Method and device for constructing partial discharge detection model

Abstract

The invention provides a method and device for constructing a partial discharge detection model, and belongs to the field of partial discharge detection.The method comprises the steps that a partial discharge detection data set is obtained, and the partial discharge detection data set comprises a plurality of partial discharge detection data; grouping the partial discharge data set to obtain a first partial discharge detection data set, a second partial discharge detection data set and a third partial discharge detection data set; and synchronously or asynchronously training the partial discharge detection model through the first partial discharge detection data set, the second partial discharge detection data set and the third partial discharge detection data set. According to the method, the partial discharge detection model is trained by grouping the partial discharge detection data, so that the efficiency of constructing the partial discharge detection model is improved.

Description

Technical Field

[0001] This application relates to the field of partial discharge detection, and in particular to a method and apparatus for constructing a partial discharge detection model. Background Art

[0002] With the development of AI technology, applying AI for partial discharge detection is a future trend. Utilizing AI technology for partial discharge detection requires the construction of a partial discharge detection model.

[0003] Supervised learning is a widely used learning method in machine learning. It trains a model using known inputs and corresponding outputs, enabling the model to predict the correct output based on new inputs. How to construct a partial discharge detection model using existing partial discharge detection data is a problem that needs to be solved. Summary of the Invention

[0004] In view of this, this application proposes a method, apparatus, electronic device and storage medium for constructing a partial discharge detection model.

[0005] To achieve the above objectives, this application provides a method for constructing a partial discharge detection model, comprising:

[0006] A partial discharge detection dataset is obtained, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals.

[0007] The partial discharge dataset is grouped to obtain a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein each of the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group includes at least one partial discharge detection data.

[0008] The partial discharge detection model is trained synchronously or asynchronously using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, so that the output result of the partial discharge detection model meets the threshold requirements.

[0009] Based on the same concept, this application also provides an apparatus for constructing a partial discharge detection model, comprising:

[0010] The acquisition module is used to acquire a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals.

[0011] The grouping module is used to group the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group each include at least one partial discharge detection data.

[0012] The training module is used to synchronously or asynchronously train the partial discharge detection model using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group, so that the output result of the partial discharge detection model meets the threshold requirements.

[0013] Based on the same concept, this application also provides an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the method of constructing a partial discharge detection model as described in any of the preceding claims.

[0014] Based on the same concept, this application also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the method of constructing a partial discharge detection model as described in any of the preceding claims. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments will be briefly described below. Obviously, the drawings described below only relate to some embodiments of the present invention and are not intended to limit the present invention.

[0016] Figure 1 This is a flowchart illustrating the method for constructing a partial discharge detection model provided in an embodiment of this application;

[0017] Figure 2 This is a schematic diagram of the system structure of the method for constructing a partial discharge detection model provided in an embodiment of this application;

[0018] Figure 3A This is a schematic diagram of partial discharge data in the initial stage of partial discharge provided in an embodiment of this application;

[0019] Figure 3B This is a schematic diagram of partial discharge data during the mid-stage of partial discharge provided in an embodiment of this application;

[0020] Figure 3C This is a schematic diagram of partial discharge data in the later stage of partial discharge provided in an embodiment of this application;

[0021] Figure 4This is another flowchart illustrating the method for constructing a partial discharge detection model provided in an embodiment of this application;

[0022] Figure 5 This is another flowchart illustrating the method for constructing a partial discharge detection model provided in the embodiments of this application;

[0023] Figure 6 This is another schematic flowchart illustrating the method for constructing a partial discharge detection model provided in the embodiments of this application;

[0024] Figure 7 This is a schematic diagram of an apparatus for constructing a partial discharge detection model according to an embodiment of this application;

[0025] Figure 8 This is a schematic diagram of an electronic device structure provided according to an embodiment of this application;

[0026] Figure 9 This is a schematic diagram of a storage medium provided according to an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the described embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains. The terms “first,” “second,” and similar terms used in the specification and claims of this patent application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as “comprising” or “including” mean that the technical feature preceding the term encompasses the technical features listed following the term and their equivalents, without excluding other technical features.

[0029] At least one embodiment of this application provides a method for constructing a partial discharge detection model. The method includes: acquiring a partial discharge detection dataset, the dataset comprising multiple partial discharge detection data points, each including a denoised partial discharge signal; grouping the dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group; and synchronously or asynchronously training the partial discharge detection model using the first, second, and third partial discharge detection data groups, such that the output of the model meets a threshold requirement. This method improves the efficiency of constructing a partial discharge detection model by grouping the partial discharge detection data for training.

