Invention
The invention aims to solve the problems and designs a direct current partial discharge signal classification and identification method based on spectrum energy distribution probability. The specific design scheme is as follows:
a direct current partial discharge signal classification identification method based on frequency spectrum energy distribution probability sequentially carries out a model building step, a data acquisition step, a smoothing noise reduction processing step, a frequency spectrum division step and a distribution probability obtaining step, wherein the model building step comprises a high-voltage direct current power supply and direct current partial discharge simulation and detection model; in the data acquisition step, a standard pulse generator is used for carrying out pulse discharge amplitude calibration on the detection system, and the amplitude of a signal displayed by the measurement system is read; the smoothing noise reduction processing part is used for carrying out noise reduction on an original partial discharge pulse waveform acquired by the high-frequency coupling current sensor by adopting a Savitzky-Golay algorithm, and in the frequency spectrum division step, the discharge type is classified and identified by adopting the frequency characteristic of a direct-current partial discharge pulse waveform; in the step of solving the distribution probability, the energy distribution probability of each frequency band is calculated and drawn in an energy distribution probability chart.
The high-voltage direct-current power supply and the direct-current partial discharge simulation and detection model comprise a high-voltage direct-current power supply, a protective resistor, a resistor voltage divider, an oil paper insulation defect model, a vacuum pump, a high-frequency coupling current sensor and an oscilloscope, one end of the high-voltage direct current power supply is connected with one end of a protection resistor, the other end of the protection resistor and one fixed end of a resistor voltage divider are connected with one end of the oiled paper insulation defect model, the other fixed end of the resistor voltage divider and the other end of the high-voltage direct current power supply are connected with a reference ground, the sliding end of the resistor voltage divider is connected with the input end of an oscilloscope, the output end of the oscilloscope and the other end of the oiled paper insulation defect model are connected with a high-frequency coupling current sensor, the interior of the oiled paper insulation defect model is communicated with a vacuum pump, the whole experiment is completed in a shielding chamber, direct-current voltage is generated by adopting a voltage doubling circuit, and the detection of partial discharge is realized by using a partial discharge detector developed by Italy TechImp company.
In the data acquisition step, firstly, a standard pulse generator is utilized to calibrate the pulse discharge capacity amplitude of the detection system, a pulse signal source with known and adjustable discharge capacity injects charges into a high-voltage electrode in a pulse mode, the gain of the partial discharge measurement system is adjusted to be maximum, a trigger value is adjusted to the pulse current signal level, then the amplitude of the pulse current signal is gradually increased until the measurement system can obtain the pulse current signal, the amplitude of the output signal of the pulse source is continuously increased, and the amplitude of the signal displayed by the measurement system is read.
The discharge amplitude-discharge intensity relation obtained by fitting is that Q is 0.77477+0.06794U-3.81892 multiplied by 10- 5U2。
Under direct current voltage, the voltage applied to two ends of a test article when partial discharge occurs is defined as partial discharge initial voltage, the direct current partial discharge has no regular repeatability, so that the discharge frequency per minute needs to be specified, the internationally recognized discharge frequency per minute is 1, namely if a step voltage boosting method is adopted in an experiment, each voltage is kept for 10min, then the discharge of more than 10 times occurs within 10min, the external application voltage at the moment is considered as partial discharge initial voltage, and according to the method, the step voltage boosting method is adopted to determine the direct current partial discharge initial voltage of the test article.
Because the partial discharge pulse waveform needs to be subjected to spectrum analysis, and the energy distribution probability of different frequency band waveforms is obtained on the basis of a spectrum function, the accurate extraction of the partial discharge waveform is a premise of all work, but in an experiment, because of the existence of environmental micro-vibration and weak electromagnetic interference, the signal waveform inevitably contains noise and is weak, so that the original signal needs to be preprocessed firstly, effective noise elimination and smoothing of data are particularly important, in the smoothing and noise reduction processing step, a group of 2M +1 data taking n as the center is considered, and the following polynomial is used for fitting
The residual of the least squares fit is:
the constant term of the fitting polynomial is obtained by convolution operation, namely the input data is weighted and averaged:
the partial derivative is calculated for the above formula, which includes:
after simplification, the following can be obtained:
where i is 0,1, N, let a be { a ═ ani},ani＝ni,-M≤n≤M,0≤i≤N,B＝ATA, then:
Ba＝AT·Aa＝ATx (1.7)
a＝(AT·A)-1·ATx＝Hx (1.8)
h is the convolution coefficient to be obtained,
the S-G algorithm is adopted to carry out smooth noise reduction on the pulse waveform, the selected fitting polynomial is a 5-degree polynomial, a single group of fitting data is 20 data points, the pulse waveform obtained through fitting is shown in figure 5, and compared with the waveforms in figures 4 and 5, the S-G algorithm has a good filtering effect and can well remove noise.
