Online identification method and device for winding state of transformer, equipment and medium

Abstract

The embodiment of the invention discloses an online recognition method and device for the winding state of a transformer, equipment and a medium. The method comprises the steps that a current frequency response curve of the transformer is obtained; determining a current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; and obtaining a current winding state of the transformer by using the current normalized frequency response mixed variance and a preset winding state identification model, the winding state identification model being obtained by pre-training a convolutional neural network model based on sample normalized frequency response mixed variances in different winding states. By means of the mode, misjudgment and potential safety risks caused by errors of the monitoring sensor can be avoided, the service life of the transformer is prolonged, and the method has the important significance on safe, reliable and efficient operation of an electric power system.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis of electric power equipment, and in particular to an online identification method and device, equipment and medium for the winding state of a transformer. Background Art

[0002] As the core equipment in the power system, the normal operation of the transformer is crucial to the safety of the power system. The transformer may be hit or impacted during transportation or installation, and may suddenly short-circuit during operation, which may cause winding deformation, thus seriously threatening the safe operation of the power system. Winding deformation is one of the common faults of transformers, which may cause permanent deformation such as local distortion, bulging or displacement, which will seriously affect the safe operation of the transformer and even cause serious power accidents.

[0003] By monitoring the working status of the transformer (winding and core condition) online through monitoring sensors, equipment failures can be discovered in a timely manner, which is of great significance for extending the service life, operational reliability and economy of the transformer, and for ensuring the safe, reliable and efficient operation of the power system. Summary of the invention

[0004] The main purpose of the present invention is to provide an online identification method and device, equipment and medium for the winding state of a transformer, which can solve the problem of monitoring the working state of the transformer in the prior art.

[0005] To achieve the above object, the first aspect of the present invention provides an online identification method for a winding state of a transformer, the method comprising:

[0006] Obtaining a current frequency response curve of the transformer;

[0007] Determine a current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm;

[0008] The current winding state of the transformer is obtained by utilizing the current normalized frequency response mixed variance and a preset winding state recognition model, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

[0009] In a feasible implementation, determining the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm includes:

[0010] Using the reference frequency response curve of the transformer, two sets of current frequency response curves of the transformer and a preset frequency response mixed variance algorithm, the current frequency response mixed variances of different frequency bands are obtained;

[0011] The current normalized frequency response mixed variances of different frequency bands are obtained by using the current frequency response mixed variance and the normalized frequency response mixed variance algorithm.

[0012] In a feasible implementation, the mathematical expression of the frequency response mixed variance algorithm is as follows:

[0013]

[0014] In the formula, S f is the mixed variance of the frequency response under frequency band f, i is the number of the selected frequency point under frequency band f, X is the total number of selected frequency points i; N 1 (i) is the amplitude of the first group of current frequency response curves corresponding to frequency point i, N 2 (i) is the amplitude of the second group of current frequency response curves corresponding to frequency point i, and M(i) is the amplitude of the reference frequency response curve corresponding to frequency point i.

[0015] In a feasible implementation, the mathematical expression of the normalized frequency response mixed variance algorithm is as follows:

[0016]

[0017] In the formula, is the normalized frequency response mixed variance in frequency band f, S f is the mixed variance of the frequency response in frequency band f, i is the frequency point number in frequency band f; N 1 (i) is the amplitude of the first group of current frequency response curves corresponding to frequency point i, N 2 (i) is the amplitude of the frequency point i corresponding to the current frequency response curve of the second group, M(i) is the amplitude of the frequency point i corresponding to the reference frequency response curve, min is the minimum value, and max is the maximum value.

[0018] In a feasible implementation, the method further includes:

[0019] Acquire a sample data set, wherein the sample data set includes a number of sample normalized frequency response mixed variances annotated with true labels of winding states;

[0020] Dividing the sample data set into a training set and a validation set by cross-validation;

[0021] Performing model training for winding state prediction according to the training set and a preset convolutional neural network model to obtain a trained convolutional neural network model;

[0022] The winding state prediction model is optimized according to the verification set and the trained convolutional neural network model to obtain a winding state recognition model.

