Electronic voltage transformer state identification method and system under limited sample

Abstract

The invention relates to a method and a system for identifying the state of an electronic voltage transformer under a limited sample. The method comprises the following steps: acquiring operation data of the electronic voltage transformer and preprocessing the operation data; creating an initial feature set based on the operation data, and setting a feature combination strategy to recombine the initial feature set to obtain a fused feature set; constructing an operation data sample expansion model, training the operation data sample expansion model based on the fused feature set, and expanding the operation data through the trained operation data sample expansion model to obtain an operation data expansion sample set; constructing an electronic voltage transformer state recognition model, and performing adaptive adjustment on parameters of the electronic voltage transformer state recognition model based on the line topology features; and applying the adaptive adjustment parameters to an electronic voltage transformer state identification model, and identifying the state of the electronic voltage transformer by operating the data expansion sample set.

Description

Technical Field

[0001] The invention relates to a method and a system for identifying the state of an electronic voltage transformer under limited samples, and belongs to the technical field of electric power metering inaccuracy identification. Background Art

[0002] Electronic Voltage Transformer (EVT) is one of the key devices in modern power systems, used to measure the voltage of high-voltage transmission lines. Compared with traditional electromagnetic voltage transformers, EVT has the advantages of small size, light weight, fast response speed, and no ferromagnetic resonance. It is widely used in smart grids, substation automation and other fields. The normal operation of EVT is crucial to the safety and stability of the power system, so real-time monitoring of its status and fault diagnosis are of great significance.

[0003] In actual operation, abnormal conditions of EVT may be rare, and abnormal signals may be very close to normal signals and difficult to distinguish. In many cases, only data of normal operating conditions can be obtained, lacking abnormal samples, which makes traditional supervised learning methods difficult to apply. Summary of the invention

[0004] In order to solve the above problems existing in the prior art, the present invention proposes a method and system for identifying the state of an electronic voltage transformer under limited samples.

[0005] The technical solution of the present invention is as follows: On the one hand, the present invention provides a method for identifying the state of an electronic voltage transformer under a limited sample, comprising the following steps: Collect the operation data of the electronic voltage transformer and pre-process the operation data; Create an initial feature set based on the preprocessed running data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; Constructing an operation data sample expansion model, training the operation data sample expansion model based on the fusion feature set, expanding the operation data through the trained operation data sample expansion model, and obtaining an operation data expansion sample set; An electronic voltage transformer state recognition model is constructed based on the DBSCAN algorithm, and the parameters of the electronic voltage transformer state recognition model are adaptively adjusted based on the line topology characteristics of the electronic voltage transformer. The adaptive adjustment parameters are applied to the electronic voltage transformer state recognition model, and the electronic voltage transformer state is identified by expanding the sample set through running data.

[0006] Preferably, the operating data of the electronic voltage transformer includes voltage waveform, current waveform, temperature and humidity; The running data is normalized as shown in the following formula: ; in: Indicates the collected operation data; represents the mean of the running data; represents the standard deviation of the running data; Represents normalized running data.

[0007] Preferably, an initial feature set is created based on the preprocessed operating data, and the initial feature set includes the mean, variance, skewness, kurtosis, waveform factor, and kurtosis index of the preprocessed operating data.

[0008] Preferably, the running data sample expansion model is constructed based on a GAN model.

[0009] Preferably, a feature combination strategy is set to recombine the initial feature set to obtain a fused feature set, specifically: For each feature sample in the initial feature set, calculate its impact on the output of the running data sample expansion model, as shown in the following formula: ; in: Indicates running the data sample expansion model function; represents the initial feature set; represents the first Feature samples; Indicates The influence of each characteristic sample; The influence of all feature samples is normalized, and based on the normalized influence of the feature samples, the weight of each feature sample is calculated and nonlinear transformation is performed, as shown in the following formula: ; in: Indicates feature sample weights; After normalization, Feature samples; After normalization, Feature samples; represents a natural constant; Each feature sample is weighted and fused with its corresponding weight to obtain a fused feature set.

[0010] Preferably, the fused feature set is introduced into the GAN model discriminator as conditional information, as shown in the following formula: ; in: represents the discriminator function; represents the fusion feature set; Indicates true or false labels; During the training process, the initial feature set is used as the conditional vector, and a random noise vector is created at the same time. The conditional vector and the random noise vector are input into the GAN model generator to generate samples. The discriminator is then used to determine the true or false labels of the generated samples. The loss is calculated based on the judgment accuracy, and the GAN model parameters are updated based on the loss through the back propagation algorithm.