In the step of dividing the frequency spectrum, the prior art has the defects that a single type of partial discharge occurs under the partial discharge starting voltage, the discharge pulse waveforms of each time are similar, and the discharge pulse waveforms of different defects are relatively close, as shown in fig. 6, so that the analysis and division of the discharge waveforms are difficult directly from the angle of the time domain, but the time domain pulse waveforms which are similar in appearance and are generated by different defect types are obviously different in the frequency domain, wherein the frequency range of the needle plate model with the higher partial discharge amplitude is 0-50MHz, and some energy distribution exists between 50MHz-150 MHz; the partial discharge energy of the internal air gap model is almost completely distributed within 0-50MHz, and other frequency ranges are not distributed; partial discharge energy is mainly distributed in the range of 50MHz-150MHz along the surface model, and is also distributed in the ranges of 0-50MHz and 150MHz-200MHz, and based on the results, the discharge types are classified and identified by adopting the frequency characteristics of direct current partial discharge pulse waveforms. In order to realize statistical analysis and to classify and identify the discharge types conveniently and quickly, 0-200MHz is divided into 3 frequency bands, namely, a low frequency band 0-50MHz, a middle frequency band 50-150 MHz and a 150MHz-200MHz, as shown in fig. 7, then for the partial discharge pulse waveform, the energy distribution probability of each frequency band can be calculated by the formula (1.9):
from Parseval energy identity, there is another expression form of equation (1.9), i.e. using a frequency spectrum function to calculate by digital-to-analog square, as shown in equation (1.10):
wherein R isL、RBAnd RHLow, medium and high band energy distribution probability, u, respectivelyL、uBAnd uHFor each frequency band, u is the full-band voltage amplitude, XL、XBAnd XHAnd (3) obtaining the modulus of each frequency band pulse spectrum function, wherein X is the modulus of a full-frequency band spectrum function, delta T is sampling interval time, N is sampling depth, and R is measurement loop resistance.
In the step of obtaining the distribution probability, the energy distribution probabilities α, β, γ of each frequency band of the partial discharge pulse waveform can be obtained by calculation according to the formula (1.10), the sum of the three is known to be 100% by the calculation formula, and the analysis shows that the energy distribution probabilities of each frequency band of the partial discharge generated by different defect types under the direct current voltage are different, and the probabilities α, β, γ are plotted in a Ternary parameter diagram (Ternary Plot) for visual analysis and convenient future identification of the discharge type by using a cluster analysis method, specifically as illustrated in fig. 8, the probability distribution of each point in the upper diagram is: a (0, 0, 100%), B (0, 100%, 0), C (100%, 0, 0), D (25%, 50%, 25%), carrying out spectrum analysis on partial discharge pulses of each defect type generated under the partial discharge initial voltage, obtaining energy distribution probability of each frequency band, drawing the energy distribution probability in an energy distribution probability chart, and then conveniently and intuitively classifying and identifying the discharge types.
The direct current partial discharge signal classification and identification method based on the frequency spectrum energy distribution probability, which is obtained by the technical scheme of the invention, has the beneficial effects that:
partial discharge under different diameter needle plate models can be identified by drawing a ternary energy probability distribution diagram. The research aims to lay a foundation for identifying the discharge type of the power equipment under the direct-current voltage and evaluating the discharge hazard.
