CN114548012B - Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis - Google Patents
Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis Download PDFInfo
- Publication number
- CN114548012B CN114548012B CN202210121763.6A CN202210121763A CN114548012B CN 114548012 B CN114548012 B CN 114548012B CN 202210121763 A CN202210121763 A CN 202210121763A CN 114548012 B CN114548012 B CN 114548012B
- Authority
- CN
- China
- Prior art keywords
- dimensional
- frequency response
- centroid
- response curve
- winding
- 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.)
- Active
Links
- 238000004804 winding Methods 0.000 title claims abstract description 125
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000003745 diagnosis Methods 0.000 title claims abstract description 21
- 238000004458 analytical method Methods 0.000 title claims abstract description 18
- 238000009826 distribution Methods 0.000 claims abstract description 32
- 238000010586 diagram Methods 0.000 claims description 13
- 238000004088 simulation Methods 0.000 claims description 13
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 238000001514 detection method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The transformer winding deformation fault diagnosis method based on the three-dimensional frequency response curve centroid analysis comprises the following steps of: establishing a transformer winding concentrated parameter model to obtain normal winding amplitude frequency data and phase frequency data; adjusting circuit parameters of a concentrated parameter model of a transformer winding to obtain amplitude and phase data of the winding in different fault states; the acquired frequency, amplitude and phase are presented in a three-dimensional Cartesian coordinate system as x, y and z values, and a three-dimensional frequency response curve is established; respectively calculating barycenter coordinates represented by the frequency, the amplitude and the phase, and visualizing the obtained coordinates to obtain barycenter distribution of each three-dimensional frequency response curve; and establishing a three-dimensional coordinate system, wherein the centroids of the three-dimensional frequency response curves of different faults are distributed around the three-dimensional coordinate system, and determining the fault types of the windings according to the differences of the centroid distribution intervals. The method can sensitively and effectively distinguish the types of faults of the windings through the difference of centroid distribution intervals of the three-dimensional frequency response curves.
Description
Technical Field
The invention relates to the field of transformer winding fault diagnosis, in particular to a transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis.
Background
Transformers are one of the most expensive assets in power systems. The cost of repairing the transformer after the transformer faults is extremely high, the transformer faults often cause interruption of a power supply, and once the faults occur, the normal operation of the transformer substation can be greatly influenced, so that larger property damage is caused. The working environment of the power transformer is generally bad, so that various faults are easy to generate, and winding deformation is one of the faults. When a short circuit accident occurs, the strong current can lead the transformer winding to generate permanent instability deformation such as distortion, bulge or displacement, and the transformer can not work normally when serious, thereby causing the accident of large-area power failure. To maintain normal and stable operation of the transformer, sensitive, reliable, cost-effective fault diagnosis techniques are required to detect any minor degree of winding deformation to avoid more serious damage to the transformer.
The frequency response method (Frequency Response Analysis, FRA) is widely used in winding deformation detection in the practical industry due to its simple and convenient instrument operation and high accuracy. To date, in order to solve the problem that the false judgment rate is high depending on manual experience when the winding deformation is diagnosed by using the frequency response method, to realize accurate diagnosis of the transformer winding deformation, researchers have made various attempts. Document [1]: zhao Zhongyong, tang Chao, li Chengxiang, etc. A transformer winding deformation fault diagnosis method based on frequency response binarized image [ J ]. High voltage technology, 2019,45 (5): 1526-1534. A transformer winding deformation classification method based on frequency response complex value and binarized processing is provided on the basis of frequency response method, and the sensitivity of winding diagnosis is improved. Document [2]: the method is characterized by comprising the following steps of (1) carrying out strong payment, (Ning Wenyao, liu Daifei) carrying out winding mixed fault analysis method [ J ]. High voltage technology based on multi-index fusion, (2021,47 (02): 537-545), and providing a method based on the ratio of the fusion Euclidean distance and the offset area ratio of a frequency response curve. At present, the detection effect on winding deformation based on the frequency response method is good and has the potential of on-line detection. However, because the frequency response detection result lacks related criterion support corresponding to the winding deformation state, the quantitative diagnosis of the type and degree of transformer winding deformation does not form a unified standard, and people with abundant professional experience can accurately judge the transformer winding deformation.
