CN112684311A - Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer - Google Patents
Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer Download PDFInfo
- Publication number
- CN112684311A CN112684311A CN202110131797.9A CN202110131797A CN112684311A CN 112684311 A CN112684311 A CN 112684311A CN 202110131797 A CN202110131797 A CN 202110131797A CN 112684311 A CN112684311 A CN 112684311A
- Authority
- CN
- China
- Prior art keywords
- discharge
- phase
- partial discharge
- cycle
- maximum
- 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
Images
Landscapes
- Testing Relating To Insulation (AREA)
Abstract
The invention relates to a characteristic quantity extraction method for identifying the oil paper insulation partial discharge type of a transformer, which comprises the following steps: step 1, building a simulation loop of a transformer oil paper insulation partial discharge laboratory, and setting a typical transformer oil paper insulation defect model; step 2, carrying out experiments aiming at different defect models, collecting partial discharge PRPD spectrogram information, and dividing the collected data into a training set and a testing set; step 3, extracting the relationship between the discharge repetition rate and the discharge phase of the positive and negative half-cycle reaction from the PRPD spectrogram obtained in the step 2 respectively
A spectrogram; step 4, obtaining from step 3
And extracting phases corresponding to 50%, 30% and 10% of maximum discharge times of the positive and negative half cycles from the spectrogram to serve as new characteristic quantities. Compared with the prior art, the method has the advantages of effectively avoiding the influence of accidental partial discharge information on the characteristic quantity value, being less in time consumption and the like.
Description
Technical Field
The invention relates to the technical field of online monitoring of insulation states of oil paper insulation electrical equipment, in particular to a characteristic quantity extraction method for identifying partial discharge types of oil paper insulation of a transformer.
Background
With the development of electric power systems in China, the total mileage of power transmission lines is continuously increased, the power transformer is used as key equipment of the electric power systems, safe and reliable operation of the power transformer has very important significance on the stability of the electric power systems, and according to statistical results, more than 50% of electric power system faults are caused by the faults of the power transformer every year in China. The failure modes of the transformer are various, and the insulation failure is the main failure mode of the transformer. Partial discharge is an important cause of insulation failure of the transformer, and the occurrence of the insulation failure can aggravate the generation of the partial discharge phenomenon. Therefore, the local discharge signal is monitored, the characteristic quantity is extracted so as to further carry out local discharge type identification, the type and the position of the local discharge of the voltage transformer are judged, the insulation fault level of the transformer can be reflected in time, and the occurrence of power accidents is avoided. The main insulating media in the power transformer are transformer oil and insulating paper boards, so that the partial discharge type identification of the transformer oil paper insulation can effectively reflect the insulation state of the transformer.
The existing method for identifying the type of the transformer oiled paper insulation is mainly based on a partial discharge phase distribution pattern (PRPD), a pulse sequence phase distribution pattern (PRPSA) and a partial discharge time distribution pattern(PRPS) based on a partial discharge phase distribution pattern (PRPD), i.e.
The type identification of the pattern mainly aims at extracting relevant characteristic quantity of a PRPD spectrogram. Such as for its sub-spectrogram
Kurtosis K extracted from n-quInclination of SkAnd degree of asymmetry Asy. Wherein kurtosis KuThe method is characterized in that the concentrated distribution degree of the spectrum ordinate scalar quantity is embodied, the larger the kurtosis value is, the more concentrated the spectrum ordinate quantity distribution is represented, and the smaller the value is, the more dispersed the spectrum ordinate quantity distribution is represented.
To is directed at
All ordinate scalars are needed in the process of extracting the kurtosis value from the n-q spectrogram. However, in the measurement of partial discharge, due to the fact that partial discharge development has a certain degree of contingency, some partial discharges with small amplitude or low occurrence frequency cannot reflect differences of different partial discharge types, but the partial discharges have large influences on kurtosis values and skewness values, and when the partial discharges are used for transformer oil-paper insulation type identification, certain interference can be generated on identification results.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a characteristic quantity extraction method for identifying the transformer oil paper insulation partial discharge type, wherein the characteristic quantity extracted by the method is the characteristic quantity which can effectively avoid the influence on the characteristic quantity value to a greater extent due to a very small number of accidental partial discharges, so that the accuracy of identifying the transformer oil paper insulation partial discharge type is not influenced.
The purpose of the invention can be realized by the following technical scheme:
a characteristic quantity extraction method for transformer oilpaper insulation partial discharge type identification comprises the following steps:
step 1, building a simulation loop of a transformer oil paper insulation partial discharge laboratory, and setting a typical transformer oil paper insulation defect model;
step 2, carrying out experiments aiming at different defect models, collecting partial discharge PRPD spectrogram information, and dividing the collected data into a training set and a testing set;
Preferably, the step 1 specifically comprises:
and (3) setting up test loops under four insulation defects of the oil paper insulation needle plate, the ball plate, the plate and the edge surface of the transformer, and respectively simulating local discharge and edge discharge under an extremely non-uniform electric field, a slightly non-uniform electric field and a uniform electric field in the oil paper insulation of the transformer.