Drawings
FIG. 1 is a schematic diagram of an experimental loop according to the present invention;
FIG. 2 is a calibration curve of the pulse current measurement system of the present invention;
FIG. 3 is a schematic illustration of an experimental electrode according to the present invention;
FIG. 4 is a graph of a partial discharge pulse waveform prior to smoothing in accordance with the present invention;
FIG. 5 is a graph of a partial discharge pulse waveform after smoothing in accordance with the present invention;
FIG. 6 is a graph of partial discharge pulse waveforms for different defect types according to the present invention;
fig. 7 is a schematic diagram of the frequency spectrum division of the partial discharge pulse according to the present invention;
FIG. 8 is a schematic diagram of the probability distribution of the partial discharge pulse spectrum energy according to the present invention;
FIG. 9 is a graph of partial discharge starting voltage and maximum field strength of needle plate electrodes of different diameters according to the present invention;
FIG. 10 is a graph of partial discharge pulse amplitudes of needle plate electrodes of different diameters at a partial discharge starting voltage according to the present invention;
FIG. 11a1 is a waveform of a 0.5mm tip diameter partial discharge pulse according to the present invention;
FIG. 11a2 is a plot of partial discharge pulse amplitude versus spectrum for a tip diameter of 0.5mm in accordance with the present invention;
FIG. 11b1 is a waveform of a 1.0mm tip diameter partial discharge pulse according to the present invention;
FIG. 11b2 is a graph of partial discharge pulse amplitude versus spectrum for a tip diameter of 1.0mm in accordance with the present invention;
FIG. 11c1 is a waveform of a 2.0mm tip diameter partial discharge pulse according to the present invention;
FIG. 11c2 is a graph of partial discharge pulse amplitude versus spectrum for a 2.0mm tip diameter according to the present invention;
FIG. 11d1 is a waveform of a 4.0mm tip diameter partial discharge pulse according to the present invention;
FIG. 11d2 is a graph of partial discharge pulse amplitude versus spectrum for a 4.0mm tip diameter according to the present invention;
FIG. 12 is a graph of the probability of distribution of the frequency spectrum energy of the electrode of the faller bars with different diameters under the partial discharge starting voltage;
in the figure, 1, a high-voltage direct-current power supply; 2. a protection resistor; 3. a resistive voltage divider; 4. oil paper insulation defect model; 5. a vacuum pump; 6. a high frequency coupled current sensor; 7. an oscilloscope.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings.
A direct current partial discharge signal classification identification method based on frequency spectrum energy distribution probability sequentially carries out a model building step, a data acquisition step, a smoothing noise reduction processing step, a frequency spectrum division step and a distribution probability obtaining step, wherein the model building step comprises a high-voltage direct current power supply and direct current partial discharge simulation and detection model; in the data acquisition step, a standard pulse generator is used for carrying out pulse discharge amplitude calibration on the detection system, and the amplitude of a signal displayed by the measurement system is read; the smoothing noise reduction processing part is used for carrying out noise reduction on an original partial discharge pulse waveform acquired by the high-frequency coupling current sensor by adopting a Savitzky-Golay algorithm, and in the frequency spectrum division step, the discharge type is classified and identified by adopting the frequency characteristic of a direct-current partial discharge pulse waveform; in the step of solving the distribution probability, the energy distribution probability of each frequency band is calculated and drawn in an energy distribution probability chart.
The high-voltage direct-current power supply and direct-current partial discharge simulation and detection model comprises a high-voltage direct-current power supply 1, a protection resistor 2, a resistor divider 3, an oil paper insulation defect model 4, a vacuum pump 5, a high-frequency coupling current sensor 6 and an oscilloscope 7, wherein one end of the high-voltage direct-current power supply 1 is connected with one end of the protection resistor 2, the other end of the protection resistor 2 and one fixed end of the resistor divider 3 are connected with one end of the oil paper insulation defect model 4, the other fixed end of the resistor divider 3 and the other end of the high-voltage direct-current power supply 1 are connected with a reference ground, the sliding end of the resistor divider 3 is connected with the input end of the oscilloscope 7, the output end of the oscilloscope 7 and the other end of the oil paper insulation defect model 4 are connected with the high-frequency coupling current sensor 6, the interior of the oil paper insulation defect model 4 is communicated with the vacuum pump 5, the whole experiment is completed in a shielding chamber, and direct-current voltage is generated by adopting a voltage doubling circuit, the detection of partial discharge was carried out using a partial discharge detector developed by techmip, italy.
In the data acquisition step, firstly, a standard pulse generator is utilized to calibrate the pulse discharge capacity amplitude of the detection system, a pulse signal source with known and adjustable discharge capacity injects charges into a high-voltage electrode in a pulse mode, the gain of the partial discharge measurement system is adjusted to be maximum, a trigger value is adjusted to the pulse current signal level, then the amplitude of the pulse current signal is gradually increased until the measurement system can obtain the pulse current signal, the amplitude of the output signal of the pulse source is continuously increased, and the amplitude of the signal displayed by the measurement system is read.