Disclosure of Invention
Aiming at the technical problems, the invention provides a transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis, which can sensitively and effectively distinguish the types of faults of windings through the difference of the centroid distribution intervals of the three-dimensional frequency response curve.
The technical scheme adopted by the invention is as follows:
the transformer winding deformation fault diagnosis method based on the three-dimensional frequency response curve centroid analysis comprises the following steps of:
step 1: establishing a transformer winding concentrated parameter model, injecting sinusoidal sweep frequency signals into the transformer winding concentrated parameter model, and obtaining winding input voltage And output voltageObtaining normal winding amplitude frequency data and phase frequency data as fingerprint data;
Step 2: adjusting circuit parameters of a concentrated parameter model of a transformer winding, realizing various fault deformation simulation of the winding, and acquiring amplitude and phase data of the winding in different fault states;
Step 3: presenting the frequency, amplitude and phase acquired in the step 1 and the step 2 as x, y and z values in a three-dimensional Cartesian coordinate system, and establishing a three-dimensional frequency response curve;
step 4: by the formula Wherein x, y and z respectively represent frequency, amplitude and phase, x o、yo、zo is the mass center, and n represents the number of points taken; respectively calculating the values of the centroid coordinates x, y and z represented by the frequency, the amplitude and the phase, and visualizing the obtained coordinates (x, y and z) to obtain centroid distribution of each three-dimensional frequency response curve;
step 5: and (3) according to the result of the centroid of the three-dimensional curve obtained in the step (4), establishing a three-dimensional coordinate system by taking the centroid of the three-dimensional frequency response curve of the normal winding as an origin, distributing the centroids of the three-dimensional frequency response curves of different faults around the centroid, and determining the fault type of the winding according to the difference of the centroid distribution intervals.
In the step 1, a transformer winding centralized parameter model is established by using circuit simulation software PSpice, and a sine sweep frequency signal with the frequency ranging from 1kHZ to 1MHZ is injected into a head winding of the transformer winding centralized parameter model.
PSpice establishes a transformer winding concentration parameter model as shown in fig. 6, wherein the transformer winding concentration parameter model is symmetrical, and windings at the first end and the second end are identical, and can be used for injecting signals from the first-stage winding so as to acquire required frequency response data at the tail end winding, and can be used for injecting signals from the tail end so as to acquire data at the first end.
In the step 2, different percent circuit parameters are used for carrying out deformation simulation of various fault degrees of the winding, and 3 types of fault types are simulated by taking 5% as a gradient within the range of 5% -95% parameter variation: axial deformation, radial deformation and axial displacement, 19 failure degrees per class.
Document [3]: understanding power transformer frequency response analysis signatures discloses that when different fault types occur, the relevant variables in the transformer concentration parameter model change as shown in table 1.
The 5% -95% parameter change specifically refers to a change range of the relevant parameter corresponding to the fault type, as shown in fig. 7, and the capacitance to ground C in the transformer concentration parameter model simulated by the invention is 121.286pF. If the axial deviation fault is simulated, only the value of the capacitance C to the ground is required to be adjusted, namely the value of the capacitance C is increased, and 19 axial deviations with different fault degrees are simulated by changing the capacitance C to the ground by 1.05 times of C, 1.1 times of C … 1.95.95 times of C and other 19 values within the range of 5-95% parameter change of the value of the capacitance C to the ground.
Therefore, the invention uses equivalent parameters with different percentages to simulate the deformation of the winding with various degrees, adjusts the state quantity of the simulation circuit of the concentrated parameter model of the transformer, and can realize the simulation of different fault states.
In the step 3, the frequency of the sweep frequency signal is used as an x-axis coordinate value, the frequency response amplitude is used as a y-axis coordinate value, and the frequency response phase value is used as a z-axis coordinate value, so that a three-dimensional frequency response curve graph is established.
In the step 5, the x coordinates of the centroid obtained by the frequency are the same, and after a three-dimensional cartesian coordinate system is established by taking the centroid of the three-dimensional frequency response curve of the normal winding as the origin, the coordinate values of the frequency response axes are all 0, as shown in fig. 3 (a) and fig. 3 (b).
