Method and device for monitoring operation state of voltage transformer

Abstract

The invention discloses a method and a device for monitoring the running state of a voltage transformer, relates to the technical field of detection, and mainly aims to solve the problem of poor effectiveness of running state monitoring of an existing voltage transformer. Comprising the following steps: acquiring voltage data of a voltage transformer, and performing principal component analysis on the voltage data based on a kernel principal component analysis model which completes model training to obtain an analysis result; determining a fault evaluation standard threshold value through residual space statistics obtained through the analysis result, and determining an abnormal operation state of the voltage transformer when the voltage data is matched with the fault evaluation standard threshold value; and classifying the voltage data of the abnormal operation state to obtain an abnormal voltage phase, and determining the abnormal voltage phase as a monitoring result of the abnormal operation state of the voltage transformer.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to a method and device for monitoring the operating status of a voltage transformer. Background Art

[0002] In recent years, the construction of UHV transmission lines in my country has been accelerating, and the requirements for safe and stable operation of power lines have been increasing. Among them, capacitive voltage transformers (CVTs) have the advantages of high insulation strength, small size, light weight, and no ferromagnetic resonance in high-voltage power systems, and are usually used in the field of voltage monitoring of transmission lines. However, due to the influence of factors such as raw materials, manufacturing experience, and system overvoltage, CVTs may have various faults during operation. How to effectively monitor the operating status of CVTs and promptly detect and eliminate CVT operating faults is of great significance to the safe and reliable operation of the power grid.

[0003] At present, the fault detection of the operating status of the existing capacitive voltage transformer CVT equipment usually requires power outage detection, and the equipment operating status is evaluated by measuring parameters such as capacitance, insulation resistance and dielectric loss in the power outage state to achieve the monitoring purpose. However, periodic power outage detection is difficult to implement in the industrial production process, especially during the peak period of electricity consumption. If the maintenance cycle is reached, but the line cannot be shut down for maintenance, it is easy to cause the equipment to expire. Once the line is powered off, both ends need to be shut down, which greatly reduces the equipment utilization rate, affects production operations, and reduces economic benefits. In addition, during the power outage calibration cycle, a large amount of manpower costs need to be invested, resulting in an increase in the workload of the staff, resulting in a large amount of human resource waste, and it is easy to cause missed detection or false detection. Therefore, there is an urgent need for a voltage transformer operating status monitoring to solve the above problems. Summary of the invention

[0004] In view of this, the present invention provides a method and device for monitoring the operating status of a voltage transformer, the main purpose of which is to solve the problem of poor effectiveness of existing voltage transformer operating status monitoring.

[0005] According to one aspect of the present invention, a method for monitoring the operating status of a voltage transformer is provided, comprising:

[0006] Acquire voltage data of the voltage transformer, and perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed model training to obtain an analysis result;

[0007] Determining a fault assessment standard threshold value through the residual space statistics obtained by the analysis result, and determining an abnormal operating state of the voltage transformer when the voltage data matches the fault assessment standard threshold value;

[0008] The voltage data of the abnormal operation state is classified to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operation state of the voltage transformer.

[0009] Furthermore, before performing principal component analysis on the voltage data based on the kernel principal component analysis model that has completed model training and obtaining the analysis result, the method further includes:

[0010] Acquiring voltage sample data of a voltage transformer, and performing normalization processing on the voltage sample data;

[0011] The kernel principal component analysis model is trained based on the voltage sample data that has undergone normalization processing to complete the model training.

[0012] Furthermore, the kernel principal component analysis model is trained based on the voltage sample data that has completed the normalization process, and completing the model training includes:

[0013] Performing high-dimensional space mapping on the voltage sample data that has completed the standardization processing based on a kernel function, and calculating the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors;

[0014] Performing principal component decomposition on the eigenvalues ​​and the unit orthogonal eigenvectors to determine a principal component space matrix and a residual space matrix;

[0015] When the cumulative variance percentage of the principal component number is greater than a preset expected value, the training of the kernel principal component analysis model is completed.

[0016] Furthermore, the normalizing process of the voltage sample data includes:

[0017] Performing data completion processing on the voltage sample data by interpolation, and deleting abnormal sample data in the voltage sample data;

[0018] The voltage sample data after the abnormal sample data is deleted is operated with the data average value and the data standard deviation to complete the normalization processing of the voltage sample data.

[0019] Furthermore, the method further comprises:

[0020] When it is detected that the voltage sample data is updated, the data sliding window width is determined, and the original voltage sample data is selected based on the data sliding window width, and the step of re-executing the model training of the kernel principal component analysis model is combined with the new voltage sample data, so as to re-perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed the model training, and dynamically update the fault assessment standard threshold.