[0023] In a feasible implementation, the model training for winding state prediction is performed according to the training set and the preset convolutional neural network model to obtain the trained convolutional neural network model, including:

[0024] The convolutional neural network model is trained using the sample normalized frequency response mixed variance in the training set to obtain a predicted label of the winding state, wherein the convolutional neural network model includes a convolutional layer and a fully connected layer connected in sequence, the number of convolutional layers is set to three, the convolution kernel size is selected as 1*1, the step size is 1, the activation function is selected as ReLU, and batch normalization is applied after the convolutional layer or the fully connected layer;

[0025] Determining whether the convolutional neural network model has converged using the predicted label and the true label of the normalized frequency response mixed variance of the samples in the training set and a preset mean square error loss function;

[0026] If the convolutional neural network model converges, a trained convolutional neural network model is obtained.

[0027] In a feasible implementation, the model optimization for fault type prediction is performed according to the verification set and the trained convolutional neural network model to obtain a winding state recognition model, including:

[0028] The model learning rate is adjusted using a linear learning rate decay strategy, an early stopping method is introduced, and the minimum change between the patience value and the verification loss of the verification set is set. After each round of training, the loss of the trained convolutional neural network model on the verification set is calculated, and the model parameters when the verification loss is the lowest are saved to obtain the winding state recognition model.

[0029] To achieve the above object, the second aspect of the present invention provides an online identification device for the winding state of a transformer, the device comprising:

[0030] Data acquisition module: used to obtain the current frequency response curve of the transformer;

[0031] Feature determination module: used to determine the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm;

[0032] State recognition module: used to obtain the current winding state of the transformer by using the current normalized frequency response mixed variance and a preset winding state recognition model, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

[0033] To achieve the above-mentioned purpose, the third aspect of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor executes the steps shown in the first aspect and any feasible implementation method.

[0034] To achieve the above-mentioned purpose, the fourth aspect of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps shown in the first aspect and any feasible implementation method.

[0035] The embodiments of the present invention have the following beneficial effects:

[0036] The present invention provides an online identification method for the winding state of a transformer, the method comprising: obtaining the current frequency response curve of the transformer; determining the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; obtaining the current winding state of the transformer using the current normalized frequency response mixed variance and a preset winding state identification model, the winding state identification model is obtained by pre-training a convolutional neural network model based on the sample normalized frequency response mixed variance under different winding states. Through the above method, it is possible to avoid misjudgment and potential safety risks caused by monitoring sensor errors, extend the service life of the transformer, etc., which is of great significance for ensuring the safe, reliable and efficient operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0038] in:

[0039] Figure 1 It is a flow chart of an online identification method of a transformer winding state in an embodiment of the present invention;

[0040] Figure 2 Another flow chart of an online identification method for a transformer winding state in an embodiment of the present invention;

[0041] Figure 3 It is a structural block diagram of an online identification device for the winding state of a transformer in an embodiment of the present invention;

[0042] Figure 4 4 is a structural block diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 , Figure 1 The flowchart of an online identification method of a transformer winding state in an embodiment of the present invention is shown in FIG. 1 . The method can be applied to both a terminal and a server. The terminal can be a desktop terminal or a mobile terminal. The mobile terminal can be at least one of a mobile phone, a tablet computer, a laptop computer, etc. The server can be implemented as an independent server or a server cluster composed of multiple servers. This embodiment is described by applying to a terminal as an example. Figure 1 The method shown includes the following steps:

[0045] 101. Obtain a current frequency response curve of the transformer;

[0046] 102. Determine a current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm;

[0047] It should be noted that the transformer is subjected to an online frequency response (frequency response) test, and the current frequency response curve is obtained through the online frequency response test. Among them, the normalized frequency response mixed variance is the normalized frequency response mixed variance, and the frequency response mixed variance is the mixed variance between the current frequency response curve and the reference frequency response curve of the transformer. After obtaining the above current frequency response curve, the normalized frequency response mixed variance algorithm can be used to obtain the current normalized frequency response mixed variance.