This is because the frequency points selected during simulation are all constant; therefore, the three-dimensional Cartesian coordinate system is projected, only the amplitude and the phase, namely the y-axis coordinate and the z-axis coordinate are reserved, the two-dimensional centroid distribution diagram is drawn by dimension reduction, the centroid deviation is shown in fig. 5, and the deviation of the three-dimensional frequency response curve under different faults is represented by the centroid deviation.
In the step 5, a cartesian two-dimensional coordinate system of centroid distribution is divided into four sections according to a two-dimensional centroid distribution diagram drawn by amplitude and phase coordinates, three faults of axial deformation, radial deformation and axis deviation are respectively distributed in different sections, and fault types are classified according to the difference of the obtained centroid distribution sections.
In the step 5, if the centroid distribution result has the following characteristics:
① . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 1, radial deformation of the transformer winding occurs;
② . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 3, the transformer winding is axially deformed;
③ . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 4, the transformer winding generates axis deviation.
The invention discloses a transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis, which has the following technical effects:
1) Based on the original frequency response curve, the method takes the phase frequency information of the frequency response curve into consideration, so that a three-dimensional frequency response curve is constructed. Compared with the traditional method that the winding state is analyzed only through the frequency response amplitude, the built three-dimensional frequency response curve has more characteristics, and the recognition accuracy of the frequency response method is improved.
2) Compared with the original two-dimensional frequency response curve, the method has the advantages that the characteristic of reflecting the winding faults is added, and the winding faults can be classified by shifting the centroid of the three-dimensional frequency response curve. The method has higher sensitivity in diagnosing the axial displacement faults of the transformer winding, and improves the accuracy of transformer winding deformation diagnosis.
3) The method calculates the mass center of each three-dimensional frequency response curve through an algorithm, and establishes a three-dimensional coordinate system by taking the mass center of the three-dimensional frequency response curve of the normal winding as an origin. And the fault classification is realized by analyzing the distribution interval of the centroid of the three-dimensional frequency response curves of the three simulated faults. The fault classification of the transformer winding can be effectively realized through the analysis of the centroid of the three-dimensional frequency response curve, and the thought is widened for the research field.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 (a) is a graph I of three-dimensional frequency response of different faults at different angles according to the present invention;
fig. 2 (b) is a graph two of three-dimensional frequency response of different faults at different angles according to the present invention.
FIG. 3 (a) is a graph showing a centroid distribution of a three-dimensional frequency response curve at different angles according to the present invention;
Fig. 3 (b) is a diagram showing a centroid distribution of a three-dimensional frequency response curve at different angles according to the present invention.
Fig. 4 is a schematic diagram of centroid offset according to the present invention.
Fig. 5 is a diagram showing a centroid offset of a three-dimensional frequency response curve according to the present invention.
Fig. 6 is a diagram of a transformer winding equivalent circuit used in the present invention.
Fig. 7 is a diagram of a transformer winding set-up parametric model built using PSpice in accordance with the present invention.
Detailed Description
A transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis establishes a three-dimensional frequency response curve graph with scanning frequency, amplitude and phase being x, y and z respectively. Compared with the traditional method that the winding state is analyzed only through the frequency response amplitude, the built three-dimensional frequency response curve has more characteristics, and the recognition precision of the frequency response method is improved; and calculating the mass center of each three-dimensional frequency response curve through an algorithm, establishing a three-dimensional coordinate system by taking the mass center of the three-dimensional frequency response curve of the normal winding as an origin, and analyzing different distributions of the mass centers of the three-dimensional frequency response curves of different fault types to perform fault diagnosis. The method comprises the following steps:
step 1: establishing a transformer winding concentrated parameter model, injecting sinusoidal sweep frequency signals into the transformer winding concentrated parameter model, and obtaining winding input voltage And output voltageAnd obtaining normal winding amplitude frequency data and phase frequency data as fingerprint data. The fingerprint data refers to the original basic data for comparison, namely the frequency response data of the healthy transformer winding.