[0021] Further, determining the data sliding window width includes:

[0022] The data sliding window width is calculated based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2, wherein the data sliding window width is negatively correlated with the residual space statistic.

[0023] Further, the voltage data of the abnormal operation state is classified to obtain the abnormal voltage phase, and the monitoring result determined as the abnormal operation state of the voltage transformer includes:

[0024] The Minkowski distance of the voltage data of the abnormal operation state is calculated based on a k-nearest neighbor algorithm; and an abnormal voltage phase is determined based on the Minkowski distance to determine the abnormal voltage phase as a monitoring result of a normal operation state.

[0025] According to another aspect of the present invention, there is provided a voltage transformer operating status monitoring device, comprising:

[0026] An acquisition module is used to acquire voltage data of a voltage transformer, and perform principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result;

[0027] A first determination module, configured to determine a fault assessment standard threshold value through a residual space statistic obtained from the analysis result, and to determine an abnormal operation state of the voltage transformer when the voltage data matches the fault assessment standard threshold value;

[0028] The second determination module is used to classify the voltage data of the abnormal operation state, obtain the abnormal voltage phase, and determine it as the monitoring result of the abnormal operation state of the voltage transformer.

[0029] Furthermore, the device also includes:

[0030] A processing module, used for acquiring voltage sample data of a voltage transformer and performing normalization processing on the voltage sample data;

[0031] The training module is used to perform model training on the kernel principal component analysis model based on the voltage sample data that has completed the normalization process, so as to complete the model training.

[0032] Furthermore, the training module is specifically used to perform high-dimensional space mapping on the voltage sample data that has completed standard processing based on a kernel function, and calculate the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors; perform principal component decomposition on the eigenvalues ​​and the unit orthogonal eigenvectors to determine the principal component space matrix and the residual space matrix; when the cumulative variance percentage of the principal component number is greater than a preset expected value, the training of the kernel principal component analysis model is completed.

[0033] Furthermore, the processing module is specifically used to complete the voltage sample data by interpolation and delete abnormal sample data in the voltage sample data; based on the voltage sample data with the abnormal sample data deleted, the voltage sample data is operated with the data average and the data standard deviation to complete the normalization processing of the voltage sample data.

[0034] Furthermore, the device also includes:

[0035] An execution module is used to determine the data sliding window width when it is detected that the voltage sample data is updated, and select the original voltage sample data based on the data sliding window width, and re-execute the step of training the kernel principal component analysis model in combination with the new voltage sample data, so as to re-perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed the model training, and dynamically update the fault assessment standard threshold.

[0036] Furthermore, the execution module is specifically used to calculate the data sliding window width based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2, wherein the data sliding window width is negatively correlated with the residual space statistic.

[0037] Furthermore, the second determination module is specifically configured to calculate the Minkowski distance of the voltage data of the abnormal operation state based on a k-nearest neighbor algorithm; and determine the abnormal voltage phase based on the Minkowski distance, so as to determine the abnormal voltage phase as a monitoring result of the normal operation state.

[0038] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned voltage transformer operating status monitoring method.

[0039] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0040] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned voltage transformer operating status monitoring method.

[0041] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0042] The present invention provides a method and device for monitoring the operating status of a voltage transformer. Compared with the prior art, the embodiment of the present invention obtains voltage data of a voltage transformer, and performs principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result; determines a fault assessment standard threshold by using a residual space statistic obtained from the analysis result, and determines the abnormal operating status of the voltage transformer when the voltage data matches the fault assessment standard threshold; classifies the voltage data of the abnormal operating status to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operating status of the voltage transformer, thereby achieving the purpose of voltage transformer detection without power outages in continuous industrial production, meeting the flexibility requirements for effectively using voltage transformers in different industrial productions, and greatly reducing the monitoring resource consumption of the voltage transformer, avoiding missed detection and false detection of the operating status of the voltage transformer, thereby improving the monitoring effectiveness of the operating status of the voltage transformer.

[0043] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0045] Figure 1 A flow chart of a method for monitoring the operating status of a voltage transformer provided by an embodiment of the present invention is shown;

[0046] Figure 2 A schematic diagram of data updating provided by an embodiment of the present invention is shown;

[0047] Figure 3 A block diagram showing a voltage transformer operating status monitoring device provided by an embodiment of the present invention is shown;

[0048] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0049] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0050] The embodiment of the present invention provides a method for monitoring the operating status of a voltage transformer, such as Figure 1 As shown, the method includes:

[0051] 101. Obtain voltage data of a voltage transformer, and perform principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result.