[0048] Among them, the current frequency response curve can be one group or multiple groups. The present application takes two groups of current frequency response curves as an example. The current normalized frequency response mixed variance can be the mixed variance between the two groups of current frequency response curves and the reference frequency response curve.

[0049] Moreover, the current normalized frequency response mixed variance may include the current normalized frequency response mixed variance of the three frequency bands of low frequency band, middle frequency band and high frequency band according to the frequency band of the current frequency response curve. For example, the current normalized frequency response mixed variance of the low frequency band is calculated by the curve data corresponding to the low frequency band of the two sets of current frequency response curves and the reference frequency response curve, the current normalized frequency response mixed variance of the middle frequency band is calculated by the curve data corresponding to the middle frequency band of the two sets of current frequency response curves and the reference frequency response curve, and the current normalized frequency response mixed variance of the high frequency band is calculated by the curve data corresponding to the high frequency band of the two sets of current frequency response curves and the reference frequency response curve. In this way, the current normalized frequency response mixed variance of the three frequency bands of the current frequency response curve is obtained. The winding state includes but is not limited to the normal state and the abnormal state, and the abnormal state includes but is not limited to various fault types.

[0050] 103. Utilize the current normalized frequency response mixed variance and a preset winding state recognition model to obtain the current winding state of the transformer, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

[0051] Furthermore, after obtaining the above-mentioned current normalized frequency response mixed variance, the current normalized frequency response mixed variance can be input into the pre-trained convolutional neural network model to obtain the current winding state of the transformer. Since the convolutional neural network model is trained based on the sample normalized frequency response mixed variance under different winding states, a large number of corresponding relationships between sample normalized frequency response mixed variances and winding states have been pre-learned, so the corresponding winding state can be output according to the current normalized frequency response mixed variance.

[0052] Among them, the current normalized frequency response mixed variance may include the current normalized frequency response mixed variance of three frequency bands (high, medium and low), thereby improving the accuracy of online identification of the winding state, and avoiding misjudgment and potential safety risks caused by monitoring sensor errors.

[0053] The present invention provides an online identification method for the winding state of a transformer, the method comprising: obtaining the current frequency response curve of the transformer; determining the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; obtaining the current winding state of the transformer using the current normalized frequency response mixed variance and a preset winding state identification model, the winding state identification model being obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states. Through the above-mentioned method, misjudgment and potential safety risks caused by monitoring sensor errors can be avoided, and the service life of the transformer can be extended, which is of great significance for ensuring the safe, reliable and efficient operation of the power system.

[0054] See also Figure 2 , Figure 2 FIG. 1 is another flow chart of an online identification method for a transformer winding state in an embodiment of the present invention, as shown in FIG. Figure 2 The method shown includes the following steps:

[0055] 201. Obtain a current frequency response curve of the transformer;

[0056] It should be noted that step 201 and Figure 1 The contents of step 101 are similar and will not be described here to avoid repetition. Figure 1 The content of step 101 is shown.

[0057] 202. Obtain current frequency response mixed variances of different frequency bands using the reference frequency response curve of the transformer, two sets of current frequency response curves of the transformer, and a preset frequency response mixed variance algorithm;

[0058] It can be understood that the mixed frequency response variance in the present application can be the mixed frequency response variance between the reference frequency response curve and the current frequency response curve, and can be the mixed frequency response variance between the reference frequency response curve and two sets of current frequency response curves. And according to different frequency bands of the curve, the mixed frequency response variance of different frequency bands can also be obtained.

[0059] Furthermore, the mathematical expression of the frequency response mixed variance algorithm is as follows:

[0060]

[0061] In the formula, S f is the mixed variance of the frequency response under frequency band f, i is the number of the selected frequency point under frequency band f, X is the total number of selected frequency points i; N 1 (i) is the amplitude of the first group of current frequency response curves corresponding to frequency point i, N 2 (i) is the amplitude of the second group of current frequency response curves corresponding to frequency point i, and M(i) is the amplitude of the reference frequency response curve corresponding to frequency point i.