Step 2: and adjusting circuit parameters of a concentrated parameter model of the transformer winding, realizing the simulation of various fault deformation of the winding, and acquiring amplitude and phase data of the winding in different fault states.
Step 3: and (3) presenting the frequency, amplitude and phase acquired in the step (1) and the step (2) as x, y and z values in a three-dimensional Cartesian coordinate system, and establishing a three-dimensional frequency response curve.
Step 4: by the formulaWherein x, y and z respectively represent frequency, amplitude and phase, x o、yo、zo is the mass center, and n represents the number of points taken; and respectively calculating the values of the centroid coordinates x, y and z represented by the frequency, the amplitude and the phase, and visualizing the obtained coordinates (x, y and z), wherein the visualization means that the data are displayed in a form of a graph, and particularly shown in fig. 3. And obtaining the centroid distribution of each three-dimensional frequency response curve.
Step 5: and (3) according to the result of the centroid of the three-dimensional curve obtained in the step (4), establishing a three-dimensional coordinate system by taking the centroid of the three-dimensional frequency response curve of the normal winding as an origin, distributing the centroids of the three-dimensional frequency response curves of different faults around the centroid, and determining the fault type of the winding according to the difference of the centroid distribution intervals.
The specific coordinates are the barycenter coordinates obtained by calculating the frequency response data obtained in the step 1 according to the formula in the step 4, and then a three-dimensional coordinate system is established by taking the barycenter coordinates of the normal windings as the origin. For example: the obtained centroid of the three-dimensional frequency response curve of the normal winding is (X, Y, Z) = (144857.4, -87.88097, -426.8259) corresponding to the frequency, the amplitude and the phase respectively, and the simulated centroid coordinates of the three-dimensional frequency response curve with faults are as follows: the centroid coordinates (X1, Y1, Z1) = (144857.4, -87.80565, -417.01324) when the radial deformation is 50% of the parameter changes, the coordinates are established by taking the centroid of the three-dimensional frequency response curve of the normal winding as the origin, namely, the (X, Y, Z) is subtracted by taking the centroid coordinates of the normal winding as the origin, and the three-dimensional coordinates established by taking the centroid coordinates of the normal winding as the origin are finally obtained as shown in fig. 3.
In the step 1, a transformer winding centralized parameter model is established by using circuit simulation software PSpice, and a sine sweep frequency signal with the frequency ranging from 1kHZ to 1MHZ is injected into a head winding of the transformer winding centralized parameter model.
In the step 2, different percent circuit parameters are used for carrying out deformation simulation of various fault degrees of the winding, and 3 types of fault types are simulated by taking 5% as a gradient within the range of 5% -95% parameter variation: axial deformation, radial deformation and axial displacement, 19 failure degrees per class.
In the step 3, the frequency of the sweep frequency signal is used as an x-axis coordinate value, the frequency response amplitude is used as a y-axis coordinate value, and the frequency response phase value is used as a z-axis coordinate value, so that a three-dimensional frequency response curve graph is established.
In the step 5, the x coordinates of the mass centers obtained by the frequencies are the same, and after a three-dimensional Cartesian coordinate system is established by taking the mass center of the three-dimensional frequency response curve of the normal winding as the origin, the coordinate values of the frequency response axes are all 0; this is because the frequency points selected during simulation are all constant; therefore, the three-dimensional Cartesian coordinate system is projected, only the amplitude and the phase, namely the y-axis coordinate and the z-axis coordinate are reserved, and the dimension reduction is carried out on the three-dimensional Cartesian coordinate system to draw a two-dimensional centroid distribution diagram; the shift of the three-dimensional frequency response curve at different faults is characterized by the centroid shift.
In the step 5, a cartesian two-dimensional coordinate system of centroid distribution is divided into four sections according to a two-dimensional centroid distribution diagram drawn by amplitude and phase coordinates, three faults of axial deformation, radial deformation and axis deviation are respectively distributed in different sections, and fault types are classified according to the difference of the obtained centroid distribution sections.
In the step 5, if the centroid distribution result has the following characteristics:
① . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 1, radial deformation of the transformer winding occurs;
② . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 3, the transformer winding is axially deformed;
③ . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 4, the transformer winding generates axis deviation.