[0030] At least one embodiment of this application also provides an apparatus for constructing a partial discharge detection model, comprising: an acquisition module for acquiring a partial discharge detection dataset, the partial discharge detection dataset including multiple partial discharge detection data, the partial discharge detection data including denoised partial discharge signals; a grouping module for grouping the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group each including at least one partial discharge detection data; and a training module for synchronously or asynchronously training the partial discharge detection model using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group, so that the output result of the partial discharge detection model meets a threshold requirement. This method improves the efficiency of constructing a partial discharge detection model by grouping and training the partial discharge detection data.

[0031] At least one embodiment of this disclosure also provides an electronic device, which includes: one or more processors and a storage device; the storage device is used to store one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the methods for constructing a partial discharge detection model provided in the embodiments of this disclosure.

[0032] At least one embodiment of this disclosure also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform any of the methods for constructing a partial discharge detection model provided in the embodiments of this disclosure.

[0033] Figure 1 This is a schematic flowchart illustrating the method for constructing a partial discharge detection model provided in an embodiment of this application. Figure 1As shown, the method includes:

[0034] Step 102: Obtain a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals;

[0035] According to embodiments of this disclosure, the partial discharge detection dataset can be an open-source dataset or a self-built dataset. The partial discharge data in the self-built dataset includes partial discharge signals and partial discharge results. The partial discharge signals are denoised and are suitable for feature extraction. The partial discharge results are verified partial discharge results and can be obtained through manual or automatic standards.

[0036] According to embodiments of this disclosure, the partial discharge signal includes one or more of the following: time-domain features (pulse amplitude, rise time, number of discharges / phase), frequency-domain features (spectrum, wavelet energy coefficient), and statistical features (skewness, kurtosis, pulse repetition rate).

[0037] Step 104: The partial discharge dataset is grouped to obtain a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group each include at least one partial discharge detection data.

[0038] Step 106: Train the partial discharge detection model synchronously or asynchronously using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group, so that the output result of the partial discharge detection model meets the threshold requirements.

[0039] According to embodiments of this disclosure, the output of the partial discharge detection model includes one or more of accuracy, precision, recall, and F1 score. Accuracy is the proportion of samples correctly predicted by the model out of the total number of samples; precision is the proportion of samples predicted as positive by the model that are actually positive; recall is the proportion of samples that are actually positive that are correctly predicted as positive; and the F1 score is the harmonic mean of precision and recall.

[0040] According to an embodiment of this disclosure, the degree of partial discharge is determined based on the partial discharge dataset, and the partial discharge dataset is grouped into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group. The first partial discharge detection data group corresponds to the late stage of partial discharge, the second partial discharge detection data group corresponds to the middle stage of partial discharge, and the third partial discharge detection data group corresponds to the early stage of partial discharge.

[0041] According to embodiments of this disclosure, the partial discharge type is determined based on the partial discharge dataset, and the partial discharge dataset is grouped into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group. The first partial discharge detection data group corresponds to the pulsed discharge, the second partial discharge detection data group corresponds to the continuous discharge, and the third partial discharge detection data group corresponds to the random discharge.

[0042] According to embodiments of this disclosure, the partial discharge dataset is grouped into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group based on the partial discharge dataset to determine the degree and type of partial discharge. For example, the first partial discharge detection data group corresponds to the late stage of pulsed discharge, the second partial discharge detection data group corresponds to the middle stage of continuous discharge, and the third partial discharge detection data group corresponds to the early stage of random discharge.

[0043] According to embodiments of this disclosure, when grouping the partial discharge dataset according to the degree and type of partial discharge, the influencing factors of the degree and type of partial discharge on the partial discharge detection model are determined. When the influence of the degree of partial discharge is greater than that of the type of partial discharge, the partial discharge dataset is preferentially grouped according to the degree of partial discharge.

[0044] According to the embodiments of this disclosure, by determining the influencing factors of partial discharge degree and partial discharge type on the partial discharge detection model, the influence of different partial discharge degree inputs on the output results of the partial discharge detection model can be obtained; at the same time, the influence of different partial discharge type inputs on the output results of the partial discharge detection model can be obtained.