The discharge amplitude-discharge intensity relation obtained by fitting is that Q is 0.77477+0.06794U-3.81892 multiplied by 10- 5U2。
Under direct current voltage, the voltage applied to two ends of a test article when partial discharge occurs is defined as partial discharge initial voltage, the direct current partial discharge has no regular repeatability, so that the discharge frequency per minute needs to be specified, the internationally recognized discharge frequency per minute is 1, namely if a step voltage boosting method is adopted in an experiment, each voltage is kept for 10min, then the discharge of more than 10 times occurs within 10min, the external application voltage at the moment is considered as partial discharge initial voltage, and according to the method, the step voltage boosting method is adopted to determine the direct current partial discharge initial voltage of the test article.
Because the partial discharge pulse waveform needs to be subjected to spectrum analysis, and the energy distribution probability of different frequency band waveforms is obtained on the basis of a spectrum function, accurate extraction of the partial discharge waveform is a premise of all work, but in an experiment, because of the existence of environmental micro-vibration and weak electromagnetic interference, a signal waveform inevitably contains noise and is weak, an original signal needs to be preprocessed firstly, effective noise elimination and smoothing of data are particularly important, and in the smoothing noise reduction processing step, a group of 2M +1 data with n-0 as a center are considered, and fitting is carried out by using the following polynomials:
the residual of the least squares fit is:
the constant term of the fitting polynomial is obtained by convolution operation, namely the input data is weighted and averaged:
the partial derivative is calculated for the above formula, which includes:
after simplification, the following can be obtained:
where i is 0,1, N, let a be { a ═ ani},ani＝ni,-M≤n≤M,0≤i≤N,B＝ATA, then:
Ba＝AT·Aa＝ATx (1.7)
a＝(AT·A)-1·ATx＝Hx (1.8)
h is the convolution coefficient to be obtained,
the S-G algorithm is adopted to carry out smooth noise reduction on the pulse waveform, the selected fitting polynomial is a 5-degree polynomial, a single group of fitting data is 20 data points, the pulse waveform obtained through fitting is shown in figure 5, and compared with the waveforms in figures 4 and 5, the S-G algorithm has a good filtering effect and can well remove noise.
In the step of dividing the frequency spectrum, the prior art has the defects that a single type of partial discharge occurs under the partial discharge starting voltage, the discharge pulse waveforms of each time are similar, and the discharge pulse waveforms of different defects are relatively close, as shown in fig. 6, so that the analysis and division of the discharge waveforms are difficult directly from the angle of the time domain, but the time domain pulse waveforms which are similar in appearance and are generated by different defect types are obviously different in the frequency domain, wherein the frequency range of the needle plate model with the higher partial discharge amplitude is 0-50MHz, and some energy distribution exists between 50MHz-150 MHz; the partial discharge energy of the internal air gap model is almost completely distributed within 0-50MHz, and other frequency ranges are not distributed; partial discharge energy is mainly distributed in the range of 50MHz-150MHz along the surface model, and is also distributed in the ranges of 0-50MHz and 150MHz-200MHz, and based on the results, the discharge types are classified and identified by adopting the frequency characteristics of direct current partial discharge pulse waveforms. In order to realize statistical analysis and to classify and identify the discharge types conveniently and quickly, 0-200MHz is divided into 3 frequency bands, namely, a low frequency band 0-50MHz, a middle frequency band 50-150 MHz and a 150MHz-200MHz, as shown in fig. 7, then for the partial discharge pulse waveform, the energy distribution probability of each frequency band can be calculated by the formula (1.9):
from Parseval energy identity, there is another expression form of equation (1.9), i.e. using a frequency spectrum function to calculate by digital-to-analog square, as shown in equation (1.10):
wherein R isL、RBAnd RHLow, medium and high band energy distribution probability, u, respectivelyL、uBAnd uHFor each frequency band, u is the full-band voltage amplitude, XL、XBAnd XHAnd (3) obtaining the modulus of each frequency band pulse spectrum function, wherein X is the modulus of a full-frequency band spectrum function, delta T is sampling interval time, N is sampling depth, and R is measurement loop resistance.