Preferably, the step 2 specifically comprises:
the partial discharge is measured by adopting a pulse current method, the detection impedance is connected in series with the coupling loop, and the partial discharge instrument is connected with the detection impedance to carry out data acquisition of the partial discharge, so that a PRPD spectrogram of the partial discharge is obtained.
Preferably, the PRPD spectrogram contains information of phase, discharge amount and discharge times of partial discharge.
Preferably, the step 3 specifically comprises:
according to the respective discharge characteristics of the transformer oil paper insulation partial discharge in the AC positive half cycle and the AC negative half cycle, the relation between the discharge times and the discharge phase of the positive half cycle and the negative half cycle is respectively extracted from the PRPD spectrogram
Spectra.
Preferably, the positive half cycle is 0-180 DEG phase,
preferably, the negative half cycle is 181-360 degrees in phase.
Preferably, the step 4 specifically includes:
step 4-1, extracting positive half-cycle characteristic parameters;
and 4-2, extracting the negative half-cycle characteristic parameters.
Preferably, the step 4-1, the positive half-cycle characteristic parameter extraction specifically comprises:
will be half-cycle positive
Finding and recording the phase corresponding to the maximum discharge times, namely the maximum discharge repetition rate in the spectrogram
The number of discharges was recorded as nmax+；
Take n50％+＝0.5nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n50％+All phases of
B is more than or equal to a, and the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of 50 percent of the maximum discharge times of the positive half cycle
Namely, it is
Take n30％+＝0.3nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n30％+All phases of
Wherein d ≧ c, the difference between the maximum phase and the minimum phase satisfying the condition is takenPhase width of 30% of maximum discharge frequency in positive half cycle
Namely, it is
Take n10％+＝0.1nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n10％+All phases of
Wherein f is more than or equal to e, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the maximum discharge frequency of 10 percent of the positive half cycle
Namely, it is
Preferably, the step 4-2 of extracting the negative half-cycle characteristic parameters specifically comprises:
will be in a negative half cycle
Finding and recording the phase corresponding to the maximum discharge times, namely the maximum discharge repetition rate in the spectrogram
The number of discharges was recorded as nmax-；
Take n50％-＝0.5nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n50％-All phases of
H is more than or equal to g, and the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of 50 percent of the maximum discharge times of the negative half cycle
Namely, it is
Take n30％-＝0.3nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n30％-All phases of
Wherein j is more than or equal to i, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the 30 percent maximum discharge times of the negative half cycle
Namely, it is
Take n10％-＝0.1nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n10％-All phases of
Wherein l is more than or equal to k, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the 10% maximum discharge times of the negative half cycle
Namely, it is
Compared with the prior art, the invention has the following advantages:
1. the method has the advantages of simple extraction process of the characteristic quantity, and less use time and skewness under the condition of the same data quantity.
2. The invention discloses the electric field distribution condition under different discharge models when partial discharge occurs in the oil paper insulation of the transformer through a simulation method, explains the theoretical basis of the characteristic quantity for identifying the partial discharge type of the oil paper insulation of the transformer by combining the theoretical analysis of the partial discharge, and proves the feasibility of the principle.
3. The new characteristic quantity extraction process needs to determine the partial discharge times of 50%, 30% and 10%, and the information of accidental partial discharge is weakened due to the fact that the partial discharge times are numerous, so that the characteristic quantity is low in sensitivity to accidental partial discharge with few occurrence times, and the characteristic quantity value cannot be influenced by accidental partial discharge of irrelevant type identification.
4. The method is used for identifying the type of the transformer paper oil insulation partial discharge, and can achieve a good partial discharge type identification effect no matter the method is used alone or used together with other kurtosis, asymmetry and skewness as input characteristic quantities.