[0045] According to embodiments of this disclosure, a partial discharge detection dataset is obtained, the partial discharge detection dataset including multiple partial discharge detection data, the partial discharge detection data including denoised partial discharge signals; the partial discharge dataset is grouped into a first partial discharge detection data group, a second partial discharge detection data group, a third partial discharge detection data group, and a fourth partial discharge detection data group, each of the first partial discharge detection data group, the second partial discharge detection data group, the third partial discharge detection data group, and the fourth partial discharge detection data group including at least one partial discharge detection data; the partial discharge detection model is trained synchronously or asynchronously using the first partial discharge detection data group, the second partial discharge detection data group, the third partial discharge detection data group, and the fourth partial discharge detection data group, so that the output result of the partial discharge detection model meets the threshold requirements.

[0046] According to embodiments of this disclosure, grouping the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, a third partial discharge detection data group, and a fourth partial discharge detection data group includes: grouping the partial discharge dataset into these groups based on the degree or type of partial discharge. The first partial discharge detection data group corresponds to the early stage of partial discharge, the second partial discharge detection data group corresponds to the middle stage of partial discharge, the third partial discharge detection data group corresponds to the middle and late stages of partial discharge, and the fourth partial discharge detection data group corresponds to the late stage of partial discharge. By dividing the partial discharge dataset into four groups, the classification of partial discharge results corresponding to the partial discharge dataset becomes more refined, facilitating more precise adjustment of the partial discharge detection model and improving the accuracy of the partial discharge detection model.

[0047] Figure 2 This is a schematic diagram of the system structure of the method for constructing a partial discharge detection model provided in an embodiment of this application. The partial discharge detection platform includes a partial discharge detection module library, a partial discharge sample library, and a partial discharge detection training platform. The partial discharge sample library includes data preprocessing, sample labeling, and sample management.

[0048] The sample management process groups the partial discharge dataset into three sets: the first partial discharge detection dataset, the second partial discharge detection dataset, and the third partial discharge detection dataset. These datasets are then used for model training and validation on the partial discharge detection training platform.

[0049] The partial discharge detection training platform includes environment creation, model development, model training, and model evaluation. The platform is built upon a machine learning framework.

[0050] Partial discharge detection models include at least one model, and different partial discharge detection models differ in the number of model parameters, convergence speed, and accuracy. Partial discharge detection models include free metal particle discharge models, corona discharge models, insulation defect models, and floating potential discharge models.

[0051] Figure 3A This is a schematic diagram of partial discharge data in the initial stage of partial discharge provided in an embodiment of this application; as shown... Figure 3A As shown, the number of partial discharges is relatively small, and the amplitude is below 5pC, indicating that it is in the early stage of partial discharge. Figure 3B This is a schematic diagram of partial discharge data in the middle stage of partial discharge provided in the embodiments of this application; the number of partial discharges increases, and the amplitude is about 5pC, which is in the middle stage of partial discharge. Figure 3CThis is a schematic diagram of partial discharge data in the later stage of partial discharge provided in an embodiment of this application; the number of partial discharges is continuous and the amplitude is greater than 5 pC.

[0052] Figure 4 This is another flowchart illustrating the method for constructing a partial discharge detection model provided in an embodiment of this application. For example... Figure 4 As shown, the method includes:

[0053] Step 402: Obtain a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals;

[0054] According to embodiments of this disclosure, the partial discharge detection dataset can be an open-source dataset or a self-built dataset. The partial discharge data in the self-built dataset includes partial discharge signals and partial discharge results. The partial discharge signals are denoised and are suitable for feature extraction. The partial discharge results are verified partial discharge results and can be obtained through manual or automatic standards.

[0055] According to embodiments of this disclosure, the partial discharge signal includes one or more of the following: time-domain features (pulse amplitude, rise time, number of discharges / phase), frequency-domain features (spectrum, wavelet energy coefficient), and statistical features (skewness, kurtosis, pulse repetition rate).

[0056] Step 404: The partial discharge dataset is grouped to obtain a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein each of the first partial discharge detection data group, the second partial discharge detection data group, or the third partial discharge detection data group includes at least one partial discharge detection data.

[0057] Step 406: Simultaneously train the partial discharge detection model using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, so that the output result of the partial discharge detection model meets the threshold requirement.

[0058] According to embodiments of this disclosure, the output of the partial discharge detection model includes one or more of accuracy, precision, recall, and F1 score. Accuracy is the proportion of samples correctly predicted by the model out of the total number of samples; precision is the proportion of samples predicted as positive by the model that are actually positive; recall is the proportion of samples that are actually positive that are correctly predicted as positive; and the F1 score is the harmonic mean of precision and recall.