The transformer windings were simulated using PSpice. At present, no study has explored the determination of the extent of winding failure in relation to its associated state quantity. Thus, the present invention uses different percent equivalent parameters for various degrees of deformation simulation of the winding. And (3) adjusting the state quantity of the simulation circuit to simulate different fault states.
Three faults were simulated. Including axial deformation, radial deformation, and axial misalignment. The present invention simulates 3 fault types with a gradient of 5% over a range of 5% -95% parameter variation based on the parameter variation indicated in table 1.
TABLE 1 model parameter value variation corresponding to fault types
And respectively establishing a three-dimensional Cartesian coordinate system by taking the frequency, amplitude and phase as x, y and z-axis coordinates. For example: the obtained centroid of the three-dimensional frequency response curve of the normal winding is (X, Y, Z) = (144857.4, -87.88097, -426.8259) corresponding to frequency, amplitude and phase respectively, and the coordinate system is established by taking the centroid of the three-dimensional frequency response curve of the normal winding as an origin, namely subtracting (X, Y, Z) from (X1, Y1, Z1), and finally the three-dimensional coordinate system established by taking the centroid coordinate of the normal winding as the origin is shown in figure 3 by taking the centroid coordinate of the simulated three-dimensional frequency response curve of the fault, such as the centroid coordinate (X1, Y1, Z1) = (144857.4, -87.80565, -417.01324) when the parameters of radial deformation change.
Fig. 2 (a) and fig. 2 (b) are three-dimensional frequency response curves of different faults at different angles. Because the traditional frequency response method only considers amplitude and frequency, the three-dimensional frequency response curve provided by the invention simultaneously considers frequency, amplitude and phase values, and the state quantity extracted from the obtained curve is more reliable than that of the traditional method.
Compared with the traditional method that the winding state is analyzed only through the frequency response amplitude, the built three-dimensional frequency response curve has more characteristics, and the recognition accuracy of the frequency response method is improved.
According to the obtained three-dimensional frequency response curves, calculating the mass center of each three-dimensional frequency response curve, wherein the calculated mass center formula is as follows: Wherein x, y and z represent frequency, amplitude and phase respectively; x o、yo、zo is the centroid; n represents the number of points taken. And according to the centroid calculation result, a three-dimensional coordinate system is established by taking the centroid of the three-dimensional frequency response curve of the normal winding as an origin. Three different fault three-dimensional frequency response curve centroids are distributed around it.
Fig. 3 (a) and 3 (b) show the centroid distribution diagrams of the three-dimensional frequency response curves at different angles. It is obvious from fig. 3 (a) and 3 (b) that the radial deformation represented by "+," + axis deviation represented by "+,".
Fig. 3 (a) and 3 (b) show radial deformation (Radialdeformation, RD), axial Offset (AO), and Axial deformation (Axial deformation, AD), respectively. As can be seen from fig. 3 (a) and 3 (b), the coordinates of the centroid obtained from the frequencies are the same, and after a three-dimensional cartesian coordinate system is established with the centroid of the three-dimensional frequency response curve of the normal winding as the origin, the coordinate values of the frequency response axes are all 0. This is because the frequency points selected in the simulation are all constant. The results shown in fig. 3 (a) and fig. 3 (b) show that the positions of the centroids in different fault states are regular. Therefore, the three-dimensional Cartesian coordinate system is projected, only the amplitude and the phase are reserved, and the two-dimensional centroid distribution diagram is drawn by dimension reduction. The shift of the three-dimensional curve of the different state windings is characterized by the centroid shift.
Fig. 4 shows a centroid offset schematic. As shown in fig. 4, the three-dimensional centroid coordinates of the winding in a normal state are taken as an origin (x o,yo), and the three-dimensional centroid coordinates of the winding in a fault state are taken as (x i,yi) and distributed around the winding. It is divided into four sections according to the cartesian coordinate system specification.
Since the x coordinate, i.e., the frequency coordinate of each fault three-dimensional curve centroid becomes 0 after the normal frequency winding three-dimensional frequency response curve centroid is taken as the origin. The centroid of the three-dimensional frequency response curve is projected to be changed into two dimensions for analysis.