[0011] Preferably, the specific steps of adaptively adjusting the parameters of the electronic voltage transformer state identification model based on the line topology characteristics of the electronic voltage transformer are: Based on the electronic voltage transformer topology information, a line topology characteristic graph of the electronic voltage transformer is constructed, wherein each node of the graph represents an electronic voltage transformer, and each edge of the graph represents a connection status between the electronic voltage transformers; For each node in the voltage transformer topology characteristic graph, its local density index is calculated, as shown in the following formula: ; in: Indicates The local density index of each node; Indicates A node and its Distances of neighboring points; Calculate the topological weight of each node as shown below: ; in: Indicates The topological weight of each node; Indicates The number of connections between a node and other nodes; Indicates the total number of nodes; , Represent the importance coefficients of the current node degree and the total number of nodes respectively; Initialize the initial radius and initial minimum number of sample points of the DBSCAN algorithm; For each node, the minimum number of sample points is adjusted according to its topological weight, as shown in the following formula: ; in: Indicates The minimum number of sample points after node adjustment; Indicates the initial minimum number of sample points; For each node, its radius is adjusted based on the local density index, as shown in the following formula: ; in: Indicates The radius of each node after adjustment; represents the initial radius; Represents the average value of local density index; represents the adjustment factor; For each node that the DBSCAN algorithm has not visited, determine whether it is a core point. If it is a core point, start a new cluster and recursively expand the cluster; Non-core points are marked as boundary points, isolated points that cannot be expanded are marked as noise points, and noise points are regarded as outliers.

[0012] On the other hand, the present invention also provides an electronic voltage transformer state recognition system under limited samples, including a data acquisition module, a feature fusion module, a sample expansion module and a state recognition module; The data acquisition module is used to collect the operation data of the electronic voltage transformer and pre-process the operation data; The feature fusion module is used to create an initial feature set based on the preprocessed operation data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; The sample expansion module is used to construct an operation data sample expansion model, train the operation data sample expansion model based on the fusion feature set, expand the operation data through the trained operation data sample expansion model, and obtain an operation data expansion sample set; The state recognition module is used to construct an electronic voltage transformer state recognition model based on the DBSCAN algorithm, and adaptively adjust the parameters of the electronic voltage transformer state recognition model based on the line topology characteristics of the electronic voltage transformer; apply the adaptively adjusted parameters to the electronic voltage transformer state recognition model, and identify the electronic voltage transformer state by expanding the sample set through running data.

[0013] On the other hand, the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the present invention when executing the program.

[0014] In yet another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the method of the present invention when executed by a processor.

[0015] The present invention has the following beneficial effects: 1. The present invention proposes a GAN algorithm based on the improved characteristic conditions of the mutual inductor signal, which performs data expansion according to the characteristics of the mutual inductor data and solves the problem of insufficient mutual inductor data samples.

[0016] 2. The present invention proposes a DBSCAN classification algorithm that performs adaptive parameter adjustment based on line topology structure, which can enhance the robustness and recognition accuracy of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0020] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0021] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0022] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0023] Embodiment 1: See also Figure 1 , a method for identifying the state of an electronic voltage transformer under finite samples, comprising the following steps: Collect the operating data of the electronic voltage transformer (these data can come from multiple sensors, including voltage sensors, current sensors, temperature sensors, humidity sensors, etc. The data is continuous time series data, collected once per second. When collecting data, each data point can be labeled with status information, such as "normal", "abnormal", etc. Although this method mainly focuses on normal data, in the subsequent verification and testing stages, the labeled data can help evaluate the performance of the model), and preprocess the operating data; Create an initial feature set based on the preprocessed running data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; Constructing an operation data sample expansion model, training the operation data sample expansion model based on the fusion feature set, expanding the operation data through the trained operation data sample expansion model, and obtaining an operation data expansion sample set; An electronic voltage transformer state recognition model is constructed based on the DBSCAN algorithm, and the parameters of the electronic voltage transformer state recognition model are adaptively adjusted based on the line topology characteristics of the electronic voltage transformer. The adaptive adjustment parameters are applied to the electronic voltage transformer state recognition model, and the electronic voltage transformer state is identified by expanding the sample set through running data.

[0024] As a preferred implementation of this embodiment, the operation data of the electronic voltage transformer includes voltage waveform, current waveform, temperature and humidity; The running data is normalized as shown in the following formula: ; in: Indicates the collected operation data; represents the mean of the running data; represents the standard deviation of the running data; Represents normalized running data.

[0025] As a preferred implementation mode of this embodiment, an initial feature set is created based on the preprocessed operating data, and the initial feature set includes the mean, variance, skewness, kurtosis, waveform factor, and kurtosis index of the preprocessed operating data; these features can provide different perspectives on the state of the transformer, but each of them may not be sufficient to fully describe complex failure modes.

[0026] As a preferred implementation of this embodiment, the running data sample expansion model is constructed based on the GAN model.

[0027] As a preferred implementation of this embodiment, a feature combination strategy is set to recombine the initial feature set to obtain a fused feature set, specifically: For each feature sample in the initial feature set, calculate its impact on the output of the running data sample expansion model, as shown in the following formula: ; in: Indicates running the data sample expansion model function; represents the initial feature set; represents the first Feature samples; Indicates The influence of each characteristic sample; Normalize the influence of all feature samples (using methods such as minimum-maximum normalization or Z-score standardization), calculate the weight of each feature sample based on the normalized influence of the feature samples, and perform nonlinear transformation to ensure that the sum of the weights is 1, as shown in the following formula: ; in: Indicates feature sample weights; After normalization, Feature samples; After normalization, Feature samples; represents a natural constant; Each feature sample is weighted and fused with its corresponding weight to obtain a fused feature set.

[0028] As a preferred implementation of this embodiment, the fused feature set is introduced into the GAN model discriminator as conditional information, as shown in the following formula: ; in: represents the discriminator function; represents the fusion feature set, , Represents the total number of samples in the initial feature set; Indicates true or false labels; The discriminator function is specifically shown in the following formula: ; During the training process, the initial feature set is used as the conditional vector, and a random noise vector is created at the same time. The conditional vector and the random noise vector are input into the GAN model generator to generate samples. The discriminator is then used to determine the true or false labels of the generated samples. The loss is calculated based on the judgment accuracy, and the GAN model parameters are updated based on the loss through the back propagation algorithm.

[0029] As a preferred implementation of this embodiment, the specific steps of adaptively adjusting the parameters of the electronic voltage transformer state identification model based on the electronic voltage transformer line topology characteristics are: Based on the electronic voltage transformer topology information, a line topology characteristic graph of the electronic voltage transformer is constructed, wherein each node of the graph represents an electronic voltage transformer, and each edge of the graph represents a connection status between the electronic voltage transformers; For each node in the voltage transformer topology characteristic graph, its local density index is calculated, as shown in the following formula: ; in: Indicates The local density index of each node; Indicates A node and its Distances of neighboring points; Calculate the topological weight of each node as shown below: ; in: Indicates The topological weight of each node; Indicates The number of connections between a node and other nodes; Indicates the total number of nodes; , Represent the importance coefficients of the current node degree and the total number of nodes respectively; Initialize the initial radius and initial minimum number of sample points of the DBSCAN algorithm; For each node, the minimum number of sample points is adjusted according to its topological weight, as shown in the following formula: ; in: Indicates The minimum number of sample points after node adjustment; Indicates the initial minimum number of sample points; For each node, its radius is adjusted based on the local density index, as shown in the following formula: ; in: Indicates The radius of each node after adjustment; represents the initial radius; Represents the average value of local density index; represents the adjustment factor; For each node that the DBSCAN algorithm has not visited, determine whether it is a core point. If it is a core point, start a new cluster and recursively expand the cluster; Non-core points are marked as boundary points, isolated points that cannot be expanded are marked as noise points, and noise points are regarded as abnormal points. The status of each electronic voltage transformer is judged by the abnormal points and normal points.

[0030] Embodiment 2: An electronic voltage transformer state recognition system under finite samples, comprising a data acquisition module, a feature fusion module, a sample expansion module and a state recognition module; The data acquisition module is used to collect the operation data of the electronic voltage transformer and pre-process the operation data; The feature fusion module is used to create an initial feature set based on the preprocessed operation data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; The sample expansion module is used to construct an operation data sample expansion model, train the operation data sample expansion model based on the fusion feature set, expand the operation data through the trained operation data sample expansion model, and obtain an operation data expansion sample set; The state recognition module is used to construct an electronic voltage transformer state recognition model based on the DBSCAN algorithm, and adaptively adjust the parameters of the electronic voltage transformer state recognition model based on the line topology characteristics of the electronic voltage transformer; apply the adaptively adjusted parameters to the electronic voltage transformer state recognition model, and identify the electronic voltage transformer state by expanding the sample set through running data.

[0031] The system is used to implement the method in Example 1, which will not be described in detail here.

[0032] Embodiment three: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present invention when executing the program.

[0033] Embodiment 4: This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present invention is implemented.

[0034] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0035] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0036] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0037] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0038] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for identifying the state of an electronic voltage transformer under finite samples, characterized in that: The following steps are involved: Collect the operation data of the electronic voltage transformer and pre-process the operation data; Create an initial feature set based on the preprocessed running data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; Constructing an operation data sample expansion model, training the operation data sample expansion model based on the fusion feature set, expanding the operation data through the trained operation data sample expansion model, and obtaining an operation data expansion sample set; An electronic voltage transformer state recognition model is constructed based on the DBSCAN algorithm, and the parameters of the electronic voltage transformer state recognition model are adaptively adjusted based on the line topology characteristics of the electronic voltage transformer. The adaptive adjustment parameters are applied to the electronic voltage transformer state recognition model, and the electronic voltage transformer state is identified by expanding the sample set through running data.

2. The method for identifying the state of an electronic voltage transformer under limited samples according to claim 1 is characterized in that: The operation data of the electronic voltage transformer includes voltage waveform, current waveform, temperature and humidity; The running data is normalized as shown in the following formula: ； in: Indicates the collected operation data; represents the mean of the running data; represents the standard deviation of the running data; Represents normalized running data.

3. The method for identifying the state of an electronic voltage transformer under finite samples according to claim 1, characterized in that: An initial feature set is created based on the preprocessed operating data, wherein the initial feature set includes a mean, a variance, a skewness, a kurtosis, a waveform factor, and a kurtosis index of the preprocessed operating data.

4. The method for identifying the state of an electronic voltage transformer under finite samples according to claim 1, characterized in that: The running data sample expansion model is constructed based on the GAN model.

5. The method for identifying the state of an electronic voltage transformer under limited samples according to claim 4 is characterized in that: Set the feature combination strategy to recombine the initial feature set to obtain the fused feature set, specifically: For each feature sample in the initial feature set, calculate its impact on the output of the running data sample expansion model, as shown in the following formula: ； in: Indicates running the data sample expansion model function; represents the initial feature set; represents the first Feature samples; Indicates The influence of each characteristic sample; The influence of all feature samples is normalized, and based on the normalized influence of the feature samples, the weight of each feature sample is calculated and nonlinear transformation is performed, as shown in the following formula: ； in: Indicates feature sample weights; After normalization, Feature samples; After normalization, Feature samples; represents a natural constant; Each feature sample is weighted and fused with its corresponding weight to obtain a fused feature set.

6. The method for identifying the state of an electronic voltage transformer under limited samples according to claim 5, characterized in that: The fused feature set is introduced into the GAN model discriminator as conditional information, as shown in the following formula: ； in: represents the discriminator function; represents the fusion feature set; Indicates true or false labels; During the training process, the initial feature set is used as the conditional vector, and a random noise vector is created at the same time. The conditional vector and the random noise vector are input into the GAN model generator to generate samples. The discriminator is then used to determine the true or false labels of the generated samples. The loss is calculated based on the judgment accuracy, and the GAN model parameters are updated based on the loss through the back propagation algorithm.

7. The method for identifying the state of an electronic voltage transformer under finite samples according to claim 1, characterized in that: The specific steps of adaptively adjusting the parameters of the electronic voltage transformer state recognition model based on the line topology characteristics of the electronic voltage transformer are as follows: Based on the electronic voltage transformer topology information, a line topology characteristic graph of the electronic voltage transformer is constructed, wherein each node of the graph represents an electronic voltage transformer, and each edge of the graph represents a connection status between the electronic voltage transformers; For each node in the voltage transformer topology characteristic graph, its local density index is calculated, as shown in the following formula: ； in: Indicates The local density index of each node; Indicates A node and its Distances of neighboring points; Calculate the topological weight of each node as shown below: ； in: Indicates The topological weight of each node; Indicates The number of connections between a node and other nodes; Indicates the total number of nodes; , Represent the importance coefficients of the current node degree and the total number of nodes respectively; Initialize the initial radius and initial minimum number of sample points of the DBSCAN algorithm; For each node, the minimum number of sample points is adjusted according to its topological weight, as shown in the following formula: ； in: Indicates The minimum number of sample points after node adjustment; Indicates the initial minimum number of sample points; For each node, its radius is adjusted based on the local density index, as shown in the following formula: ； in: Indicates The radius of each node after adjustment; represents the initial radius; Represents the average value of local density index; represents the adjustment factor; For each node that the DBSCAN algorithm has not visited, determine whether it is a core point. If it is a core point, start a new cluster and recursively expand the cluster; Non-core points are marked as boundary points, isolated points that cannot be expanded are marked as noise points, and noise points are regarded as outliers.

8. An electronic voltage transformer state identification system under finite samples, characterized in that: It includes data acquisition module, feature fusion module, sample expansion module and state recognition module; The data acquisition module is used to collect the operation data of the electronic voltage transformer and pre-process the operation data; The feature fusion module is used to create an initial feature set based on the preprocessed running data, set a feature combination strategy to recombine the initial feature set, and obtain a fused feature set; The sample expansion module is used to construct an operation data sample expansion model, train the operation data sample expansion model based on the fusion feature set, expand the operation data through the trained operation data sample expansion model, and obtain an operation data expansion sample set; The state recognition module is used to construct an electronic voltage transformer state recognition model based on the DBSCAN algorithm, and adaptively adjust the parameters of the electronic voltage transformer state recognition model based on the line topology characteristics of the electronic voltage transformer; apply the adaptively adjusted parameters to the electronic voltage transformer state recognition model, and identify the electronic voltage transformer state by expanding the sample set through running data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images