CN117171616A - Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement - Google Patents
Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement Download PDFInfo
- Publication number
- CN117171616A CN117171616A CN202311128203.4A CN202311128203A CN117171616A CN 117171616 A CN117171616 A CN 117171616A CN 202311128203 A CN202311128203 A CN 202311128203A CN 117171616 A CN117171616 A CN 117171616A
- Authority
- CN
- China
- Prior art keywords
- voltage
- data
- fault
- voltage transformer
- graph
- 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.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000003860 storage Methods 0.000 title claims abstract description 12
- 238000003062 neural network model Methods 0.000 claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000013499 data model Methods 0.000 claims abstract description 16
- 230000001052 transient effect Effects 0.000 claims abstract description 16
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 8
- 230000002708 enhancing effect Effects 0.000 claims abstract description 8
- 238000005096 rolling process Methods 0.000 claims abstract description 6
- 239000003990 capacitor Substances 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 19
- 238000005070 sampling Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 230000015556 catabolic process Effects 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 17
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000004075 alteration Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000007621 cluster analysis Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The embodiment of the application provides a voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement, which comprises the following steps: acquiring historical voltage data of a secondary side of the voltage transformer; establishing a transient data model of the voltage transformer, and enhancing fault voltage data; constructing a graph rolling neural network model, and inputting the enhanced fault voltage data into the graph rolling neural network model for training; performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolution neural network model; according to the application, the fault of the voltage transformer is diagnosed through the graph convolutional neural network model, the complexity degree and the calculated amount of the model depend on the size of the image, and the complexity degree and the calculated amount are irrelevant to the selection of the historical data, so that the excessive calculated amount caused by excessive historical data can be effectively avoided; a large amount of fault data is obtained by enhancing the fault voltage data, so that a model is optimized, and the fault diagnosis rate is improved.
Description
Technical Field
The application belongs to the technical field of voltage transformers, and particularly relates to a voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement.
Background
The capacitive voltage transformer is widely used as important equipment for metering, measuring and relay protection of a 500kV transformer substation, and whether the potential fault hazard exists or not directly affects the safe and stable operation of a power grid. Various sensors such as temperature, infrared, vibration, gas sensitivity and humidity sensitivity are added to monitor state characteristic quantity signals reflecting CVT faults, but the method needs to additionally add sensors, and the types, the numbers and the positions of the sensors cannot be installed as required under the safety constraint of a power grid, so that the required fault state characteristic quantity signals cannot be obtained.
The secondary side voltage of the CVT can be obtained by using a D5000 intelligent power grid dispatching system (hereinafter referred to as a D5000 system), and can be used as a fault state characteristic quantity signal to realize CVT fault diagnosis through data processing and analysis. The common data processing and analyzing method has the advantages of a threshold method and a cluster analysis method, the threshold method has the advantages of clear physical meaning, easy realization and the like, but has the defect of difficult selection of a critical value, frequently causes missed diagnosis and misdiagnosis of faults due to improper selection of the critical value, and in addition, the traditional threshold method only depends on current time information and can not effectively utilize history information, so the method has no global property; the cluster analysis method classifies voltages into two types, and fault diagnosis is achieved by comparing the voltages with the Euclidean distance between centers of the two types, but when the data amount is large, it is difficult to determine that the classification into several types is optimal. Based on the above problems, a more sophisticated method is needed to solve the fault diagnosis problem of the voltage transformer.
Disclosure of Invention
In order to solve one of the technical defects, the embodiment of the application provides a voltage transformer fault diagnosis method, device and storage medium based on data enhancement, which can improve the accuracy of the voltage transformer fault diagnosis.
According to a first aspect of the embodiment of the present application, there is provided a fault diagnosis method for a voltage transformer based on data enhancement, including:
acquiring historical voltage data of a secondary side of the voltage transformer;
establishing a transient data model of the voltage transformer, and enhancing fault voltage data;
constructing a graph rolling neural network model, and inputting the enhanced fault voltage data into the graph rolling neural network model for training;
and performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolution neural network model.