[0052] In an embodiment of the present invention, the voltage transformer is a capacitive voltage transformer CVT, and the high voltage is measured to obtain voltage data in real time. The current execution end serves as the execution subject for monitoring the operating status of the voltage transformer, which can be a server or a processing terminal, and the embodiment of the present invention does not make specific limitations. After the voltage data is obtained, the voltage data is subjected to principal component analysis based on the kernel principal component analysis model that completes the model training to obtain the analysis result. In an embodiment of the present invention, kernel principal component analysis (KPCA) is a technology used in the field of multivariate statistics. It is an extension of principal component analysis (PCA). It achieves dimensionality reduction of nonlinear data by introducing kernel methods. It is not only applicable to linearly separable data sets, but also can process linearly inseparable data sets. It maps data from the input space to a high-dimensional or even infinite-dimensional feature space through nonlinear mapping, and then performs PCA dimensionality reduction in the high-dimensional space. At this time, the kernel principal component analysis model is a model constructed on the kernel principal component analysis algorithm for performing principal component analysis of voltage data.

[0053] In another embodiment of the present invention, for further definition and explanation, the step performs principal component analysis on the voltage data based on the kernel principal component analysis model that completes model training, and before obtaining the analysis result, the method further includes:

[0054] Acquiring voltage sample data of a voltage transformer, and performing normalization processing on the voltage sample data;

[0055] The kernel principal component analysis model is trained based on the voltage sample data that has undergone normalization processing to complete the model training.

[0056] In order to analyze the voltage data based on the principal component analysis method to divide the effective principal component space and residual space, specifically, the current execution end also obtains the voltage sample data of the voltage transformer to perform model training of the kernel principal component analysis model. Among them, the voltage sample data is the collected secondary voltage sample data of the voltage transformer. At the same time, in order to improve the accuracy of the model training, the current execution end also normalizes the voltage sample data so that the normalized voltage sample data is input into the kernel principal component analysis model for training.

[0057] In another embodiment of the present invention, for further definition and explanation, the step of performing model training on the kernel principal component analysis model based on the voltage sample data that has completed the normalization process, and completing the model training includes:

[0058] Performing high-dimensional space mapping on the voltage sample data that has completed the standardization processing based on a kernel function, and calculating the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors;

[0059] Performing principal component decomposition on the eigenvalues ​​and the unit orthogonal eigenvectors to determine a principal component space matrix and a residual space matrix;

[0060] When the cumulative variance percentage of the principal component number is greater than a preset expected value, the training of the kernel principal component analysis model is completed.

[0061] In order to achieve effective training based on the kernel principal component analysis model, specifically, when the current execution end is performing model training, specifically, firstly, based on the kernel function, the voltage sample data that has completed the standard processing is mapped to a high-dimensional space. Among them, the kernel function is to use the calculation on the low-dimensional feature space to avoid the excessive amount of calculation of the vector inner product in the high-dimensional feature space. The kernel function in the embodiment of the present invention can be a linear kernel function, a Gaussian kernel function, etc., without specific limitation, so as to map the voltage sample data in the high-dimensional space. Furthermore, the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors are currently calculated. At this time, the calculation method of the eigenvalues ​​and unit orthogonal eigenvectors of the autocovariance matrix is ​​not specifically limited in the embodiment of the present invention. After obtaining the eigenvalues ​​and unit orthogonal eigenvectors, the current execution end performs principal component decomposition on the eigenvalues ​​and unit orthogonal unit vectors through kernel principal component analysis to determine the principal component space matrix and the residual space matrix.

[0062] In a specific embodiment, the kernel function maps the data set X to a new high-dimensional feature space, and the mapped data set Ψ is expressed as: Ψ = [σ(x1), σ(x2), ...σ(x n )]; where σ(x i ) is the standard deviation,

[0063] μ is the mean value, m represents the number of variables in the sample, and n is the number of samples.

[0064] It should be noted that after determining the principal component space matrix and the residual space matrix, the cumulative variance percentage of the principal element number in the principal component space matrix is ​​counted and compared with the preset expected value. When the cumulative variance percentage is greater than the preset expected variance, the model training is stopped and the training of the kernel principal component analysis model is completed. Otherwise, the model training can be continued. Among them, the expected expected variance can be configured according to the training request requirements of the model, such as 85%, which is not specifically limited in the embodiment of the present invention.