Drawings
FIG. 1 is a model of partial discharge defects in four types of transformer oiled paper insulation according to an embodiment, wherein (a) is an oiled paper insulation needle plate, (b) is a ball plate, (c) is a plate, and (d) is a plane;
FIG. 2 is a simulation result of electric field distribution between the upper and lower electrodes of the transformer oiled paper insulation partial discharge pin plate and the ball plate defect model in the embodiment, wherein (a) is the pin plate and (b) is the ball plate;
FIG. 3 is a simulation result of electric field distribution between the transformer oiled paper insulation partial discharge plate and the upper and lower electrodes of the creeping discharge defect model in the embodiment, in which (a) is the plate and (b) is the creeping surface;
FIG. 4 is a diagram illustrating the condition of feature quantities extracted from different defect models for identifying the type of partial discharge in paper-oil insulation of a transformer according to the present invention in an exemplary embodiment;
FIG. 5 shows the partial discharge of the ball-panel defect model in the experiment of example 1
Spectra.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
The method comprises the steps of constructing a transformer oil paper insulation defect model in a laboratory to obtain PRPD spectrograms of different types of partial discharge according to the PRPD spectrogram
And extracting phase information corresponding to the maximum discharge times from the spectrogram, taking the phase widths corresponding to 50%, 30% and 10% of the maximum discharge times of the positive and negative half cycles as new characteristic quantities, and describing theoretical bases of the new characteristic quantities by means of simulation verification. And the effect of the characteristic quantity as the identification characteristic quantity of the transformer oil paper insulation partial discharge type is verified through an experimental laboratory system, so that the feasibility and the accuracy of the characteristic quantity used for the identification of the transformer oil paper insulation partial discharge type are proved.
First, the analysis method of the present invention is described.
The invention explains the theoretical basis of the partial discharge mechanism of the transformer oilpaper insulation partial discharge by analyzing the discharge mechanism and assisting simulation verification. Compared with spectrogram characteristics of partial discharge of different transformer oil paper insulation, the characteristic quantity for identifying the type of the partial discharge of the transformer oil paper insulation is provided. And the effect of the characteristic quantity as the identification characteristic quantity of the transformer oil paper insulation partial discharge type is verified through an experimental laboratory system, so that the feasibility and the accuracy of the characteristic quantity used for the identification of the transformer oil paper insulation partial discharge type are proved. The specific analysis and verification steps are as follows:
the method comprises the steps of firstly analyzing the mechanism of the oil paper insulation partial discharge of the transformer, carrying out simulation, and verifying the field intensity distribution conditions of a needle plate electrode, a ball plate electrode, a plate electrode and an edge electrode.
The condition that partial discharge occurs in the oil paper insulation of the transformer is that partial discharge voltage is reached, the electric field distribution between the upper electrode and the lower electrode is not even due to the irregular shape of the electrodes, and partial area between the upper electrode and the lower electrode meets the requirement of partial discharge under the condition that certain voltage is applied to the outside, so the partial discharge occurs; the voltage of partial area does not reach the requirement of partial discharge, so the partial discharge does not occur, which results in that partial discharge with different times occurs at different parts of the same electrode, so the total discharge times are different due to different defect models.
In order to verify the above analysis, electric field distribution simulation tests between pin plate electrodes, between ball plate electrodes, between plate electrodes and between along-plane defect model electrodes were performed, and the results shown in fig. 2 and 3 were obtained to verify the correctness of the above analysis.
Partial discharge is analyzed from the perspective of a power frequency period, and the instantaneous voltage peak value of the partial discharge also changes along with the change of the voltage phase. In the process of changing the phase from 0 to 90 degrees, the voltage amplitude is continuously increased along with the increase of the phase, and in the process of increasing the voltage amplitude, because the electric field intensity at the sharp corners of the electrodes is concentrated, the sharp corners between the electrodes firstly reach the partial discharge voltage, and partial discharge occurs; with the voltage amplitude rising again, partial discharge occurs at other positions successively.
In the process of changing the phase from 90 degrees to 180 degrees, the voltage amplitude is continuously reduced along with the increase of the phase, and in the process of reducing the voltage amplitude, firstly, the voltage at a non-sharp angle can not meet the requirement of self-sustaining discharge any more, so that the partial discharge disappears, and then the total discharge frequency under the model is changed. The process analysis from 180 ° to 360 ° is as above.
From the above analysis, it is known that the electric field intensity at different positions between the electrodes changes to different degrees with the change of the voltage phase, and the number of discharges changes accordingly.
The electric field distribution uniformity among the pin plate electrodes, among the ball plate electrodes, among the plate electrodes and among the along-surface defect model electrodes is different, so that in the process that the voltage amplitude changes along with the phase, the voltage change amplitude among the electrodes is different, the voltage change of different parts under the same electrode is also different, the obvious change of the discharge frequency along with the phase is reflected, and the response sensitivity of the change of the discharge frequency under different electrodes to the phase change is different. Therefore, the correspondence relationship between the discharge frequency and the phase, i.e., the phase widths corresponding to the 50%, 30%, and 10% maximum discharge frequency of the positive and negative half cycles can be used as a type of characteristic quantity for identifying the partial discharge type.
Secondly, comparing discharge spectrograms of different defect models under laboratory conditions, extracting phase widths corresponding to 50%, 30% and 10% of maximum discharge times of the positive half cycle and the negative half cycle, and verifying the conclusion.