[0059] According to an embodiment of this disclosure, the step of synchronously training the partial discharge detection model using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group includes: synchronously training the partial discharge detection model using partial partial discharge detection data from the first partial discharge detection data group, partial partial discharge detection data from the second partial discharge detection data group, and partial partial discharge detection data from the third partial discharge detection data group.

[0060] According to embodiments of this disclosure, by simultaneously training the partial discharge detection model with the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, the model can be made compatible with various partial discharge scenarios, adapt to different partial discharge detection data, and improve the robustness of the constructed partial discharge detection model.

[0061] Figure 5 This is another flowchart illustrating the method for constructing a partial discharge detection model provided in an embodiment of this application. Figure 4 As shown, the method includes:

[0062] Step 502: Obtain a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals;

[0063] Step 504: The partial discharge dataset is grouped to obtain a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein each of the first partial discharge detection data group, the second partial discharge detection data group, or the third partial discharge detection data group includes at least one partial discharge detection data.

[0064] Step 506: Train the partial discharge detection model asynchronously using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, so that the output of the partial discharge detection model meets the threshold requirements.

[0065] According to embodiments of this disclosure, the partial discharge detection model is trained stepwise using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set. A first partial discharge detection model is obtained using the first partial discharge detection data set, and the output accuracy of the first partial discharge detection model satisfies a first threshold. A second partial discharge detection model is obtained using the second partial discharge detection data set, and the output accuracy of the second partial discharge detection model satisfies a second threshold. A third partial discharge detection model is obtained using the third partial discharge detection data set, and the output accuracy of the third partial discharge detection model satisfies a third threshold.

[0066] According to an embodiment of this disclosure, obtaining a second partial discharge detection model using the second partial discharge detection data set includes: training the first partial discharge detection model using the second partial discharge detection data set to obtain the second partial discharge detection model, wherein the second threshold is less than the first threshold.

[0067] According to an embodiment of this disclosure, obtaining a third partial discharge detection model using the third partial discharge detection data set includes: training a second partial discharge detection model using the third partial discharge detection data set to obtain the third partial discharge detection model, wherein the third threshold is less than the second threshold.

[0068] According to embodiments of this disclosure, by training the partial discharge detection model step by step with the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, the module can start from a dataset with obvious features, thereby accelerating the model convergence speed and improving the speed of constructing the partial discharge detection model.

[0069] Figure 6 This is another schematic flowchart illustrating the method for constructing a partial discharge detection model provided in the embodiments of this application. Figure 4 As shown, the method includes:

[0070] Step 602: Obtain a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals;

[0071] Step 604: The partial discharge results are classified into pulsed discharge, continuous discharge and random discharge according to the discharge waveform; the first partial discharge detection data group corresponds to the pulsed discharge, the second partial discharge detection data group corresponds to the continuous discharge, and the third partial discharge detection data group corresponds to the random discharge.

[0072] Step 606: Train the partial discharge detection model synchronously or asynchronously using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set.

[0073] Figure 7 An apparatus 700 for constructing a partial discharge detection model is shown, comprising:

[0074] The acquisition module 710 is used to acquire a partial discharge detection dataset, which includes multiple partial discharge detection data and the partial discharge detection data includes noise-reduced partial discharge signals.

[0075] The grouping module 720 is used to group the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein each of the first partial discharge detection data group, the second partial discharge detection data group, or the third partial discharge detection data group includes at least one partial discharge detection data.

[0076] The training module 730 is used to train the partial discharge detection model synchronously or asynchronously using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group, so that the output result of the partial discharge detection model meets the threshold requirements.

[0077] The device system provided in this disclosure can achieve the same working process and technical effect as the embodiment of the ultra-fast transient overvoltage automatic detection method, which will not be repeated here. Please refer to the previous description.

[0078] At least one embodiment of this disclosure also provides an electronic device 800, see reference. Figure 8 The electronic device 800 includes one or more processors 810 and a storage device 820; the storage device 820 is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement any of the methods for constructing a partial discharge detection model provided in the embodiments of this disclosure.

[0079] At least one embodiment of this application also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform any of the methods for constructing a partial discharge detection model provided in the embodiments of this disclosure.