In the step of obtaining the distribution probability, the energy distribution probabilities α, β, γ of each frequency band of the partial discharge pulse waveform can be obtained by calculation according to the formula (1.10), the sum of the three is known to be 100% by the calculation formula, and the analysis shows that the energy distribution probabilities of each frequency band of the partial discharge generated by different defect types under the direct current voltage are different, and for visual analysis and convenience of identifying the discharge type by a cluster analysis method in the future, the probabilities α, β, γ are plotted in a Ternary parameter diagram Ternary Plot, as shown in fig. 8 specifically, the probability distributions of each point in the upper diagram are: a (0, 0, 100%), B (0, 100%, 0), C (100%, 0, 0), D (25%, 50%, 25%), carrying out spectrum analysis on partial discharge pulses of each defect type generated under the partial discharge initial voltage, obtaining energy distribution probability of each frequency band, drawing the energy distribution probability in an energy distribution probability chart, and then conveniently and intuitively classifying and identifying the discharge types.
Example 1 partial discharge initiation of the electrode of the needle plate at different tip diameters.
The step boosting method is adopted in the experiment, the partial discharge initial voltage of the needle plate electrode structure when the diameters of the needle electrodes are different is determined, the Ansoft simulation software is used for calculating the internal maximum field intensity of the model under the partial discharge initial voltage, and the experimental results are shown in the following table and fig. 9.
It can be seen from the experimental results that as the diameter of the needle electrode increases, the partial discharge onset voltage increases due to the more and more uniform electric field in the gap. On the other hand, however, as the diameter of the electrode increases, the maximum field strength in the gap at the partial discharge start voltage decreases. The analysis shows that the phenomenon is caused by that the effective action area of the electrode surface is increased after the diameter of the needle electrode is increased, and the strong field area is more widely distributed, so that ionization can occur in a larger range, and discharge does not depend on overhigh electric field intensity. In addition, as the diameter of the electrode increases, the dispersion of the partial discharge starting voltage is larger, and the phenomenon is that the field intensity of the needle tip is reduced under the large diameter, so that the ionization process is more random, and the instability of the partial discharge process is further caused. This process also results in a large dispersion of the partial discharge pulse amplitude for large diameter electrode structures at the partial discharge start voltage, as shown in fig. 10. The larger the diameter of the needle electrode is, the higher the amplitude of the partial discharge pulse under the partial discharge starting voltage is, and the larger the dispersibility is. Considering the equation (1.11) for calculating the apparent discharge amount when the gap is partially discharged, it can be seen that, assuming that the capacitance between the gaps is not changed, the larger the diameter of the needle electrode is, the larger the voltage drop across the gap is when the partial discharge occurs, and therefore, the larger the apparent discharge amount is.
Q＝CxΔU
Wherein Q is the apparent discharge amount when partial discharge occurs in the gap, CxFor the inter-gap capacitance, Δ U is the voltage change over the gap when the discharge occurs.
Example 2 partial discharge pulse frequency spectrum and energy distribution probability chart of needle plate electrode under different needle tip diameters.
Fig. 11a1 to 11d2 are graphs of average waveforms and corresponding amplitudes-spectra of partial discharge pulses of faller electrode electrodes with different diameters at the partial discharge starting voltage, respectively, measured experimentally. In fact, when the probability of the discharge energy distribution in each frequency band is obtained and the energy probability distribution map is drawn, the partial discharge pulse waveforms at all partial discharge starting voltages are processed, and the final processing result is shown in fig. 12.
As can be seen from FIG. 12, as the diameter of the needle electrode increases, the energy of the middle frequency band of the partial discharge pulse increases, and the energy of the low frequency band decreases, but the discharge energy distribution is concentrated in the range (not less than 50%, not more than 10%).
The technical solutions described above only represent the preferred technical solutions of the present invention, and some possible modifications to some parts of the technical solutions by those skilled in the art all represent the principles of the present invention, and fall within the protection scope of the present invention.
Claims (8)
1. A direct current partial discharge signal classification identification method based on frequency spectrum energy distribution probability sequentially carries out a model building step, a data acquisition step, a smoothing noise reduction processing step, a frequency spectrum division step and a distribution probability solving step, and is characterized in that the model building step comprises a high-voltage direct current power supply and direct current partial discharge simulation and detection model; in the data acquisition step, a standard pulse generator is used for carrying out pulse discharge amplitude calibration on the detection system, and the amplitude of a signal displayed by the measurement system is read; the smoothing noise reduction processing part is used for carrying out noise reduction on an original partial discharge pulse waveform acquired by the high-frequency coupling current sensor by adopting a Savitzky-Golay algorithm, and in the frequency spectrum division step, the discharge type is classified and identified by adopting the frequency characteristic of a direct-current partial discharge pulse waveform; in the step of solving the distribution probability, the energy distribution probability of each frequency band is calculated and drawn in an energy distribution probability chart.