A transformer oil paper insulation partial discharge laboratory loop is built under laboratory conditions, four defect models of a typical transformer oil paper insulation needle plate, a ball plate, a plate and an edge surface are arranged as shown in figure 1, and partial discharge and edge discharge which occur under an extremely non-uniform electric field, a slightly non-uniform electric field and a uniform electric field in transformer oil paper insulation are simulated respectively. Extracting the discharge times and discharge phase information of the obtained partial discharge information spectrogram to obtain positive and negative half cycles under various defect models
Spectra.
Extracting positive half-cycle characteristic parameters:
will be half-cycle positive
Finding and recording the phase corresponding to the maximum discharge times, namely the maximum discharge repetition rate in the spectrogram
The number of discharges was recorded as nmax+。
Take n50％+＝0.5nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n50％+All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of 50% of the maximum discharge times in the positive half cycle
Namely, it is
Take n30％+＝0.3nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n30％+All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the maximum discharge frequency of 30% in the positive half cycle
Namely, it is
Take n10％+＝0.1nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n10％+All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the maximum discharge frequency of 10% in the positive half cycle
Namely, it is
Extracting negative half-cycle characteristic parameters:
will be in a negative half cycle
Maximum number of discharges in spectrumThe phase position corresponding to the number, i.e. the maximum discharge repetition rate, is found and recorded
The number of discharges was recorded as nmax-。
Take n50％-＝0.5nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n50％-All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the 50% maximum discharge times of the negative half cycle
Namely, it is
Take n30％-＝0.3nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n30％-All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the maximum discharge frequency of 30% in the negative half cycle
Namely, it is
Take n10％-＝0.1nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n10％-All phases of
The difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the maximum discharge frequency of 10% in the negative half cycle
Namely, it is
Each defect model was characterized by 15 sets of spectral information collected, which was used for feature extraction
The correspondence between the feature quantity and the defect type is shown in FIG. 4, which shows that
The six characteristic quantities have good discrimination among defect models and can be used as a class of characteristic quantities for identifying the type of the partial discharge.
Therefore, can be used
The characteristic quantity is used for identifying the oil paper insulation type of the transformer.
In order to verify the conclusion, experiments are carried out and verified in a transformer oil paper insulation partial discharge laboratory experiment system, and the feasibility and the accuracy of the characteristic quantity used for identifying the transformer oil paper insulation partial discharge type are proved.
DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION
A loop of a transformer oil paper insulation partial discharge laboratory is built, four defect models of a typical transformer oil paper insulation needle plate, a ball plate, a plate and an edge surface are arranged as shown in figure 1, and partial discharge and surface discharge which occur under an extremely non-uniform electric field, a slightly non-uniform electric field and a uniform electric field in the transformer oil paper insulation are simulated respectively.
The transformer oil paper insulation voltage withstand test is respectively carried out on different defect models, PRPD spectrograms of partial discharge of various defect models are obtained based on pulse current method measurement, partial discharge data of each defect model are collected, the collection time of each group of data is 5 minutes, and 40 groups of PRPD spectrogram information of partial discharge are collected under each defect model. Of these, 35 sets of data were used as training sets and 5 sets of data were used to test the final classification results. The following description of specific steps for extracting characteristic quantity and identifying the transformer oil-paper insulation partial discharge type is carried out by taking a first group of data of a ball plate defect model as an example:
and performing a loop experiment of an oiled paper insulation partial discharge laboratory to obtain a PRPD spectrogram under a needle plate defect model, wherein the partial discharge data of the group totally comprises 146294 partial discharges, and spectrogram information also comprises discharge phase, discharge magnitude and discharge time information. Extracting positive and negative half cycles to show discharge repetition rate and discharge phase relation
The spectrogram, namely the spectrogram of the corresponding relationship between the discharge times and the phase is shown in FIG. 5.
Extracting positive half-cycle characteristic parameters:
positive half cycle
The phase corresponding to the maximum discharge frequency in the spectrogram, namely the maximum discharge repetition rate is 114 degrees, and is recorded as
The number of discharges is 5693, which is denoted as nmax+＝5693。
Take n50％+＝0.5nmax+2846.5, all phases 102 DEG, 104 DEG, … … and 125 DEG of which the positive half-cycle discharge frequency is equal to or more than 2846.5 are found, and the difference between the maximum phase and the minimum phase which satisfy the condition is taken as the phase width of the positive half-cycle 50% of the maximum discharge frequency
I.e. the process is repeated.
Take n30％+＝0.3nmax+When the phase width of the maximum discharge frequency of 30% in the positive half cycle is determined as 1707.9, all phases 95 °, 97 °, … … and 130 ° in which the number of positive half cycle discharges is 1707.9 or more are found, and the difference between the maximum phase and the minimum phase satisfying the condition is regarded as the phase width of the maximum discharge frequency of 30% in the positive half cycle
I.e. the process is repeated.