Preferably, the establishing a transient data model of the voltage transformer, and the enhancing fault voltage data includes:
deducing an equivalent circuit of the voltage transformer according to the physical structure of the voltage transformer, and constructing a transient data model of the voltage transformer;
converting historical voltage data of a secondary side of the voltage transformer into primary side, and inputting the converted primary side voltage data into a transient data model of the voltage transformer;
breakdown of different degrees occurs to a high-voltage capacitor and a medium-voltage capacitor in the voltage transformer at different moments, so that enhanced fault voltage data are obtained;
and processing the enhanced fault voltage data to obtain a sample graph with enhanced data, and extracting fault state characteristic quantity according to the sample graph to obtain a voltage picture with the fault state characteristic quantity.
Preferably, the processing the enhanced fault voltage data to obtain a sample graph with enhanced data, extracting a fault state feature quantity according to the sample graph, and obtaining a voltage picture with the fault state feature quantity includes:
obtaining a sample graph after data enhancement at intervals of T of each fault according to the enhanced fault voltage data, wherein the sample graph comprises a normal data picture and a fault data picture;
and calculating the time domain statistical state quantity by the sample graph according to the sampling window w and the step length s to obtain the voltage picture with the fault state characteristic quantity.
Preferably, the step of inputting the enhanced fault voltage data into the graph convolution neural network model for training is specifically: and inputting the voltage picture with the fault state characteristic quantity into a graph convolutional neural network model for training.
Preferably, the structure for constructing the graph roll-up neural network model comprises 1 convolution layer, 1 pooling layer, 1 convolution layer, one pooling layer, one flattening layer, one full-connection layer number 128 and 1 full-connection layer number 1.
Preferably, the fault diagnosis for the voltage data of the current voltage transformer by using the trained graph convolution neural network model includes:
acquiring secondary side voltage data of a current voltage transformer;
calculating time domain statistical state quantity by using a sampling window w and a step length s to finish the extraction of fault state characteristic quantity and obtain a current voltage picture with the fault state characteristic quantity;
and inputting the obtained voltage picture into a GNN model for fault diagnosis.
Preferably, the converting the historical voltage data of the secondary side of the voltage transformer to the primary side includes: converting historical voltage data of a secondary side of the voltage transformer to a primary side through the voltage transformer (1):
in the formula (1), C 1 Is of voltageHigh-voltage capacitor in mutual inductor, C 2 Is the medium voltage capacitor in the voltage transformer, u 1 For the converted primary-side input voltage, C e Is equivalent capacitance.
Preferably, the time domain statistical state quantity includes a maximum value, a minimum value, a peak-to-peak value, an average value, an absolute average value, a root mean square, a standard deviation, a skewness, a variance, a kurtosis, a coefficient of variation, a peak factor, a waveform factor, a pulse factor, a square root amplitude, a margin index, and a clearance factor. A kind of module is assembled in the module and the module is assembled in the module.
According to a second aspect of an embodiment of the present application, there is provided a computer apparatus comprising:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the data enhancement based voltage transformer fault diagnosis method of any of the above.
According to a third aspect of an embodiment of the present application, there is provided a computer-readable storage medium having a computer program stored thereon; the computer program is executed by a processor to implement the data enhancement based voltage transformer fault diagnosis method as claimed in any one of the above.
By adopting the voltage transformer fault diagnosis method, the voltage transformer fault diagnosis equipment and the storage medium based on data enhancement provided by the embodiment of the application, a voltage transformer transient data model is built according to the basic structure of the voltage transformer, and fault voltage data is enhanced; training a graph convolution neural network model through the enhanced fault voltage data, and performing fault diagnosis on the voltage data of the current voltage transformer by utilizing the trained graph convolution neural network model; according to the application, the fault of the voltage transformer is diagnosed through the graph convolutional neural network model, the complexity degree and the calculated amount of the model depend on the size of the image, and the complexity degree and the calculated amount are irrelevant to the selection of the historical data, so that the excessive calculated amount caused by excessive historical data can be effectively avoided; according to the application, a large amount of fault data is obtained by enhancing the fault voltage data, so that the model is optimized, and the fault diagnosis rate is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a schematic flow chart of a fault diagnosis method for a voltage transformer based on data enhancement according to an embodiment of the present application;
fig. 2 is a schematic flow chart II of a fault diagnosis method for a voltage transformer based on data enhancement according to an embodiment of the present application;
fig. 3 is a basic structural diagram of a voltage transformer according to an embodiment of the present application;
fig. 4 is an equivalent circuit diagram of a voltage transformer according to an embodiment of the present application;
FIG. 5 is a schematic diagram of enhanced data of fault data according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a time domain statistics extraction failure feature sample according to an embodiment of the present application
FIG. 7 is a schematic diagram of an algorithm diagnosis result consisting of a training set of 468 measured normal eigenvalues and enhanced fault eigenvalues and a test set of 231 measured faults;
fig. 8 is a schematic diagram showing the summary of experimental results of each example.
Detailed Description
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following detailed description of exemplary embodiments of the present application is provided in conjunction with the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application and not exhaustive of all embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
In the process of realizing the application, the inventor finds that the diagnosis of the traditional voltage transformer needs a large amount of fault data for model training by adopting a neural network algorithm, and the fault data is not easy to obtain in practice.
In view of the above problems, an embodiment of the present application provides a fault diagnosis method for a voltage transformer based on data enhancement, as shown in fig. 1, where the fault diagnosis method includes:
s10, acquiring historical voltage data of a secondary side of a voltage transformer; the method comprises the following steps: carrying out data acquisition on bus voltage in a D5000 system of the 500kV transformer substation at a sampling frequency fs;
s20, establishing a transient data model of the voltage transformer, and enhancing fault voltage data;
s30, constructing a graph convolution neural network model, and inputting the enhanced fault voltage data into the graph convolution neural network model for training;
s40, performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolutional neural network model.
In the embodiment, data acquisition is performed on bus voltage in a D5000 system of a 500kV transformer substation at a sampling frequency of 1Hz, and 88500 data are assumed to be included, wherein the number of normal data is 74640, and the number of fault data is 13860 (from 74641 th point to 88500 th point); according to the basic structure of the voltage transformer, a transient data model of the voltage transformer is established, and fault voltage data are enhanced; training a graph convolution neural network model through the enhanced fault voltage data, and performing fault diagnosis on the voltage data of the current voltage transformer by utilizing the trained graph convolution neural network model;
according to the application, the fault of the voltage transformer is diagnosed through the graph convolutional neural network model, the complexity degree and the calculated amount of the model depend on the size of the image, and the complexity degree and the calculated amount are irrelevant to the selection of the historical data, so that the excessive calculated amount caused by excessive historical data can be effectively avoided; by adopting the neural network algorithm, a large amount of fault data is required for model training, and the fault data is not easy to obtain in practice.
The embodiment of the application provides a voltage transformer fault diagnosis method based on data enhancement, as shown in fig. 2, on the basis of fig. 1, a voltage transformer transient data model is established, and the enhanced fault voltage data comprises the following steps:
s201, deducing an equivalent circuit of the voltage transformer according to the physical structure of the voltage transformer, and constructing a transient data model of the voltage transformer; FIG. 3 is a basic structure diagram of a voltage transformer, and FIG. 4 is an equivalent circuit of the voltage transformer; as shown in fig. 3 and 4, the voltage transformer is composed of a capacitive voltage divider and an electromagnetic unit, wherein the capacitive voltage divider is composed of a high-voltage capacitor C1 and a medium-voltage capacitor C2, and an equivalent circuit diagram of the voltage transformer can be obtained according to a basic structure diagram of the voltage transformer, so that a mathematical model of the voltage transformer can be obtained. In FIG. 4, L k And R is k The inductance and the resistance of the compensating reactor and the intermediate transformer are respectively; l (L) m And R is m The inductance and the resistance of the excitation branch of the intermediate transformer are respectively; l (L) 2 And R is 2 The inductance and the resistance are respectively converted into a primary side load; c (C) f 、R f 、L f 、r f The capacitor, the resistor, the inductor and the small resistor of the damper are respectively arranged; i.e m Is L m 、R m The sum of the currents flowing through; i.e 1 、i f 、i 2 Respectively is R k 、R f 、R 2 Is set to be a current of (a); u (u) 2 The converted secondary side output voltage is obtained, and W is the number of turns of the primary side of the intermediate transformer; f (ψm) is the excitation characteristic of the intermediate transformer.
As shown in fig. 4, the differential equation of state of the voltage transformer circuit is derived from the equivalent circuit of the voltage transformer as follows:
wherein the method comprises the steps of：
S202, converting historical voltage data of a secondary side of the voltage transformer into primary side through the method (1), and inputting the converted primary side voltage data into a transient data model of the voltage transformer;
in the formula (1), C 1 Is the high-voltage capacitor in the voltage transformer, C 2 Is the medium voltage capacitor in the voltage transformer, u 1 For the converted primary-side input voltage, C e Is equivalent capacitance; in this embodiment, the voltage transformer is used in a transformer station with a voltage class of 500kV, so that the primary side line voltage is 500kV, the phase voltage is 500/. V3 kV, and the secondary side voltage of the voltage transformer is a fixed value of 100V, so that the transformation ratio is 500×10 3 V 3:100; according to a transformation ratio of 500 x 10 3 V 3:100, converting the historical voltage of the CVT secondary side obtained by the D5000 system, including normal data and fault data, to the primary side through the formula (1), and then inputting the historical voltage into a transient data model of a voltage transformer.
S203, breakdown of the high-voltage capacitor and the medium-voltage capacitor in the voltage transformer occurs to different degrees at different moments, enhanced fault voltage data are obtained, and the enhanced fault voltage data are shown in fig. 5; illustrating: the voltage amplitude in the sampling interval is unchanged, so that the high-voltage capacitor C1 and the medium-voltage capacitor C2 of the voltage transformer respectively break down (age) differently, the high-voltage capacitor C1 and the medium-voltage capacitor C2 respectively occur in 7 th minutes, the degree is 1%, 2% and … …%, and 20 faults are generated in total; the high-voltage capacitor C1 occurs at the 7 th minute, the 12 th minute and the 17 th minute respectively, and the degrees of the three time points are respectively 2%, 4%, 6% and 1%, 3% and 6%, and the two groups have 6 faults; the medium voltage capacitor C2 occurs at the 7 th minute, the 12 th minute and the 17 th minute respectively, and the degrees of three time points are respectively 1%, 2%, 3% and 1%, 3% and 6%, and the two groups have 6 faults in total; the high-voltage capacitor C1 occurs at the 7 th minute and the 10 th minute respectively, the degrees of the two time points are 1%, 2%,1%, 4% and 2%, 4% respectively, and three groups have 6 faults in total; the medium voltage capacitor C2 occurs at 7 and 10 minutes, respectively, with two time points levels of 1%, 2%,1%, 3% and 2%, 4%, respectively, three groups of 6 faults altogether; and adding up 44 faults to obtain data with enhanced fault data.
S204, processing the enhanced fault voltage data to obtain a sample graph with enhanced data, and extracting fault state characteristic quantities according to the sample graph to obtain a voltage picture with the fault state characteristic quantities.
The embodiment realizes fault data enhancement and solves the problems of insufficient training of the graph convolution neural network model and low fault detection rate caused by the lack of fault data.
Further, processing the enhanced fault voltage data to obtain a sample graph with enhanced data, extracting fault state feature quantity according to the sample graph, and obtaining a voltage picture with the fault state feature quantity, wherein the method comprises the following steps:
obtaining a sample graph after data enhancement at intervals of T of each fault according to the enhanced fault voltage data, wherein the sample graph comprises a normal data picture and a fault data picture; the following is exemplified according to the data enhanced by the fault data: taking 30 minutes as a sampling interval, obtaining a picture every 1 minute of each fault, and obtaining 29 sample pictures after data enhancement, wherein each picture contains history information of 5 minutes, and since step S203 is to set faults from the 7 th minute, the first three pictures do not contain fault information, and 1-5min,2-6min,3-7min are free of fault information, so that 3 normal data pictures and 26 fault data pictures can be obtained;
calculating time domain statistical state quantity by a sampling window w and a step length s to obtain a voltage picture with fault state characteristic quantity, wherein the voltage picture is shown in fig. 6; the time domain statistical state quantity comprises a maximum value, a minimum value, a peak value, an average value, an absolute average value, a root mean square, a standard deviation, a skewness, a variance, a kurtosis, a variation coefficient, a peak factor, a waveform factor, a pulse factor, a square root amplitude value, a margin index and a clearance factor; illustrating: according to each of the 29 sample images after data enhancement, taking a sampling window of 300 seconds, a step length of 60 seconds, calculating 17 time domain statistical state quantities such as maximum value and the like, and completing extraction of fault state characteristic quantities to obtain a voltage image with the fault state characteristic quantities; the method and the device solve the problem that the fault diagnosis of the voltage transformer can not be realized under complex and changeable power grid operation conditions by only using the secondary side voltage of the voltage transformer as a fault state characteristic quantity signal by extracting time domain statistics capable of reflecting the fault time characteristics of the voltage transformer.
Further, the training of the enhanced fault voltage data input into the graph convolution neural network model is specifically as follows: inputting the voltage picture with the fault state characteristic quantity into a graph convolutional neural network model for training; specifically, the fault diagnosis rate of the model is detected under 6 different calculation examples, the iteration times are 50 times, the initial learning rate is 0.006, the memory is 16GB, and the operation time is 195 seconds; in this embodiment, the more the historical data of the graph convolution neural network model is trained, the higher the fault diagnosis accuracy is, and compared with the traditional neural network, the increase of the historical data does not increase the calculation amount of the graph convolution neural network model, so that the CVT fault diagnosis through the graph convolution neural network model has extremely strong practicability.
Further, the structure for constructing the graph convolutional neural network model comprises 1 convolutional layer, 1 pooling layer, 1 convolutional layer, one pooling layer, one flattening layer, one full-connection layer number 128 and 1 full-connection layer number 1.
Further, performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolution neural network model, including: acquiring secondary side voltage data of a current voltage transformer; calculating time domain statistical state quantity by using a sampling window w and a step length s to finish the extraction of fault state characteristic quantity and obtain a current voltage picture with the fault state characteristic quantity; the time domain statistical state quantity comprises a maximum value, a minimum value, a peak value, an average value, an absolute average value, a root mean square, a standard deviation, a skewness, a variance, a kurtosis, a variation coefficient, a peak factor, a waveform factor, a pulse factor, a square root amplitude value, a margin index and a clearance factor; inputting the obtained voltage picture into a GNN model for fault diagnosis; for example, according to the current voltage data, taking a sampling window of 300 seconds, a step length of 60 seconds, calculating 17 time domain statistical state quantities such as the maximum value, completing the extraction of fault state characteristic quantities, obtaining a voltage picture with the fault state characteristic quantities, inputting the picture into a graph convolution neural network model, and judging whether the fault exists.
According to the application, the secondary side voltage of the voltage transformer is formed into a group of images in a certain sampling window, and is used as a graph neural network, the complexity and the calculated amount of the graph neural network model depend on the size of the images, and the complexity and the calculated amount are irrelevant to the selection of historical data, so that the excessive calculated amount caused by excessive historical data can be effectively avoided; by adopting a neural network algorithm, a large amount of fault data is required for model training, the fault data is not easy to obtain in practice, and the problem is effectively solved by small sample learning; fig. 7 is a schematic diagram of an algorithm diagnosis result composed of a training set of 468 pieces of actually measured normal feature values and enhanced fault feature values and a test set of 231 pieces of actually measured faults, and fig. 8 is a schematic diagram showing summary of experimental results of each example; as shown in fig. 7 and 8, the experimental results show that the diagnosis method provided by the application has small calculated amount and good performance, effectively solves the problem that the fault diagnosis method of the existing voltage transformer has relatively high misdiagnosis rate, and is suitable for fault diagnosis of the voltage transformer.
A computer device, comprising:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the data enhancement based voltage transformer fault diagnosis method as described above.
A computer-readable storage medium having a computer program stored thereon; the computer program is executed by a processor to implement the data enhancement based voltage transformer fault diagnosis method as described above.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The scheme in the embodiment of the application can be realized by adopting various computer languages, such as C language, VHDL language, verilog language, object-oriented programming language Java, an transliteration script language JavaScript and the like.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. The fault diagnosis method of the voltage transformer based on data enhancement is characterized by comprising the following steps of:
acquiring historical voltage data of a secondary side of the voltage transformer;
establishing a transient data model of the voltage transformer, and enhancing fault voltage data;
constructing a graph rolling neural network model, and inputting the enhanced fault voltage data into the graph rolling neural network model for training;
and performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolution neural network model.
2. The method for diagnosing a fault of a voltage transformer based on data enhancement according to claim 1, wherein the establishing a transient data model of the voltage transformer, enhancing the fault voltage data comprises:
deducing an equivalent circuit of the voltage transformer according to the physical structure of the voltage transformer, and constructing a transient data model of the voltage transformer;
converting historical voltage data of a secondary side of the voltage transformer into primary side, and inputting the converted primary side voltage data into a transient data model of the voltage transformer;
breakdown of different degrees occurs to a high-voltage capacitor and a medium-voltage capacitor in the voltage transformer at different moments, so that enhanced fault voltage data are obtained;
and processing the enhanced fault voltage data to obtain a sample graph with enhanced data, and extracting fault state characteristic quantity according to the sample graph to obtain a voltage picture with the fault state characteristic quantity.
3. The method for diagnosing a fault of a voltage transformer based on data enhancement according to claim 2, wherein the processing the enhanced fault voltage data to obtain a sample graph with enhanced data, extracting a fault state feature according to the sample graph, and obtaining a voltage picture with the fault state feature comprises:
obtaining a sample graph after data enhancement at intervals of T of each fault according to the enhanced fault voltage data, wherein the sample graph comprises a normal data picture and a fault data picture;
and calculating the time domain statistical state quantity by the sample graph according to the sampling window w and the step length s to obtain the voltage picture with the fault state characteristic quantity.
4. The method for diagnosing a fault of a voltage transformer based on data enhancement according to claim 2, wherein the step of inputting the enhanced fault voltage data into a graph convolution neural network model for training is specifically as follows: and inputting the voltage picture with the fault state characteristic quantity into a graph convolutional neural network model for training.
5. The data enhancement-based voltage transformer fault diagnosis method according to claim 1, wherein the structure for constructing the graph roll-up neural network model comprises 1 convolution layer, 1 pooling layer, 1 convolution layer, one pooling layer, one flattening layer, one full connection layer number 128 and 1 full connection layer number 1.
6. The method for diagnosing a fault of a voltage transformer based on data enhancement according to claim 1, wherein the step of performing fault diagnosis on the voltage data of the current voltage transformer by using the trained graph convolution neural network model comprises the steps of:
acquiring secondary side voltage data of a current voltage transformer;
calculating time domain statistical state quantity by using a sampling window w and a step length s to finish the extraction of fault state characteristic quantity and obtain a current voltage picture with the fault state characteristic quantity;
and inputting the obtained voltage picture into a GNN model for fault diagnosis.
7. The method for diagnosing a fault in a voltage transformer based on data enhancement according to claim 2, wherein converting the historical voltage data of the secondary side of the voltage transformer to the primary side comprises: converting historical voltage data of a secondary side of the voltage transformer to a primary side through the voltage transformer (1):
in the formula (1), C 1 Is the high-voltage capacitor in the voltage transformer, C 2 Is the medium voltage capacitor in the voltage transformer, u 1 For the converted primary-side input voltage, C e Is equivalent capacitance.
8. The data enhancement-based voltage transformer fault diagnosis method according to claim 3 or 6, wherein the time domain statistical state quantity comprises a maximum value, a minimum value, a peak-to-peak value, an average value, an absolute average value, a root mean square, a standard deviation, a skewness, a variance, a kurtosis, a coefficient of variation, a peak factor, a waveform factor, a pulse factor, a square root amplitude, a margin index, and a clearance factor.
9. A computer device, comprising:
a memory;
a processor; and
a computer program;
wherein the computer program is stored in the memory and configured to be executed by the processor to implement the data enhancement-based voltage transformer fault diagnosis method of any one of claims 1 to 8.
10. A computer-readable storage medium, characterized in that a computer program is stored thereon; the computer program is executed by a processor to implement the data enhancement-based voltage transformer fault diagnosis method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311128203.4A CN117171616A (en) | 2023-09-01 | 2023-09-01 | Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311128203.4A CN117171616A (en) | 2023-09-01 | 2023-09-01 | Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117171616A true CN117171616A (en) | 2023-12-05 |
Family
ID=88935938
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311128203.4A Pending CN117171616A (en) | 2023-09-01 | 2023-09-01 | Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117171616A (en) |
-
2023
- 2023-09-01 CN CN202311128203.4A patent/CN117171616A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11474163B2 (en) | Power transformer winding fault positioning method based on deep convolutional neural network integrated with visual identification | |
| Abbasi et al. | Diagnosis and clustering of power transformer winding fault types by cross‐correlation and clustering analysis of FRA results | |
| CN110780146B (en) | Transformer fault identification and positioning diagnosis method based on multi-stage transfer learning | |
| Behjat et al. | Statistical approach for interpretation of power transformers frequency response analysis results | |
| Duan et al. | Method of inter‐turn fault detection for next‐generation smart transformers based on deep learning algorithm | |
| Saleh et al. | Proposed diagnostic methodology using the cross‐correlation coefficient factor technique for power transformer fault identification | |
| CN108196143A (en) | Power Transformer Faults depth diagnostic method and terminal device | |
| Bagheri et al. | Effect of transformer winding mechanical defects, internal and external electrical faults and inrush currents on performance of differential protection | |
| Nosratian Ahour et al. | Determination and localisation of turn‐to‐turn fault in transformer winding using frequency response analysis | |
| WO2023083366A1 (en) | Method and apparatus for identifying running state of lightning arrester on basis of on-line monitoring data | |
| Jin-qiang | Fault prediction of a transformer bushing based on entropy weight TOPSIS and gray theory | |
| CN112180204A (en) | Power grid line fault diagnosis method based on electric quantity information | |
| CN112069930A (en) | Vibration signal processing method and device for improving GIS equipment fault diagnosis accuracy | |
| Guomin et al. | Deep learning‐based fault location of DC distribution networks | |
| Moradzadeh et al. | Location of disk space variations in transformer winding using convolutional neural networks | |
| CN116008731A (en) | Power distribution network high-resistance fault identification method and device and electronic equipment | |
| Zhou et al. | Detection of transformer winding faults using FRA and image features | |
| Samet et al. | A statistical-based criterion for incipient fault detection in underground power cables established on voltage waveform characteristics | |
| CN116990625B (en) | Function switching system and method of intelligent quick-checking device of distribution transformer | |
| CN105488270A (en) | Multiattribute comprehensive method for structural fault diagnosis of transformer | |
| Chen et al. | Location for single‐phase grounding fault in distribution network based on equivalent admittance distortion rate | |
| CN105741184A (en) | Transformer state evaluation method and apparatus | |
| Bin et al. | Research on mechanical fault diagnosis method of power transformer winding | |
| CN117171616A (en) | Voltage transformer fault diagnosis method, equipment and storage medium based on data enhancement | |
| Daryani et al. | Evidence theory‐based identification of aging for capacitive voltage transformers |
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 |