Fig. 5 is a diagram of a three-dimensional frequency response curve centroid offset. As shown in fig. 5, it is apparent that the distribution of different faults is regular, and three faults are distributed in different intervals, respectively. As can be seen from fig. 5, the radial deformation is in zone 1, the axial displacement is in zone 4, and the axial deformation is in zone 3. The method for analyzing the faults of the transformer winding based on the centroid characteristics of the three-dimensional frequency response curve is proved to be capable of being used for diagnosing the faults of the transformer winding, and the degree of distinction is quite obvious.
Claims (5)
1. The transformer winding deformation fault diagnosis method based on the three-dimensional frequency response curve centroid analysis is characterized by comprising the following steps of:
Step 1: establishing a transformer winding concentrated parameter model, injecting a sinusoidal sweep frequency signal into the transformer winding concentrated parameter model, and obtaining winding input voltage U i (j omega) and output voltage U 0 (j omega) to obtain normal winding amplitude frequency data and phase frequency data;
Step 2: adjusting circuit parameters of a concentrated parameter model of a transformer winding, realizing various fault deformation simulation of the winding, and acquiring amplitude and phase data of the winding in different fault states;
Step 3: presenting the frequency, amplitude and phase acquired in the step 1 and the step 2 as x, y and z values in a three-dimensional Cartesian coordinate system, and establishing a three-dimensional frequency response curve;
step 4: by the formula Wherein x, y and z respectively represent frequency, amplitude and phase, x o、yo、zo is the mass center, and n represents the number of points taken; respectively calculating the values of the centroid coordinates x, y and z represented by the frequency, the amplitude and the phase, and visualizing the obtained coordinates (x, y and z) to obtain centroid distribution of each three-dimensional frequency response curve;
step 5: according to the three-dimensional curve centroid result obtained in the step 4, a three-dimensional coordinate system is established by taking the normal winding three-dimensional frequency response curve centroid as an origin, the centroids of different fault three-dimensional frequency response curves are distributed around the three-dimensional coordinate system, and the fault types of the windings are determined according to the differences of centroid distribution intervals;
In the step 5, the x coordinates of the mass centers obtained by the frequency are the same, and after a three-dimensional Cartesian coordinate system is established by taking the mass center of the three-dimensional frequency response curve of the normal winding as an origin, the coordinate values of the frequency response axes are all 0; projecting a three-dimensional Cartesian coordinate system, only reserving amplitude and phase, namely a y-axis coordinate and a z-axis coordinate, and performing dimension reduction on the three-dimensional Cartesian coordinate system to draw a two-dimensional centroid distribution map; characterizing the offset of the three-dimensional frequency response curve under different faults through centroid offset;
In the step 5, a cartesian two-dimensional coordinate system of centroid distribution is divided into four sections according to a two-dimensional centroid distribution diagram drawn by amplitude and phase coordinates, three faults of axial deformation, radial deformation and axis deviation are respectively distributed in different sections, and fault types are classified according to the difference of the obtained centroid distribution sections.
2. The transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis according to claim 1, wherein the method comprises the following steps of: in the step 1, a transformer winding concentrated parameter model is established by using circuit simulation software PSpice, and a sine sweep frequency signal with the frequency range of 1kHZ and 1MHZ is injected into a head end winding of the transformer winding concentrated parameter model.
3. The transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis according to claim 1, wherein the method comprises the following steps of: in the step 2, different percent circuit parameters are used for carrying out deformation simulation of various fault degrees of the winding, and 3 types of fault types are simulated by taking 5% as a gradient within the range of 5% -95% parameter variation: axial deformation, radial deformation and axial displacement, 19 failure degrees per class.
4. The transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis according to claim 1, wherein the method comprises the following steps of: in the step 3, the frequency of the sweep frequency signal is used as an x-axis coordinate value, the frequency response amplitude is used as a y-axis coordinate value, and the frequency response phase value is used as a z-axis coordinate value, so that a three-dimensional frequency response curve graph is established.