2. The direct current partial discharge signal classification and identification method based on the frequency spectrum energy distribution probability is characterized in that the high-voltage direct current power supply and the direct current partial discharge simulation and detection model comprise a high-voltage direct current power supply (1), a protection resistor (2), a resistor divider (3), an oil paper insulation defect model (4), a vacuum pump (5), a high-frequency coupling current sensor (6) and an oscilloscope (7), one end of the high-voltage direct current power supply (1) is connected with one end of the protection resistor (2), the other end of the protection resistor (2) and one fixed end of the resistor divider (3) are connected with one end of the oil paper insulation defect model (4), the other fixed end of the resistor divider (3) and the other end of the high-voltage direct current power supply (1) are connected with a reference ground, and the sliding end of the resistor divider (3) is connected with the input end of the oscilloscope (7), the output end of the oscilloscope (7) and the other end of the oil paper insulation defect model (4) are connected with a high-frequency coupling current sensor (6), and the interior of the oil paper insulation defect model (4) is communicated with the vacuum pump (5).
3. The method for classifying and identifying the DC partial discharge signal based on the probability of spectral energy distribution as claimed in claim 1, wherein in the data acquisition step, a standard pulse generator is used to calibrate the amplitude of the pulse discharge of the detection system, a pulse signal source with known and adjustable discharge injects charges into the high voltage electrode in the form of pulses, the gain of the partial discharge measurement system is adjusted to the maximum, the trigger value is adjusted to the level of the pulse current signal, then the amplitude of the pulse current signal is gradually increased until the measurement system can obtain the pulse current signal, the amplitude of the output signal of the pulse source is continuously increased, and the amplitude of the signal displayed by the measurement system is read.
4. The method as claimed in claim 1, wherein the fitted relation of discharge amplitude versus discharge intensity is 0.77477+0.06794U-3.81892 x 10-5U2。
5. The method according to claim 1, wherein the dc partial discharge signal classification and identification method based on spectral energy distribution probability is characterized in that, under dc voltage, the voltage applied to two ends of the sample when partial discharge occurs is defined as the partial discharge starting voltage, and the dc partial discharge starting voltage of the sample is determined by using a step-up boosting method.
6. The method for classifying and identifying DC partial discharge signals based on spectral energy distribution probability as claimed in claim 1, wherein in the step of smoothing and noise reduction, a set of 2M +1 data with n-0 as the center is considered, and the following polynomial is used for fitting
The residual of the least squares fit is:
the constant term of the fitting polynomial is obtained by convolution operation, namely the input data is weighted and averaged:
the partial derivative is calculated for the above formula, which includes:
after simplification, the following can be obtained:
where i is 0,1, N, let a be { a ═ ani},ani＝ni,-M≤n≤M,0≤i≤N,B＝ATA, then:
Ba＝AT·Aa＝ATx (1.7)
a＝(AT·A)-1·ATx＝Hx (1.8)
h is the convolution coefficient.
7. The method for classifying and identifying dc partial discharge signals according to claim 1, wherein in the step of dividing the frequency spectrum, for the partial discharge pulse waveform, the energy distribution probability of each frequency band can be calculated by equation (1.9):
from Parseval energy identity, there is another expression form of equation (1.9), i.e. using a frequency spectrum function to calculate by digital-to-analog square, as shown in equation (1.10):
wherein R isL、RBAnd RHLow, medium and high band energy distribution probability, u, respectivelyL、uBAnd uHFor each frequency band, u is the full-band voltage amplitude, XL、XBAnd XHAnd (3) obtaining the modulus of each frequency band pulse spectrum function, wherein X is the modulus of a full-frequency band spectrum function, delta T is sampling interval time, N is sampling depth, and R is measurement loop resistance.