Take n10％+＝0.1nmax+569.3, all phases 90 °, 92 °, … …, 135 ° in which the number of positive half-cycle discharges is 569.3 or more are found, and the difference between the maximum phase and the minimum phase satisfying the condition is defined as the positive half-cycle 10% maximum discharge number phase width
Namely, it is
Extracting negative half-cycle characteristic parameters:
negative half cycle
The phase corresponding to the maximum discharge frequency, i.e. the maximum discharge repetition rate, in the spectrogram is 294 DEG and is found out and recorded
The number of discharges was 2990, which is recorded as nmax-＝2990。
Take n50％-＝0.5nmax-All phases 292 °, 294 °, … …, and 306 ° in which the number of times of discharge in the negative half cycle is equal to or greater than 1495 are found, and the difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the maximum number of times of discharge in the negative half cycle of 50%
Namely, it is
Take n30％-＝0.3nmax-When the phase widths of the maximum discharge times of 30% in the negative half cycle are determined as 897, all phases 292 °, 293 °, … …, and 308 ° in which the negative half cycle discharge times are equal to or more than 897 are found out
Namely, it is
Take n10％-＝0.1nmax-In 299, all phases 256 DEG 259 DEG … … DEG 313 DEG of negative half-cycle discharge frequency equal to or more than 299 are found, and the difference between the maximum phase and the minimum phase satisfying the condition is taken as the phase width of the negative half-cycle 10% maximum discharge frequency
Namely, it is
The characteristic parameter extraction is carried out on 146294 partial discharges of the group by using an algorithm, the kurtosis extraction time is 0.251677 seconds, the kurtosis extraction time is 0.274802 seconds, and the skewness extraction time is 0.279103 seconds. The extraction of characteristic quantities proposed by the invention is significantly less time-consuming.
And sequentially carrying out the characteristic quantity extraction for identifying the transformer oiled paper insulation partial discharge type by using 160 groups of data which are counted by a training set and a test set under the ball plate defect, the needle plate defect and the plate defect model. The characteristic quantities of the 20 test sets and the corresponding defect types are shown in table 1 below, and the abscissa 1 to 6 represents the positive half cycle 50% maximum discharge time phase width, the positive half cycle 30% maximum discharge time phase width, the positive half cycle 10% maximum discharge time phase width, the negative half cycle 50% maximum discharge time phase width, the negative half cycle 30% maximum discharge time phase width, and the negative half cycle 10% maximum discharge time phase width, respectively.
TABLE 1
And then inputting the extracted characteristic quantity for identifying the transformer oil paper insulation partial discharge type into a BP neural network to identify the transformer oil paper insulation partial discharge type. Setting the number of input layer nodes of the BP neural network as 6, the number of output layer nodes as 4, the number of hidden layer nodes as 10, and outputting 20 groups of test data as the following results: correct 19 groups were identified, incorrect 1 group was identified, and the correct identification rate was 95%.
The above 160 groups of data are processed
Extracting the kurtosis of the spectrogram, inputting the kurtosis as a characteristic quantity into a BP neural network for pattern recognition, wherein the correct recognition rate is 75%; the above 160 groups of data are processed
And extracting the skewness of the spectrogram, inputting the skewness as a characteristic quantity into a BP neural network for pattern recognition, wherein the correct recognition rate is 65%.
The experimental result in the embodiment 1 proves that the new characteristic quantity for identifying the partial discharge type of the oil paper insulation of the transformer can effectively reflect the difference of different discharge types of the oil paper insulation of the transformer; the new characteristic quantity for identifying the transformer oil paper insulation partial discharge type provided by the invention has good effect when being used for identifying the transformer oil paper insulation partial discharge type.
The invention is based on the partial discharge experiment under the transformer oil paper insulation laboratory condition, measures the partial discharge
In the spectrogram, the intensity of the partial discharge times changing along with the change of the voltage phase can reflect different partial discharge types, so that the partial discharge can be utilizedThe intensity of the change of the number of times with the change of the voltage phase serves as a characteristic quantity indirectly characterizing the partial discharge type.
Quantifying the intensity of the partial discharge frequency changing with the voltage phase, using positive and negative half cycles
The phase width corresponding to 50% of maximum discharge times, the phase width corresponding to 30% of maximum discharge times and the phase width corresponding to 10% of maximum discharge times in the spectrogram are used as characteristic quantities for identifying the oil paper insulation partial discharge type of the transformer, and when the partial discharge types are different, the positive half cycle and the negative half cycle are carried out
The phase width corresponding to 50% of the maximum discharge times, the phase width corresponding to 30% of the maximum discharge times and the phase width corresponding to 10% of the maximum discharge times in the spectrogram are greatly different.