5. The transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis according to claim 1, wherein the method comprises the following steps of: in the step 5, if the centroid distribution result has the following characteristics:
① . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 1, radial deformation of the transformer winding occurs;
② . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 3, the transformer winding is axially deformed;
③ . If the centroid of the obtained three-dimensional frequency response curve is located in the interval 4, the transformer winding generates axis deviation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210121763.6A CN114548012B (en) | 2022-02-09 | 2022-02-09 | Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210121763.6A CN114548012B (en) | 2022-02-09 | 2022-02-09 | Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114548012A CN114548012A (en) | 2022-05-27 |
| CN114548012B true CN114548012B (en) | 2024-07-05 |
Family
ID=81673105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210121763.6A Active CN114548012B (en) | 2022-02-09 | 2022-02-09 | Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114548012B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115389985A (en) * | 2022-09-07 | 2022-11-25 | 国网安徽省电力有限公司电力科学研究院 | Frequency response method-based transformer winding deformation degree online detection method |
| CN115856724B (en) * | 2022-11-24 | 2024-07-12 | 西南交通大学 | Transformer winding fault identification method considering temperature factors |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543730A (en) * | 2019-09-05 | 2019-12-06 | 长沙理工大学 | Transformer winding deformation fault determination method |
| CN111983524A (en) * | 2020-08-26 | 2020-11-24 | 西南交通大学 | Transformer winding fault assessment method based on oscillatory wave time-frequency transformation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110514924B (en) * | 2019-08-12 | 2021-04-27 | 武汉大学 | Power transformer winding fault positioning method based on deep convolutional neural network fusion visual identification |
-
2022
- 2022-02-09 CN CN202210121763.6A patent/CN114548012B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543730A (en) * | 2019-09-05 | 2019-12-06 | 长沙理工大学 | Transformer winding deformation fault determination method |
| CN111983524A (en) * | 2020-08-26 | 2020-11-24 | 西南交通大学 | Transformer winding fault assessment method based on oscillatory wave time-frequency transformation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114548012A (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114548012B (en) | Transformer winding deformation fault diagnosis method based on three-dimensional frequency response curve centroid analysis | |
| US20210231584A1 (en) | Smart defect calibration system in semiconductor wafer manufacturing | |
| CN110119680B (en) | Automatic error checking system of regulator cubicle wiring based on image recognition | |
| US7065461B2 (en) | Point cloud measuring system and method | |
| CN103759758B (en) | A kind of method for detecting position of the automobile meter pointer based on mechanical angle and scale identification | |
| CN105701766A (en) | Image matching method and device | |
| CN110909738A (en) | Automatic reading method of pointer instrument based on key point detection | |
| CN110211178A (en) | A kind of pointer instrument recognition methods calculated using projection | |
| CN109241043A (en) | A kind of data quality checking method and device | |
| CN113570550B (en) | Gold wire bonding size detection method based on three-dimensional point cloud | |
| CN114067307A (en) | Transformer substation instrument scale calibration method and system | |
| CN109115127A (en) | A kind of sub-pix peak point extraction algorithm based on Bezier | |
| CN116678368B (en) | BIM technology-based intelligent acquisition method for assembled steel structure data | |
| CN110030937A (en) | Motor-car rod piece measurement method based on artificial intelligence | |
| CN114973017A (en) | High-precision positioning method and system for power transmission and transformation equipment integrating remote sensing and GIS technology | |
| CN107421440A (en) | A kind of method of three dimensional optical measuring square hole vertical difference | |
| CN113869112A (en) | Instrument automatic reading method and device based on machine vision | |
| CN110021027A (en) | A kind of trimming point calculating method based on binocular vision | |
| CN111191562B (en) | Pointer instrument indication value reading method based on CNN neural network | |
| CN110411372A (en) | It is a kind of that room collecting method is tested based on total station | |
| CN115656632A (en) | Curve scanning track tracing and comparing algorithm suitable for impedance analyzer | |
| CN109444622A (en) | A kind of shaft angle sends system failure automatic testing method, system and device | |
| Sun et al. | Research and Development of Integral Test System for Transformer Calibrator Based on Machine Vision and Servo Control | |
| CN113393450B (en) | Data quality inspection method and system for digital line drawing and readable storage medium | |
| CN108098120A (en) | A kind of evaluating method for welding wire feed rate stability |
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 |