[0080] The storage medium for the computer-executable instructions provided in at least one embodiment of this disclosure may be a non-transitory computer-readable storage medium. Exemplarily, Figure 9 This is a schematic diagram of a non-transitory computer-readable storage medium provided for at least one embodiment of the present disclosure.

[0081] For example, such as Figure 9 As shown, computer-readable instructions 901 are non-transitory stored on computer-readable storage medium 900. For example, when computer-readable instructions 901 are executed by a processor, one or more steps in the ultrafast transient overvoltage automatic detection method described above can be performed.

[0082] For example, the storage medium 900 can be used in the aforementioned electronic device 800. For example, the storage medium 900 may include the memory 820 in the electronic device 800.

[0083] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0084] The following points need to be explained:

[0085] (1) The accompanying drawings of the embodiments of this disclosure only involve the structures involved in the embodiments of this disclosure. Other structures can be referred to the general design.

[0086] (2) Where there is no conflict, the embodiments of this disclosure and the features in the embodiments can be combined with each other to obtain new embodiments.

[0087] The above description is merely an exemplary embodiment of the present invention and is not intended to limit the scope of protection of the present invention, which is determined by the appended claims.

Claims

1. A method for constructing a partial discharge detection model, characterized in that, include: A partial discharge detection dataset is obtained, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals. The partial discharge dataset is grouped to obtain a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group. Each of the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group includes multiple partial discharge detection data. The partial discharge detection model is trained synchronously or asynchronously using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set, so that the output result of the partial discharge detection model meets the threshold requirements.

2. The method according to claim 1, characterized in that, The step of grouping the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group includes: The degree of partial discharge is determined based on the partial discharge dataset. The partial discharge dataset is then grouped into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group. The first partial discharge detection data group corresponds to the late stage of partial discharge, the second partial discharge detection data group corresponds to the middle stage of partial discharge, and the third partial discharge detection data group corresponds to the early stage of partial discharge.

3. The method according to claim 1, characterized in that, The step of grouping the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group includes: Based on the partial discharge dataset, the partial discharge type is determined, and the partial discharge dataset is grouped into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group. The first partial discharge detection data group corresponds to the pulsed discharge, the second partial discharge detection data group corresponds to the continuous discharge, and the third partial discharge detection data group corresponds to the random discharge.

4. The method according to claim 2 or 3, characterized in that, The step of asynchronously training the partial discharge detection model using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set includes: The partial discharge detection model is trained step by step using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set.

5. The method according to claim 4, characterized in that, The step of training the partial discharge detection model stepwise using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set includes: A first partial discharge detection model is obtained from the first partial discharge detection data set, and the output accuracy of the first partial discharge detection model meets a first threshold. A second partial discharge detection model is obtained through the second partial discharge detection data set, and the output accuracy of the second partial discharge detection model meets the second threshold. The third partial discharge detection model is obtained through the third partial discharge detection data set, and the output accuracy of the third partial discharge detection model meets the third threshold.

6. The method according to claim 5, wherein obtaining the second partial discharge detection model through the second partial discharge detection data set comprises: The second partial discharge detection model is obtained by training the first partial discharge detection model with the second partial discharge detection data set, and the second threshold is less than the first threshold. The process of obtaining the third partial discharge detection model from the third partial discharge detection data set includes: The third partial discharge detection model is obtained by training the second partial discharge detection model using the third partial discharge detection data set, wherein the third threshold is less than the second threshold.

7. The method according to claim 2 or 3, characterized in that, The step of simultaneously training the partial discharge detection model using the first partial discharge detection data set, the second partial discharge detection data set, and the third partial discharge detection data set includes: The partial discharge detection model is trained synchronously using partial discharge detection data from the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group.

8. A device for constructing a partial discharge detection model, characterized in that, include: The acquisition module is used to acquire a partial discharge detection dataset, which includes multiple partial discharge detection data, and the partial discharge detection data includes noise-reduced partial discharge signals. The grouping module is used to group the partial discharge dataset into a first partial discharge detection data group, a second partial discharge detection data group, and a third partial discharge detection data group, wherein the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group each include multiple partial discharge detection data; The training module is used to synchronously or asynchronously train the partial discharge detection model using the first partial discharge detection data group, the second partial discharge detection data group, and the third partial discharge detection data group, so that the output result of the partial discharge detection model meets the threshold requirements.

9. An electronic device, characterized in that, include: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for constructing a partial discharge detection model as described in any one of claims 1-7.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the method for constructing a partial discharge detection model as described in any one of claims 1-7.

Images