8. The method for classifying and identifying the dc partial discharge signal based on the spectrum energy distribution probability as claimed in claim 7, wherein in the step of obtaining the distribution probability, the energy distribution probabilities α, β, γ of the partial discharge pulse waveform in each frequency band are obtained by calculating according to the formula (1.10), and the probabilities α, β, γ are plotted in a Ternary parameter diagram (Ternary Plot), where the probability distribution of each point in the diagram is: a (0, 0, 100%), B (0, 100%, 0), C (100%, 0, 0), D (25%, 50%, 25%), carrying out spectrum analysis on partial discharge pulses of each defect type generated under the partial discharge initial voltage, obtaining energy distribution probability of each frequency band, drawing the energy distribution probability in an energy distribution probability chart, and then conveniently and intuitively classifying and identifying the discharge types.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111016060.9A CN113848435B (en) | 2021-08-31 | 2021-08-31 | Direct-current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111016060.9A CN113848435B (en) | 2021-08-31 | 2021-08-31 | Direct-current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113848435A true CN113848435A (en) | 2021-12-28 |
| CN113848435B CN113848435B (en) | 2023-12-15 |
Family
ID=78976556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111016060.9A Active CN113848435B (en) | 2021-08-31 | 2021-08-31 | Direct-current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113848435B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2009278834A1 (en) * | 2008-08-06 | 2010-02-11 | Eskom Holdings Limited | Partial discharge monitoring method and system |
| JP2011215067A (en) * | 2010-04-01 | 2011-10-27 | Hitachi Ltd | Insulation diagnosis method, insulation diagnosis system, and rotating electric machine |
| US20130179100A1 (en) * | 2012-01-11 | 2013-07-11 | Utilx Corporation | System for analyzing and locating partial discharges |
| CN103257306A (en) * | 2012-10-25 | 2013-08-21 | 国网电力科学研究院武汉南瑞有限责任公司 | Method for diagnosing direct current partial discharging insulation state of converter transformer and measurement system |
| US20150120218A1 (en) * | 2010-05-31 | 2015-04-30 | Universidad Politecnica De Madrid | Novel method for real time tests and diagnosis of partial discharge sources in high voltage equipment and installations, which are in service or out of service, and physical system for the practical use of the method |
| KR101553005B1 (en) * | 2015-04-20 | 2015-10-01 | 지투파워 (주) | A partial discharge monitoring and diagnosis system for power devices |
| CN106503612A (en) * | 2016-09-12 | 2017-03-15 | 国网浙江省电力公司舟山供电公司 | Direct current cables local discharge signal mode identification method and system based on compressive sensing theory |
| CN108761287A (en) * | 2018-05-28 | 2018-11-06 | 深圳供电局有限公司 | A kind of partial discharge of transformer kind identification method |
| CN109164362A (en) * | 2018-09-25 | 2019-01-08 | 上海交通大学 | A kind of recognition methods and system of direct current cables shelf depreciation defect failure |
| JP2019045401A (en) * | 2017-09-06 | 2019-03-22 | 株式会社明電舎 | Partial discharge diagnostic device and partial discharge diagnostic method |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
| CN110824310A (en) * | 2019-11-21 | 2020-02-21 | 深圳供电局有限公司 | Partial discharge type judgment method and device |
| CN111929553A (en) * | 2020-10-19 | 2020-11-13 | 四川大学 | Partial discharge positioning method based on phase velocity frequency-varying characteristics |
| CN112083298A (en) * | 2020-09-08 | 2020-12-15 | 四川大学 | Oil-immersed insulating paperboard partial discharge failure probability characterization method based on double-risk competition model |
| CN112147474A (en) * | 2020-10-30 | 2020-12-29 | 广东电网有限责任公司阳江供电局 | XLPE power cable typical defect partial discharge type identification system and method |
-
2021
- 2021-08-31 CN CN202111016060.9A patent/CN113848435B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2009278834A1 (en) * | 2008-08-06 | 2010-02-11 | Eskom Holdings Limited | Partial discharge monitoring method and system |
| JP2011215067A (en) * | 2010-04-01 | 2011-10-27 | Hitachi Ltd | Insulation diagnosis method, insulation diagnosis system, and rotating electric machine |
| US20150120218A1 (en) * | 2010-05-31 | 2015-04-30 | Universidad Politecnica De Madrid | Novel method for real time tests and diagnosis of partial discharge sources in high voltage equipment and installations, which are in service or out of service, and physical system for the practical use of the method |
| US20130179100A1 (en) * | 2012-01-11 | 2013-07-11 | Utilx Corporation | System for analyzing and locating partial discharges |