The extracted characteristic quantity can effectively avoid the influence of accidental partial discharge information on the characteristic quantity value, can be effectively used for identifying the type of the transformer paper oil insulation partial discharge, can be used as the characteristic quantity of the type identification of the transformer paper oil insulation partial discharge alone or can be used for identifying the type of the transformer paper oil insulation with the characteristic quantity which is widely applied such as kurtosis, skewness and the like to obtain a better identification result, and meanwhile, the extraction time consumption of the characteristic quantity is obviously lower than that of the kurtosis, skewness and the like.
While the invention has been described with reference to specific embodiments, the invention is not limited thereto, and various equivalent modifications and substitutions can be easily made by those skilled in the art within the technical scope of the invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A characteristic quantity extraction method for transformer oilpaper insulation partial discharge type identification is characterized by comprising the following steps:
step 1, building a simulation loop of a transformer oil paper insulation partial discharge laboratory, and setting a typical transformer oil paper insulation defect model;
step 2, carrying out experiments aiming at different defect models, collecting partial discharge PRPD spectrogram information, and dividing the collected data into a training set and a testing set;
step 3, extracting the relationship between the discharge repetition rate and the discharge phase of the positive and negative half-cycle reaction from the PRPD spectrogram obtained in the step 2 respectively
A spectrogram;
step 4, obtaining from step 3
And extracting phases corresponding to 50%, 30% and 10% of maximum discharge times of the positive and negative half cycles from the spectrogram to serve as new characteristic quantities.
2. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 1, wherein the step 1 specifically comprises the following steps:
and (3) setting up test loops under four insulation defects of the oil paper insulation needle plate, the ball plate, the plate and the edge surface of the transformer, and respectively simulating local discharge and edge discharge under an extremely non-uniform electric field, a slightly non-uniform electric field and a uniform electric field in the oil paper insulation of the transformer.
3. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 1, wherein the step 2 specifically comprises the following steps:
the partial discharge is measured by adopting a pulse current method, the detection impedance is connected in series with the coupling loop, and the partial discharge instrument is connected with the detection impedance to carry out data acquisition of the partial discharge, so that a PRPD spectrogram of the partial discharge is obtained.
4. The method as claimed in claim 3, wherein the PRPD spectrogram contains information of phase, discharge amount and discharge times of partial discharge.
5. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 1, wherein the step 3 specifically comprises:
according to the respective discharge characteristics of the transformer oil paper insulation partial discharge in the AC positive half cycle and the AC negative half cycle, the relation between the discharge times and the discharge phase of the positive half cycle and the negative half cycle is respectively extracted from the PRPD spectrogram
Spectra.
6. The method for extracting the characteristic quantity for the transformer oil paper insulation partial discharge type identification according to claim 5, wherein the positive half cycle is 0-180 ° in phase.
7. The method for extracting the characteristic quantity for the transformer oil paper insulation partial discharge type identification as claimed in claim 5, wherein the negative half cycle is 181-360 ° in phase.
8. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 1, wherein the step 4 specifically comprises:
step 4-1, extracting positive half-cycle characteristic parameters;
and 4-2, extracting the negative half-cycle characteristic parameters.
9. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 8, wherein the step 4-1, the positive half cycle characteristic parameter extraction specifically comprises:
will be rightHalf cycle
Finding and recording the phase corresponding to the maximum discharge times, namely the maximum discharge repetition rate in the spectrogram
The number of discharges was recorded as nmax+；
Take n50％+＝0.5nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n50％+All phases of
B is more than or equal to a, and the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of 50 percent of the maximum discharge times of the positive half cycle
Namely, it is
Take n30％+＝0.3nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n30％+All phases of
D is more than or equal to c, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of 30 percent of the maximum discharge times of the positive half cycle
Namely, it is
Take n10％+＝0.1nmax+Finding out the number of positive half-cycle discharge times greater than or equal to n10％+All phases of
Wherein f is more than or equal to e, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the maximum discharge frequency of 10 percent of the positive half cycle
Namely, it is
10. The method for extracting the characteristic quantity for identifying the transformer oil paper insulation partial discharge type according to claim 9, wherein the step 4-2, negative half cycle characteristic parameter extraction specifically comprises:
will be in a negative half cycle
Finding and recording the phase corresponding to the maximum discharge times, namely the maximum discharge repetition rate in the spectrogram
The number of discharges was recorded as nmax-；
Take n50％-＝0.5nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n50％-All phases of
H is more than or equal to g, and the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of 50 percent of the maximum discharge times of the negative half cycle
Namely, it is
Take n30％-＝0.3nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal ton30％-All phases of
Wherein j is more than or equal to i, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the 30 percent maximum discharge times of the negative half cycle
Namely, it is
Take n10％-＝0.1nmax-Finding out the number of times of negative half-cycle discharge is greater than or equal to n10％-All phases of
Wherein l is more than or equal to k, the difference between the maximum phase and the minimum phase meeting the condition is taken as the phase width of the 10% maximum discharge times of the negative half cycle
Namely, it is
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110131797.9A CN112684311B (en) | 2021-01-30 | 2021-01-30 | Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110131797.9A CN112684311B (en) | 2021-01-30 | 2021-01-30 | Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112684311A true CN112684311A (en) | 2021-04-20 |
| CN112684311B CN112684311B (en) | 2023-04-07 |
Family
ID=75459564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110131797.9A Active CN112684311B (en) | 2021-01-30 | 2021-01-30 | Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112684311B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114113942A (en) * | 2021-11-25 | 2022-03-01 | 国网四川省电力公司成都供电公司 | Oil paper insulation partial discharge identification and characteristic correction method and experimental device |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1580797A (en) * | 2004-05-17 | 2005-02-16 | 西安交通大学 | Power cable local discharge on-line monitoring method and device |
| CN2718593Y (en) * | 2004-05-17 | 2005-08-17 | 西安交通大学 | Electric cable local discharge on-line monitoring apparatus |
| CN101408580A (en) * | 2008-11-21 | 2009-04-15 | 重庆大学 | Method for evaluating oil paper insulation ageing state based on local discharge characteristic parameter |
| CN102147440A (en) * | 2011-03-10 | 2011-08-10 | 华北电力大学 | Method for diagnosing partial discharge defect severity of paper oil insulating edge surface of transformer |
| CN104198899A (en) * | 2014-08-04 | 2014-12-10 | 西安交通大学 | Partial discharge type identifying method for transformer under multi-discharge source situation |
| US20140372052A1 (en) * | 2011-09-19 | 2014-12-18 | Korea Electric Power Corporation | Method and device for determining the defect type of a partial discharge |
| US20150185270A1 (en) * | 2012-12-28 | 2015-07-02 | State Grid Corporation Of China | Method for recognizing transformer partial discharge pattern based on singular value decomposition algorithm |
| CN104849633A (en) * | 2015-05-15 | 2015-08-19 | 国家电网公司 | Switchgear partial discharge mode recognition method |
| CN104991171A (en) * | 2015-06-25 | 2015-10-21 | 国家电网公司 | Method for drawing GIS partial discharge frequency division fault spectrogram based on ultrahigh frequency signal |
| CN105044567A (en) * | 2015-06-29 | 2015-11-11 | 许继集团有限公司 | GIS partial discharge on-line monitoring mode identification method and GIS partial discharge on-line monitoring mode identification system |
| CN108805107A (en) * | 2018-07-12 | 2018-11-13 | 华南理工大学 | A kind of inside GIS shelf depreciation defect identification method based on PRPS signals |
| US20190041450A1 (en) * | 2015-12-01 | 2019-02-07 | Yandong LV | An intelligent assessment method of main insulation condition of transformer oil paper insulation |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
| CN109709459A (en) * | 2019-01-28 | 2019-05-03 | 国网安徽省电力有限公司电力科学研究院 | A kind of atlas analysis method for partial discharge monitoring data |
| CN111999382A (en) * | 2020-09-17 | 2020-11-27 | 海南电网有限责任公司电力科学研究院 | Cable partial discharge characteristic parameter extraction method considering insulation aging |
-
2021
- 2021-01-30 CN CN202110131797.9A patent/CN112684311B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN2718593Y (en) * | 2004-05-17 | 2005-08-17 | 西安交通大学 | Electric cable local discharge on-line monitoring apparatus |
| CN1580797A (en) * | 2004-05-17 | 2005-02-16 | 西安交通大学 | Power cable local discharge on-line monitoring method and device |
| CN101408580A (en) * | 2008-11-21 | 2009-04-15 | 重庆大学 | Method for evaluating oil paper insulation ageing state based on local discharge characteristic parameter |
| CN102147440A (en) * | 2011-03-10 | 2011-08-10 | 华北电力大学 | Method for diagnosing partial discharge defect severity of paper oil insulating edge surface of transformer |
| US20140372052A1 (en) * | 2011-09-19 | 2014-12-18 | Korea Electric Power Corporation | Method and device for determining the defect type of a partial discharge |
| US20150185270A1 (en) * | 2012-12-28 | 2015-07-02 | State Grid Corporation Of China | Method for recognizing transformer partial discharge pattern based on singular value decomposition algorithm |
| CN104198899A (en) * | 2014-08-04 | 2014-12-10 | 西安交通大学 | Partial discharge type identifying method for transformer under multi-discharge source situation |
| CN104849633A (en) * | 2015-05-15 | 2015-08-19 | 国家电网公司 | Switchgear partial discharge mode recognition method |
| CN104991171A (en) * | 2015-06-25 | 2015-10-21 | 国家电网公司 | Method for drawing GIS partial discharge frequency division fault spectrogram based on ultrahigh frequency signal |
| CN105044567A (en) * | 2015-06-29 | 2015-11-11 | 许继集团有限公司 | GIS partial discharge on-line monitoring mode identification method and GIS partial discharge on-line monitoring mode identification system |
| US20190041450A1 (en) * | 2015-12-01 | 2019-02-07 | Yandong LV | An intelligent assessment method of main insulation condition of transformer oil paper insulation |
| CN108805107A (en) * | 2018-07-12 | 2018-11-13 | 华南理工大学 | A kind of inside GIS shelf depreciation defect identification method based on PRPS signals |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
| CN109709459A (en) * | 2019-01-28 | 2019-05-03 | 国网安徽省电力有限公司电力科学研究院 | A kind of atlas analysis method for partial discharge monitoring data |
| CN111999382A (en) * | 2020-09-17 | 2020-11-27 | 海南电网有限责任公司电力科学研究院 | Cable partial discharge characteristic parameter extraction method considering insulation aging |
Non-Patent Citations (4)
| Title |
|---|
| RUI GUO等: "Partial discharge characteristics and classification in oil-paper insulation under DC voltage", 《2016 IEEE INTERNATIONAL CONFERENCE ON HIGH VOLTAGE ENGINEERING AND APPLICATION (ICHVE)》 * |
| XUTAO HAN等: "Partial Discharge Characteristics of Surface Defect in SF6 Under Oscillating Impulse Voltage", 《2018 IEEE INTERNATIONAL CONFERENCE ON HIGH VOLTAGE ENGINEERING AND APPLICATION (ICHVE)》 * |
| 唐炬等: "高压电缆中间接头局部放电统计特性分析", 《重庆大学学报》 * |
| 廖瑞金等: "交流电压下油纸绝缘针板放电特性", 《高电压技术》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114113942A (en) * | 2021-11-25 | 2022-03-01 | 国网四川省电力公司成都供电公司 | Oil paper insulation partial discharge identification and characteristic correction method and experimental device |
| CN114113942B (en) * | 2021-11-25 | 2023-11-10 | 国网四川省电力公司成都供电公司 | Recognition and characteristic correction method and experimental device for oil paper insulation partial discharge |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112684311B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108008244B (en) | A kind of small current grounding fault progressive classifying identification method at many levels | |
| Gulski et al. | Computer-aided recognition of discharge sources | |
| CN110376497B (en) | Low-voltage distribution system series fault electric arc identification method based on full-phase deep learning | |
| CN105425076A (en) | Method of carrying out transformer fault identification based on BP neural network algorithm | |
| WO2014101579A1 (en) | Method for identifying a transformer local discharge mode based on a singular value decomposition algorithm | |
| CN104198898A (en) | Local discharge development process diagnosis method based on pulse-train analysis | |
| CN108120907B (en) | Partial discharge diagnosis method based on feature extraction under power frequency-low frequency voltage | |
| CN109284933B (en) | Electronic transformer state evaluation system and method based on mathematical statistics | |
| CN112149549A (en) | GIS partial discharge type identification method based on depth residual error network | |
| CN109100627A (en) | A kind of power equipment partial discharges fault diagnostic method based on end-to-end mode | |
| CN104849686B (en) | Partial discharge detector's performance evaluation system | |
| CN110261746B (en) | Cable defect detection method based on periodic attenuation characteristics of oscillating wave voltage | |
| CN112684311B (en) | Characteristic quantity extraction method for identifying oil paper insulation partial discharge type of transformer | |
| CN113533910A (en) | Method and system suitable for converter transformer partial discharge early warning | |
| JP2019191142A (en) | Winding insulation deterioration diagnostic device of rotary machine | |
| US11693041B2 (en) | Method for monitoring the electric insulation status of a piece of equipment for MV or HV electric systems | |
| CN117630800A (en) | Fault diagnosis method and system for automatic calibrating device of electric energy meter | |
| CN112881879A (en) | High-voltage cable terminal partial discharge mode identification method, device and equipment | |
| CN113052249B (en) | Transformer winding fault type identification method based on support vector machine and current deviation coefficient | |
| CN105182116B (en) | A kind of transformer winding working state detecting method based on weighted gradient structural similarity | |
| CN109917245B (en) | Ultrasonic detection partial discharge signal mode identification method considering phase difference | |
| Miao et al. | Study on PD Pattern Recognition of Power Transformer Considering External Corona Interference Signal | |
| JPH11258296A (en) | Automatic discrimination method for partial discharge | |
| Saucedo-Dorantes et al. | Novelty Detection Methodology Based on Self-Organizing Maps for Power Quality Monitoring | |
| CN117748507B (en) | Distribution network harmonic access uncertainty assessment method based on Gaussian regression model |
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 |