[0062] For example, the reference frequency response curve of the transformer and two sets of current frequency response curves are selected at 100 frequency points in the low frequency band (1kHz-100kHz), the middle frequency band (100kHz-600kHz), and the high frequency band (600kHz-1MHz), that is, X is 100, and the mixed variance of the current frequency response of each fault type is calculated, then formula (1) is transformed into:

[0063]

[0064] 203. Obtain current normalized frequency response mixed variances of different frequency bands by using the current frequency response mixed variance and the normalized frequency response mixed variance algorithm;

[0065] Furthermore, after obtaining the current frequency response mixed variance, normalization processing is performed to obtain the normalized frequency response mixed variance.

[0066] Exemplarily, the mathematical expression of the normalized frequency response mixed variance algorithm is as follows:

[0067]

[0068] In the formula, is the normalized frequency response mixed variance in frequency band f, S f is the mixed variance of the frequency response in frequency band f, i is the frequency point number in frequency band f; N 1 (i) is the amplitude of the first group of current frequency response curves corresponding to frequency point i, N 2 (i) is the amplitude of the second group of current frequency response curves corresponding to frequency point i, M(i) is the amplitude of the reference frequency response curve corresponding to frequency point i, min is the minimum value, and max is the maximum value. It can be understood that according to different frequency bands, the current normalized frequency response mixed variance under different frequency bands can be obtained.

[0069] 204. Utilize the current normalized frequency response mixed variance and a preset winding state recognition model to obtain the current winding state of the transformer, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

[0070] It should be noted that step 204 and Figure 1 The contents of step 103 are similar, and are not described here to avoid repetition. For details, please refer to Figure 1 The content of step 103 is shown.

[0071] In a feasible implementation, it is necessary to pre-train the convolutional neural network model so that the convolutional neural network model can determine the winding state based on the normalized frequency response mixed variance. Therefore, the above method further includes steps A01 to A04:

[0072] A01. Obtain a sample data set, wherein the sample data set includes a number of sample normalized frequency response mixed variances annotated with real labels of winding states;

[0073] It is understandable that it is necessary to collect a number of normalized frequency response mixed variances of different winding states as training samples to obtain a sample data set required for model training, wherein the sample data set includes a number of sample normalized frequency response mixed variances annotated with the true labels of the winding states.

[0074] For example, the transformer working reference frequency response curve and two groups of fault frequency response curves under a certain degree and type are selected at 100 frequency points in the low frequency band (1kHz-100kHz), the medium frequency band (100kHz-600kHz), and the high frequency band (600kHz-1MHz), and the normalized frequency response mixed variance of each fault type is calculated. The data under the three frequency bands is a sample, and 100 samples under each fault type are used as a data set. Then, the model can be trained using the sample data set to obtain a target model that can predict the winding state.

[0075] A02, dividing the sample data set into a training set and a validation set by cross-validation;

[0076] Furthermore, in order to improve the accuracy of the model, the present application adopts a cross-validation method to train the model, that is, the data of the training set is used for preliminary training, and then the preliminary training model is verified using the verification set. Specifically, the sample data set is divided into a training set and a verification set using cross-validation, and the division ratio is not limited here.

[0077] A03. Performing model training for winding state prediction according to the training set and a preset convolutional neural network model to obtain a trained convolutional neural network model;

[0078] After the training set is divided, the convolutional neural network model is initially trained using the training set, and the winding state prediction model is trained based on the training set and the preset convolutional neural network model to obtain the trained convolutional neural network model. Specifically, step A03 includes:

[0079] B01. The convolutional neural network model is trained using the sample normalized frequency response mixed variance in the training set to obtain a predicted label of the winding state, wherein the convolutional neural network model includes a convolutional layer and a fully connected layer connected in sequence, the number of convolutional layers is set to three, the convolution kernel size is selected as 1*1, the step size is 1, the activation function is selected as ReLU, and batch normalization is applied after the convolutional layer or the fully connected layer;

[0080] B02, using the predicted label and the real label of the sample normalized frequency response mixed variance in the training set and a preset mean square error loss function to determine whether the convolutional neural network model converges;

[0081] B03. If the convolutional neural network model converges, a trained convolutional neural network model is obtained.