| CN103257306A (en) * | 2012-10-25 | 2013-08-21 | 国网电力科学研究院武汉南瑞有限责任公司 | Method for diagnosing direct current partial discharging insulation state of converter transformer and measurement system |
| KR101553005B1 (en) * | 2015-04-20 | 2015-10-01 | 지투파워 (주) | A partial discharge monitoring and diagnosis system for power devices |
| CN106503612A (en) * | 2016-09-12 | 2017-03-15 | 国网浙江省电力公司舟山供电公司 | Direct current cables local discharge signal mode identification method and system based on compressive sensing theory |
| JP2019045401A (en) * | 2017-09-06 | 2019-03-22 | 株式会社明電舎 | Partial discharge diagnostic device and partial discharge diagnostic method |
| CN108761287A (en) * | 2018-05-28 | 2018-11-06 | 深圳供电局有限公司 | A kind of partial discharge of transformer kind identification method |
| CN109164362A (en) * | 2018-09-25 | 2019-01-08 | 上海交通大学 | A kind of recognition methods and system of direct current cables shelf depreciation defect failure |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
| CN110824310A (en) * | 2019-11-21 | 2020-02-21 | 深圳供电局有限公司 | Partial discharge type judgment method and device |
| CN112083298A (en) * | 2020-09-08 | 2020-12-15 | 四川大学 | Oil-immersed insulating paperboard partial discharge failure probability characterization method based on double-risk competition model |
| CN111929553A (en) * | 2020-10-19 | 2020-11-13 | 四川大学 | Partial discharge positioning method based on phase velocity frequency-varying characteristics |
| CN112147474A (en) * | 2020-10-30 | 2020-12-29 | 广东电网有限责任公司阳江供电局 | XLPE power cable typical defect partial discharge type identification system and method |
Non-Patent Citations (3)
| Title |
|---|
| 李蓉 等: "基于输入阻抗谱的电力电缆本体局部缺陷类型识别及定位", 《电工技术学报》, vol. 36, no. 8, pages 1743 - 1751 * |
| 秦冰阳;司雯;李斯盟;李岩;梁园斌;亓占华;: "局部放电检测用D-dot探头的可行性研究", 电测与仪表, no. 06, pages 108 - 112 * |
| 陈峰 等: "基于小波包分解和支持向量机的局部放电识别方法研究", 《山东电力技术》, vol. 47, no. 271, pages 5 - 9 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113848435B (en) | 2023-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Forssén et al. | Partial discharges in a cavity at variable applied frequency part 1: measurements | |
| Esfahani et al. | Investigation of corona partial discharge characteristics under variable frequency and air pressure | |
| CN107607837A (en) | A kind of cable insulation degree of aging method of testing and device based on surge voltage | |
| Chen et al. | Comparison of methods for determining corona inception voltages of transmission line conductors | |
| Wang et al. | UHF PD measurement system with scanning and comparing method | |
| CN108548996B (en) | Method for detecting defects of switch cabinet based on transient ground voltage and ultrahigh frequency | |
| CN113848435B (en) | Direct-current partial discharge signal classification and identification method based on frequency spectrum energy distribution probability | |
| He et al. | Characteristics of alternating current corona discharge pulses and its radio interference level in a coaxial wire-cylinder gap | |
| Ren et al. | Partial discharges in SF 6 gas filled void under standard oscillating lightning and switching impulses in uniform and non-uniform background fields | |
| CN211293129U (en) | Partial discharge detection device with combined action of alternating current and impulse voltage | |
| CN114355270B (en) | Ultrahigh frequency electromagnetic pulse sensor characteristic calibration method and system | |
| Petit | Comparison of PD amplitudes of stator bars taken with different instruments | |
| CN110596545A (en) | Partial discharge detection device and detection method based on combined action of alternating current and impulse voltage | |
| Hui et al. | Cluster analysis on signals from XLPE cable partial discharge detection | |
| Li et al. | Influence of wave propagation process on measurement of corona current | |
| CN212723049U (en) | Ionic current and discharge current acquisition and detection device in aging and testing processes and aging equipment | |
| CN110632396B (en) | Cable dielectric loss measuring method | |
| Waldi et al. | Development of rogowski coil sensor for partial discharge detection | |
| Dobrzycki et al. | Analysis of acoustic emission signals accompanying the process of electrical treeing of epoxy resins | |
| Wu et al. | Analysis of current characteristics of corona discharge in high voltage transmission | |
| Ma et al. | Analysis of Ultraviolet Characteristics of Partial Discharge for Insulators | |
| Liu et al. | Study on the relationship between photon number and discharge magnitude of corona discharge based on ultraviolet imaging method | |
| CN203595767U (en) | Internal fault test system of high voltage capacitor | |
| He et al. | Characteristics of AC corona discharge pulses and RI levels in a coaxial wire-cylinder gap | |
| Dezenzo et al. | A new method for the evaluation and recognition of corona under direct voltage stress |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |