Transformer partial discharge identification method and system based on phase holding data enhancement

Abstract

The invention discloses a transformer partial discharge identification method and system based on phase preserving data enhancement, and the method comprises the steps: collecting the pulse discharge information of a transformer, converting the pulse discharge information into a first map, and carrying out the first preprocessing of the first map, and obtaining a second map; performing second preprocessing on the second atlas, dividing the second atlas into a training set and a test set, and performing data enhancement processing on the training set to generate a third atlas; and constructing a transformer partial discharge identification model, carrying out training based on the third atlas, obtaining an optimized transformer partial discharge identification model, and identifying a partial discharge type based on the optimized transformer partial discharge identification model. According to the method, the key phase information of the PRPD atlas is reserved, the recognition accuracy of the model and the recognition accuracy and robustness of various partial discharge types are improved, the method can better adapt to data changes in practical application, and the engineering applicability and reliability of partial discharge detection and fault diagnosis of power equipment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of partial discharge detection and fault diagnosis of electric power equipment, and in particular to a transformer partial discharge identification method and system based on phase retention data enhancement. Background Art

[0002] At present, transformer partial discharge detection is mainly achieved through ultra-high frequency (UHF), ultrasonic method, optical method and oil-soluble gas analysis method. Although these methods can effectively detect the existence of partial discharge phenomenon, they still face many challenges in identifying and classifying different types of partial discharge. With the rapid development of deep learning technology, convolutional neural networks (CNN) have made significant progress in the field of image classification and recognition. CNN can autonomously learn and extract multi-level features of image data, avoiding the limitations of manual feature selection in traditional methods, so its application in discharge pattern recognition is becoming more and more extensive. In particular, the phase resolved partial discharge (PRPD) mode based on phase analysis by drawing the phase distribution of transformer partial discharge voltage signal is more mature and stable than other analysis modes.

[0003] Since the PRPD spectrum has clear phase distribution characteristics, the existing discharge pattern recognition methods based on PRPD spectrum use common methods such as rotation, folding and flipping to enhance the image data after preprocessing the PRPD spectrum data, which will destroy the phase information in the spectrum, resulting in the model being unable to accurately learn the key features of the discharge pattern during training, resulting in feature loss, thus affecting the accuracy of recognition. Therefore, how to enrich the model's training samples through data enhancement without destroying the phase information, and improve the model's generalization ability and recognition accuracy, has become an urgent problem to be solved. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a transformer partial discharge identification method and system based on phase-preserving data enhancement to solve the problems mentioned in the background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a transformer partial discharge identification method based on phase-preserving data enhancement, comprising: collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum;

[0009] Performing a second preprocessing on the second atlas and dividing it into a training set and a test set, performing data enhancement processing on the training set, and generating a third atlas;

[0010] A transformer partial discharge identification model is constructed, and training is performed based on the third graph to obtain an optimized transformer partial discharge identification model, and the partial discharge type is identified based on the optimized transformer partial discharge identification model.

[0011] As a preferred solution of the transformer partial discharge identification method based on phase-preserving data enhancement described in the present invention, wherein: the data enhancement processing of the training set to generate the third spectrum includes: adding Gaussian noise to the second spectrum, recording the second spectrum as I(x, y), where x and y represent the rows and columns of the image respectively, generating a Gaussian noise matrix of the same size as the first spectrum, recorded as N(x, y), each element of which obeys a normal distribution N(μ, σ 2 ), the calculation formula is expressed as:

[0012]

[0013] Among them, I a (x, y) is the third spectrum after adding Gaussian noise, I b (x, y) is the third atlas data after the pixel value is enhanced within the legal range through cropping operation.

[0014] As a preferred solution of the transformer partial discharge identification method based on phase retention data enhancement described in the present invention, it also includes: performing phase drift processing on the second spectrum, translating the phase axis of the spectrum within a specific angle range, expressed as:

[0015] I c (x,y)=I(x,(y+Δy)mod360)

[0016] Among them, I c (x, y) is the third spectrum after adding phase drift processing, and Δy is the shift phase angle.

[0017] As a preferred solution of the transformer partial discharge identification method based on phase-preserving data enhancement described in the present invention, it also includes: performing segmented disturbance processing on the second spectrum to change the amplitude intensity of the local area of ​​the spectrum, which is expressed as:

[0018]

[0019] Among them, I d (x, y) is the third spectrum after adding segmented perturbation processing, i is the current segment number, n is the total number of segments, α i is the disturbance factor of the i-th segment.

[0020] As a preferred solution of the transformer partial discharge identification method based on phase-preserving data enhancement described in the present invention, the transformer partial discharge identification model is constructed, including: 3 convolutional layers and corresponding 3 pooling layers, 3 nonlinear activation layers, 2 fully connected layers and a 50% Dropout layer in the fully connected layers, wherein the pooling layer selects the maximum pooling function and the activation function selects the ReLU function.

[0021] As a preferred solution of the transformer partial discharge identification method based on phase-preserving data enhancement described in the present invention, wherein: training is performed based on the third graph to obtain an optimized transformer partial discharge identification model, including: the batch size of the training set and the test set is 32, the learning rate is 0.01, the training cycle is 100, the classification loss function is the cross entropy loss function, the optimization algorithm is the stochastic gradient descent algorithm, with minimizing the loss function as the goal, and the model parameters are iteratively updated after each training cycle to obtain the optimized transformer partial discharge identification model.

[0022] As a preferred solution of the transformer partial discharge identification method based on phase-preserving data enhancement described in the present invention, the partial discharge types identified based on the optimized transformer partial discharge identification model include: suspended discharge, corona discharge, insulation discharge and free discharge.

[0023] In a second aspect, the present invention provides a transformer partial discharge identification system based on phase-preserving data enhancement, comprising:

[0024] A first preprocessing module, used for collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum;

[0025] A second preprocessing module, used for performing a second preprocessing on the second atlas, dividing it into a training set and a test set, performing data enhancement processing on the training set, and generating a third atlas;

[0026] The partial discharge identification module is used to construct a transformer partial discharge identification model, and to perform training based on the third graph to obtain an optimized transformer partial discharge identification model, and to identify the partial discharge type based on the optimized transformer partial discharge identification model.

[0027] In a third aspect, the present invention provides an electronic device, comprising:

[0028] Memory and processor;

[0029] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the transformer partial discharge identification method based on phase retention data enhancement are implemented.

[0030] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the transformer partial discharge identification method based on phase-preserving data enhancement.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention can effectively retain the key phase information in the PRPD spectrum, ensure the diversity and consistency of the enhanced data, and thus improve the recognition accuracy of the model. Through the phase-preserving data enhancement technology, the model can better identify complex local discharge patterns. Compared with traditional methods, the recognition accuracy of various local discharge types is significantly improved, especially in complex noise and changing environments. The present invention optimizes the data enhancement technology so that the model has a stronger generalization ability when dealing with different discharge types, can better adapt to changes in data distribution in practical applications, avoid overfitting problems, and ensure that the model has reliable performance in actual engineering environments. Since the phase-preserving data enhancement technology can effectively deal with the interference of noise and complex data, the present invention has broad application prospects in the field of local discharge detection and fault diagnosis of power equipment, which helps to improve the safety and stability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0033] Figure 1 A schematic diagram of a method flow of a transformer partial discharge identification method and system based on phase-preserving data enhancement according to an embodiment of the present invention;

[0034] Figure 2 Schematic diagram of four typical PRPD spectra after preprocessing of a transformer partial discharge identification method and system based on phase-preserving data enhancement according to an embodiment of the present invention, wherein Figure 2 (a) is the PRPD spectrum of suspended discharge. Figure 2 (b) is the PRPD spectrum of the tip corona discharge. Figure 2 (c) is the PRPD spectrum of dielectric discharge. Figure 2 (d) is the PRPD spectrum of free particle discharge;

[0035] Figure 3 A schematic diagram of the network structure of a convolutional neural network of a transformer partial discharge identification method and system based on phase-preserving data enhancement according to an embodiment of the present invention;

[0036] Figure 4 An enhanced confusion matrix diagram generated by unused phase-preserving data in a transformer partial discharge identification method and system based on phase-preserving data enhancement according to an embodiment of the present invention;

[0037] Figure 5 An enhanced confusion matrix diagram generated by using phase-preserving data in a transformer partial discharge identification method and system based on phase-preserving data enhancement according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0039] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0040] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0041] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0042] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0043] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0044] Example 1

[0045] Reference Figure 1 to Figure 3 , is an embodiment of the present invention, which provides a transformer partial discharge identification method based on phase-preserving data enhancement, comprising:

[0046] S100: collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum;

[0047] It should be noted that, in the embodiments of the present application, the first spectrum, the second spectrum and the third spectrum are all PRPD spectra, and based on different processing results of the PRPD spectrum, the PRPD spectrum is represented as the first spectrum, the second spectrum and the third spectrum.

[0048] In an optional embodiment, the first preprocessing of the first atlas includes quality assessment and data cleaning.

[0049] S200: performing a second preprocessing on the second atlas, dividing it into a training set and a test set, performing data enhancement processing on the training set, and generating a third atlas;

[0050] In an optional embodiment, performing a second preprocessing on the second atlas includes standardizing the second atlas, and then performing grayscale and normalization processing.

[0051] For example, the types of partial discharges occurring in transformers are independently labeled, and the labels of suspended discharge, corona discharge, insulation discharge, and free discharge are set to 0, 1, 2, and 3, respectively. The images of various real PRPD maps are standardized using the Python-based OpenCV module and scaled to 224×224×3 dimensions. Grayscale and normalization are then performed. The preprocessed high-quality map data is randomly divided into a training set and a test set according to a certain ratio, with a training set: test set = 3:1. The typical PRPD maps after preprocessing are as follows: Figure 2 shown.

[0052] In an embodiment of the present application, data enhancement processing is performed on the training set to generate a third atlas, including: adding Gaussian noise to the second atlas, recording the second atlas as I(x, y), where x and y represent the rows and columns of the image, respectively, generating a Gaussian noise matrix of the same size as the first atlas, recorded as N(x, y), each element of which obeys a normal distribution N(μ, σ 2 ), the calculation formula is expressed as:

[0053]

[0054] Among them, I a (x, y) is the third spectrum after adding Gaussian noise, I b (x, y) is the third atlas data after the pixel value is enhanced within the legal range through cropping operation.

[0055] In the embodiment of the present application, the method further includes: performing phase shift processing on the second spectrum, and translating the phase axis of the spectrum within a specific angle range, which is expressed as:

[0056] I c (x,y)=I(x,(y+Δy)mod360)

[0057] Among them, I c (x, y) is the third spectrum after adding phase drift processing, and Δy is the shift phase angle.

[0058] It should be noted that mod360 ensures that when the value of (y+Δy) exceeds [0°, 360°), it is normalized to the range of 0 to 360.

[0059] In the embodiment of the present application, it also includes: performing segmented perturbation processing on the second spectrum to change the amplitude intensity of the local area of ​​the spectrum, which is expressed as:

[0060]

[0061] Among them, I d(x, y) is the third spectrum after adding segmented perturbation processing, i is the current segment number, n is the total number of segments, α i is the disturbance factor of the i-th segment.

[0062] It should be noted that the disturbance factor is a random number within a preset range, and in the embodiment of the present application, it is set between 0.9 and 1.1.

[0063] Furthermore, the second spectrum is subjected to segmented perturbation processing by dividing the phase axis y into several segments for perturbation, and the phase range of each segment is [y i ,y i+1 ], changing the amplitude intensity of the local area of ​​the spectrum.

[0064] It should be noted that this application abandons traditional data enhancement operations such as rotation, folding and flipping, and uses phase-preserving data enhancement to add Gaussian noise to the PRPD spectrum. The enhancement method can help the model improve its robustness to noisy data; perform phase drift processing on the PRPD spectrum, and translate the phase axis of the spectrum within a specific angle range; perform segmented perturbation processing on the PRPD spectrum to change the amplitude intensity of the local area of ​​the spectrum. This makes the model have stronger generalization ability when dealing with different discharge types, can better adapt to changes in data distribution in practical applications, avoid overfitting problems, and ensure that the model has reliable performance in actual engineering environments.

[0065] S300: constructing a transformer partial discharge identification model, and performing training based on the third graph to obtain an optimized transformer partial discharge identification model, and identifying the partial discharge type based on the optimized transformer partial discharge identification model.

[0066] In the embodiment of the present application, a transformer partial discharge identification model is constructed as follows: Figure 3 Shown include:

[0067] There are 3 convolutional layers and corresponding 3 pooling layers, 3 non-linear activation layers, 2 fully connected layers and a 50% Dropout layer in the fully connected layers. The pooling layer selects the maximum pooling function and the activation function selects the ReLU function.

[0068] In an embodiment of the present application, training is performed based on the third graph to obtain an optimized transformer partial discharge identification model, including: the batch size of the training set and the test set is 32, the learning rate is 0.01, the training cycle is 100, the classification loss function is the cross entropy loss function, and the optimization algorithm is a stochastic gradient descent algorithm. The goal is to minimize the loss function, and the model parameters are iteratively updated after each training cycle to obtain an optimized transformer partial discharge identification model.

[0069] It should be noted that the classification loss function in the embodiment of the present application selects the cross entropy loss function, which calculates the Softmax function internally to evaluate the difference between the model output and the actual label. The goal is to minimize the loss function. After each training cycle, the model parameters are iteratively updated to improve the recognition performance of the model.

[0070] Specifically, the iteration condition is to update the weights of the model by minimizing the cross entropy loss so that the output prediction probability is closer to the true label.

[0071] The formula for the cross entropy loss function is:

[0072]

[0073] Among them, p i is the distribution of true labels, q i is the predicted probability output by the model.

[0074] In the embodiment of the present application, the types of partial discharge identified based on the optimized transformer partial discharge identification model include: suspended discharge, corona discharge, insulation discharge and free discharge.

[0075] Furthermore, the test set is input into the trained transformer partial discharge recognition model for classification, and the confusion matrix is ​​used to evaluate and analyze the recognition effect of various discharge types.

[0076] The above is a schematic scheme of a transformer partial discharge identification method based on phase-preserving data enhancement in this embodiment. It should be noted that the technical scheme of the transformer partial discharge identification system based on phase-preserving data enhancement and the technical scheme of the transformer partial discharge identification method based on phase-preserving data enhancement belong to the same concept, and the details not described in detail in the technical scheme of the transformer partial discharge identification system based on phase-preserving data enhancement in this embodiment can all be referred to the description of the technical scheme of the transformer partial discharge identification method based on phase-preserving data enhancement.

[0077] In this embodiment, a transformer partial discharge identification system based on phase-preserving data enhancement includes:

[0078] A first preprocessing module, used for collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum;

[0079] A second preprocessing module is used to perform a second preprocessing on the second atlas, divide it into a training set and a test set, perform data enhancement processing on the training set, and generate a third atlas;

[0080] The partial discharge identification module is used to construct a transformer partial discharge identification model, and to perform training based on the third graph to obtain an optimized transformer partial discharge identification model, and to identify the partial discharge type based on the optimized transformer partial discharge identification model.

[0081] This embodiment further provides an electronic device, which is applicable to the transformer partial discharge identification method based on phase retention data enhancement, and includes:

[0082] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the transformer partial discharge identification method based on phase retention data enhancement as proposed in the above embodiment.

[0083] This embodiment further provides a storage medium having a computer program stored thereon. When the program is executed by a processor, the method for identifying partial discharge of a transformer based on phase-preserving data enhancement as proposed in the above embodiment is implemented.

[0084] The storage medium proposed in this embodiment and the transformer partial discharge identification method based on phase retention data enhancement proposed in the above embodiment belong to the same inventive concept. The technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0085] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0086] Example 2

[0087] Reference Figure 4-5 , is an embodiment of the present invention. This embodiment is different from the first embodiment in that this embodiment verifies the beneficial effects of the present invention through specific examples.

[0088] In this embodiment, UHF sensors are used to capture and collect pulse information of various partial discharge conditions occurring in the transformer. A total of 4732 pieces of raw data are collected, and the collected raw data information is converted into PRPD spectra. Quality assessment and data cleaning are performed to remove spectra without obvious characteristics. There are 4648 high-quality data remaining, 1217 suspended discharge data, 1128 tip corona discharge data, 1154 insulation discharge data, and 1149 free particle discharge data.

[0089] The types of partial discharge occurring in the transformer are independently labeled, and the labels of suspended discharge, corona discharge, insulation discharge, and free discharge are set to 0, 1, 2, and 3, respectively. The Python-based OpenCV module is used to standardize the images of various real PRPD maps and scale them to 224×224×3 dimensions. The training set and test set are divided into the ratio of 3:1.

[0090] The confusion matrix of the classification results obtained after training is as follows Figure 5 As shown, through Figure 5 The results of partial discharge type recognition show that the highest accuracy is 92.25% without data enhancement. After adding the phase-preserving data enhancement method designed by the present invention, the recognition accuracy is significantly improved to 98.84%, indicating that this operation affects the accuracy while effectively improving the generalization ability and recognition accuracy of the model, and significantly improving the reliability of transformer partial discharge recognition based on PRPD maps.

[0091] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A transformer partial discharge identification method based on phase-preserving data enhancement, characterized in that: include: Collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum; Performing a second preprocessing on the second atlas and dividing it into a training set and a test set, performing data enhancement processing on the training set, and generating a third atlas; A transformer partial discharge identification model is constructed, and training is performed based on the third graph to obtain an optimized transformer partial discharge identification model, and the partial discharge type is identified based on the optimized transformer partial discharge identification model.

2. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 1 is characterized in that: The performing data enhancement processing on the training set to generate a third atlas comprises: Add Gaussian noise to the second spectrum, and record the second spectrum as I(x,y), where x and y represent the rows and columns of the image respectively. Generate a Gaussian noise matrix of the same size as the first spectrum, record it as N(x,y), and each element of it obeys the normal distribution N(μ,σ 2 ), the calculation formula is expressed as: Among them, I a (x, y) is the third spectrum after adding Gaussian noise, I b (x, y) is the third atlas data after the pixel value is enhanced within the legal range through cropping operation.

3. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 2 is characterized in that: The method further includes: performing phase shift processing on the second spectrum, translating the phase axis of the spectrum within a specific angle range, which is expressed as: I c (x,y)=I(x,(y+Δy)mod360) Among them, I c (x, y) is the third spectrum after adding phase drift processing, and Δy is the shift phase angle.

4. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 3 is characterized in that: The method further includes: performing a segmented disturbance process on the second spectrum to change the amplitude intensity of a local area of ​​the spectrum, which is expressed as: Among them, I d (x, y) is the third spectrum after adding segmented perturbation processing, i is the current segment number, n is the total number of segments, α i is the disturbance factor of the i-th segment.

5. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 4 is characterized in that: The construction of transformer partial discharge identification model includes: There are 3 convolutional layers and corresponding 3 pooling layers, 3 non-linear activation layers, 2 fully connected layers and a 50% Dropout layer in the fully connected layers. The pooling layer selects the maximum pooling function and the activation function selects the ReLU function.

6. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 5 is characterized in that: Based on the third graph, the optimized transformer partial discharge identification model is obtained through training, including: the batch size of the training set and the test set is 32, the learning rate is 0.01, the training cycle is 100, the classification loss function is the cross entropy loss function, and the optimization algorithm is the stochastic gradient descent algorithm. The goal is to minimize the loss function. After each training cycle, the model parameters are iteratively updated to obtain the optimized transformer partial discharge identification model.

7. The transformer partial discharge identification method based on phase-preserving data enhancement according to claim 6 is characterized in that: The types of partial discharge identified based on the optimized transformer partial discharge identification model include: suspended discharge, corona discharge, insulation discharge and free discharge.

8. A transformer partial discharge identification system based on phase-preserving data enhancement, characterized in that: include: A first preprocessing module, used for collecting transformer pulse discharge information, converting the pulse discharge information into a first spectrum, and performing a first preprocessing on the first spectrum to obtain a second spectrum; A second preprocessing module, used for performing a second preprocessing on the second atlas, dividing it into a training set and a test set, performing data enhancement processing on the training set, and generating a third atlas; The partial discharge identification module is used to construct a transformer partial discharge identification model, and to perform training based on the third graph to obtain an optimized transformer partial discharge identification model, and to identify the partial discharge type based on the optimized transformer partial discharge identification model.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the transformer partial discharge identification method based on phase retention data enhancement as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the transformer partial discharge identification method based on phase-preserving data enhancement as claimed in any one of claims 1 to 7.

Images