[0082] The predicted label is used to indicate the predicted winding state. The loss value of the convolutional neural network model at this time is obtained by the predicted label and the real label to determine whether the convolutional neural network model converges. If it converges, it means that the difference between the predicted label and the real label is very small, and the prediction of the real label can almost be completed. Then the training can be stopped and step B03 is executed; if it does not converge, it means that the difference between the predicted label and the real label is very large, the prediction accuracy is very low, or the prediction requirements are not met, then the model parameters of the current convolutional neural network model are adjusted using the loss value of the predicted label and the real label, and the step B01 is returned to execute until the convolutional neural network model converges to obtain the trained convolutional neural network model. The calculation of the loss value includes the mean square error loss function.

[0083] A04. Optimize the model for winding state prediction based on the verification set and the trained convolutional neural network model to obtain a winding state recognition model.

[0084] Furthermore, the trained convolutional neural network model is verified by using the validation set, and the winding state prediction ability of the trained convolutional neural network model is evaluated by using samples that did not participate in the preliminary training. That is, the model of winding state prediction is optimized according to the validation set and the trained convolutional neural network model to obtain the winding state recognition model.

[0085] In a feasible implementation, step A04 includes: adjusting the model learning rate using a linear learning rate decay strategy, introducing an early stopping method, setting the minimum change between the patience value and the verification loss of the verification set, and after each round of training, calculating the loss of the trained convolutional neural network model on the verification set, saving the model parameters when the verification loss is the lowest, and obtaining the winding state recognition model.

[0086] It should be noted that after the initial training, the model will be optimized. The linear learning rate decay strategy is used to adjust the model learning rate, the early stopping method is introduced, the patience value and the minimum change in the validation loss are set, and after each round of training, the model loss on the validation set is calculated, and the model parameters when the validation loss is the lowest are saved.

[0087] Exemplarily, this application uses a linear learning rate decay strategy to adjust the model learning rate, introduces an early stopping method, sets the patience value and the minimum change in validation loss, calculates the model loss on the validation set after each round of training, and saves the model parameters when the validation loss is the lowest. This can be done in the following steps:

[0088] 1) The first step, initialization parameters:

[0089] Set the initial learning rate `initial_lr`.

[0090] Set the learning rate decay rate `lr_decay`.

[0091] Set the patience value of the early stopping method, that is, the number of consecutive rounds (epochs) allowed when the loss on the validation set does not improve.

[0092] Set the minimum change in validation loss, `min_delta`, which is the loss change threshold for a significant improvement in model performance.

[0093] 2) The second stage: setting up the training environment:

[0094] Select an optimizer such as SGD, Adam, etc., and set its initial learning rate.

[0095] Prepare training data and validation data.

[0096] 3) The third stage, training model:

[0097] For each epoch:

[0098] Calculate the learning rate of the current round, usually using the linear decay formula: `current_lr = initial_lr * (1-epoch / total_epochs)`.

[0099] Update the optimizer's learning rate.

[0100] Train the model for one epoch.

[0101] Evaluate the model performance on the validation set and calculate the validation loss.

[0102] 4) The fourth step: Apply the early stopping method:

[0103] Maintain a variable to record the validation loss of the best model `best_val_loss` and initialize it to a large number.

[0104] Maintain a counter `counter` to record the number of consecutive rounds in which the validation loss has not improved.

[0105] After each epoch:

[0106] If the current validation loss is lower than `best_val_loss` and the improvement is greater than `min_delta`, update `best_val_loss`, save the current model parameters, and reset `counter`.

[0107] If the current validation loss has not improved, or the improvement is less than min_delta , then counter is incremented.

[0109] If `counter` reaches `patience`, stop training.

[0110] 5) The fifth step: saving the best model:

[0111] During training, whenever there is a significant improvement in validation loss, save the model parameters.

[0112] After the training is completed, the saved optimal model parameters are used to restore the model state to obtain the winding state recognition model.

[0113] Furthermore, after the winding state recognition model is obtained, the winding state recognition model can be used to monitor the winding state online, that is, to execute the process of steps 201 to 204 above. Specifically, the trained winding state recognition model is integrated into the hardware device or deployed on the cloud server, and the data from the sensor is received and processed in real time. Key features are extracted from the collected data, and the normalized frequency response mixed variance of different frequency bands of the current state is calculated using formula (1) and formula (3);

[0114] Using the extracted features as input, the trained winding state recognition model is used for inference to obtain the transformer winding state assessment result. Based on the output of the model, the transformer operating state is adjusted in real time.

[0115] In addition, new monitoring data will be used regularly to retrain and optimize the model through online learning or incremental learning, so that the model can adapt to changes in the transformer status.

[0116] The present invention provides an online identification method for the winding state of a transformer, the method comprising: obtaining the current frequency response curve of the transformer; using the reference frequency response curve of the transformer, two sets of current frequency response curves of the transformer and a preset frequency response mixed variance algorithm to obtain the current frequency response mixed variance of different frequency bands; using the current frequency response mixed variance and the normalized frequency response mixed variance algorithm to obtain the current normalized frequency response mixed variance of different frequency bands; using the current normalized frequency response mixed variance and a preset winding state identification model to obtain the current winding state of the transformer, the winding state identification model is obtained by pre-training a convolutional neural network model based on the sample normalized frequency response mixed variance under different winding states. In the above manner, the normalized frequency response mixed variance is calculated according to the frequency response curve, the model is trained using a convolutional neural network, and then the model is optimized, and finally the model is used to monitor the winding state online, and retrain and re-optimize regularly. The patent of the present invention has the advantages of avoiding misjudgment and potential safety risks caused by sensor errors, extending the service life of the transformer, etc., and is of great significance for ensuring the safe, reliable and efficient operation of the power system.

[0117] See also Figure 3 , Figure 3 FIG. 1 is a structural block diagram of an online identification device for a transformer winding state in an embodiment of the present invention, such as Figure 3 The device shown comprises:

[0118] Data acquisition module 301: used to obtain the current frequency response curve of the transformer;

[0119] Feature determination module 302: used to determine the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm;

[0120] State recognition module 303: used to obtain the current winding state of the transformer using the current normalized frequency response mixed variance and a preset winding state recognition model, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

[0121] It should be noted that Figure 3 The functions of each module in the device shown are Figure 1 The contents of each step in the method shown are similar, so they are not described here to avoid repetition. For details, please refer to Figure 1 The content of each step in the method shown.

[0122] The present invention provides an online identification device for the winding state of a transformer, the device comprising: a data acquisition module: used to obtain the current frequency response curve of the transformer; a feature determination module: used to determine the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; a state identification module: used to obtain the current winding state of the transformer using the current normalized frequency response mixed variance and a preset winding state identification model, the winding state identification model is obtained by pre-training a convolutional neural network model based on the sample normalized frequency response mixed variance under different winding states. Through the above-mentioned method, it is possible to avoid misjudgment and potential safety risks caused by monitoring sensor errors, extend the service life of the transformer, etc., which is of great significance for ensuring the safe, reliable and efficient operation of the power system.

[0123] Figure 4 FIG. 1 shows an internal structure diagram of a computer device in an embodiment. The computer device may be a terminal or a server. Figure 4As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement the above method. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can implement the above method. Those skilled in the art can understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0124] In one embodiment, a computer device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: Figure 1 or Figure 2 Steps shown.

[0125] In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processor performs the following steps: Figure 1 or Figure 2 Steps shown.

[0126] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0127] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0128] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for online identification of the winding state of a transformer, characterized in that: The method comprises: Obtaining a current frequency response curve of the transformer; Determine a current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; The current winding state of the transformer is obtained by utilizing the current normalized frequency response mixed variance and a preset winding state recognition model, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

2. The method according to claim 1, characterized in that: Determining the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm includes: Using the reference frequency response curve of the transformer, two sets of current frequency response curves of the transformer and a preset frequency response mixed variance algorithm, current frequency response mixed variances of different frequency bands are obtained; The current normalized frequency response mixed variances of different frequency bands are obtained by using the current frequency response mixed variance and the normalized frequency response mixed variance algorithm.

3. The method according to claim 2, characterized in that: The mathematical expression of the frequency response mixed variance algorithm is as follows: In the formula, S f is the mixed variance of the frequency response under the frequency band f, i is the number of the selected frequency point under the frequency band f, X is the total number of the selected frequency point i; N1(i) is the amplitude of the first group of current frequency response curves corresponding to the frequency point i, N2(i) is the amplitude of the second group of current frequency response curves corresponding to the frequency point i, and M(i) is the amplitude of the reference frequency response curve corresponding to the frequency point i.

4. The method according to claim 3, characterized in that: The mathematical expression of the normalized frequency response mixed variance algorithm is as follows: In the formula, is the normalized frequency response mixed variance in frequency band f, S f is the mixed variance of the frequency response under the frequency band f, i is the frequency point number under the frequency band f; N1(i) is the amplitude of the first group of current frequency response curves corresponding to the frequency point i, N2(i) is the amplitude of the second group of current frequency response curves corresponding to the frequency point i, M(i) is the amplitude of the reference frequency response curve corresponding to the frequency point i, min is the minimum value, and max is the maximum value.

5. The method according to claim 1, characterized in that: The method further comprises: Acquire a sample data set, wherein the sample data set includes a number of sample normalized frequency response mixed variances annotated with true labels of winding states; Dividing the sample data set into a training set and a validation set by cross-validation; Performing model training for winding state prediction according to the training set and a preset convolutional neural network model to obtain a trained convolutional neural network model; The winding state prediction model is optimized according to the verification set and the trained convolutional neural network model to obtain a winding state recognition model.

6. The method according to claim 5, characterized in that: The model training for winding state prediction is performed according to the training set and the preset convolutional neural network model to obtain the trained convolutional neural network model, including: The convolutional neural network model is trained using the sample normalized frequency response mixed variance in the training set to obtain a predicted label of the winding state, wherein the convolutional neural network model includes a convolutional layer and a fully connected layer connected in sequence, the number of convolutional layers is set to three, the convolution kernel size is selected as 1*1, the step size is 1, the activation function is selected as ReLU, and batch normalization is applied after the convolutional layer or the fully connected layer; Determining whether the convolutional neural network model has converged using the predicted labels and the true labels of the normalized frequency response mixed variance of the samples in the training set and a preset mean square error loss function; If the convolutional neural network model converges, a trained convolutional neural network model is obtained.

7. The method according to claim 6, characterized in that: The model optimization for fault type prediction is performed according to the verification set and the trained convolutional neural network model to obtain a winding state recognition model, including: The model learning rate is adjusted using a linear learning rate decay strategy, an early stopping method is introduced, and the minimum change between the patience value and the verification loss of the verification set is set. After each round of training, the loss of the trained convolutional neural network model on the verification set is calculated, and the model parameters when the verification loss is the lowest are saved to obtain the winding state recognition model.

8. An online identification device for the winding state of a transformer, characterized in that: The device comprises: Data acquisition module: used to obtain the current frequency response curve of the transformer; Feature determination module: used to determine the current normalized frequency response mixed variance according to the current frequency response curve and a preset normalized frequency response mixed variance algorithm; State recognition module: used to obtain the current winding state of the transformer by using the current normalized frequency response mixed variance and a preset winding state recognition model, wherein the winding state recognition model is obtained by pre-training a convolutional neural network model based on the normalized frequency response mixed variance of samples under different winding states.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images