[0065] In another embodiment of the present invention, for further definition and explanation, the step of normalizing the voltage sample data includes:

[0066] Performing data completion processing on the voltage sample data by interpolation, and deleting abnormal sample data in the voltage sample data;

[0067] The voltage sample data after the abnormal sample data is deleted is operated with the data average value and the data standard deviation to complete the normalization processing of the voltage sample data.

[0068] In order to achieve the purpose of normalization processing of voltage sample data, thereby improving the training accuracy of the model, when the current execution end performs normalization processing, specifically, the voltage sample data is supplemented based on the interpolation method. At this time, the interpolation method can be linear interpolation, polynomial interpolation, etc., which is not specifically limited in the embodiment of the present invention. Furthermore, the abnormal sample data is deleted from the voltage sample data after the supplementation processing. At this time, the abnormal sample data can be determined by the historical fault assessment standard threshold, and can also be screened based on the pre-configured abnormal threshold. The embodiment of the present invention is not specifically limited. At the same time, after the current execution end deletes the abnormal sample data, the data average value and data standard deviation of the voltage sample data are calculated to complete the normalization processing. At this time, the calculation based on the voltage sample data, the data average value and data standard deviation of the voltage sample data is to use the voltage sample data value minus the data average value, and then divide it by the data standard deviation, which can make the sample data more standardized as a whole, reduce the fluctuation of data changes, prevent the interference caused by the inconsistency of variables, ensure the representativeness of the sample data, and ensure the accuracy of the fault assessment standard threshold obtained when training the principal component analysis model.

[0069] 102. Determine a fault assessment standard threshold value through the residual space statistics obtained by the analysis result, and determine an abnormal operating state of the voltage transformer when the voltage data matches the fault assessment standard threshold value.

[0070] In an embodiment of the present invention, the current execution end can obtain the analysis results including the principal component space and the residual space after the principal component analysis is processed, and then the statistics of the residual space are calculated. Among them, the fault assessment standard threshold is used to determine whether there is an abnormal threshold in the voltage transformer, and it is determined in real time based on the residual space statistics, that is, after each kernel principal component analysis is performed on the collected voltage data, the residual space statistics are calculated, so as to update the fault assessment standard threshold in real time, so that the abnormal operation judgment is more accurate. At this time, if the voltage data matches the fault assessment standard threshold, it is determined that the operating state of the voltage transformer is an abnormal operating state. Among them, it can be configured as a fault assessment standard threshold according to the residual space statistics, or it can be queried and configured based on the correspondence between the residual space statistics and each preset threshold, and the embodiment of the present invention does not make specific limitations.

[0071] It should be noted that when determining the fault assessment standard threshold value through the residual space statistics obtained through the analysis result, since the residual space statistics are obtained based on the analysis result, the specific steps include:

[0072] (1) Calculate the eigenvalues ​​of the autocovariance matrix corresponding to the sample data and its corresponding unit orthogonal eigenvectors to obtain the score matrix and loading matrix.

[0073] (2) The coefficient matrix is ​​subjected to principal component decomposition to obtain a principal component space matrix and a residual space matrix. The coefficient matrix is ​​the result of the transpose of the load matrix multiplied by the score matrix.

[0074] (3) Determine the number of principal components n. The number of principal components is taken from the principal component space matrix. Take the first n largest load vectors and the first n eigenvalues. λ is the eigenvalue. Calculate the percentage of the cumulative variance of the first n principal components so that it is greater than or equal to 85%. 85% is determined based on experience.

[0075] (4) Calculate the residual space statistic Q to determine the fault assessment standard threshold.

[0076] 103. Classify the voltage data of the abnormal operation state to obtain an abnormal voltage phase, and determine it as a monitoring result of the abnormal operation state of the voltage transformer.

[0077] In an embodiment of the present invention, after the current execution end determines that the state of the voltage transformer is an abnormal operating state, in order to further obtain the abnormal situation of the voltage transformer, the current execution end classifies the voltage data belonging to the abnormal operating state to determine the abnormal voltage phase, and uses this abnormal voltage phase as the monitoring result of the abnormal operating state of the voltage transformer, thereby greatly improving the state detection accuracy of the voltage transformer.

[0078] The embodiment of the present invention provides a method for monitoring the operating status of a voltage transformer. Compared with the prior art, the embodiment of the present invention obtains voltage data of a voltage transformer, and performs principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result; determines a fault assessment standard threshold through a residual space statistic obtained from the analysis result, and determines the abnormal operating status of the voltage transformer when the voltage data matches the fault assessment standard threshold; classifies the voltage data of the abnormal operating status to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operating status of the voltage transformer, thereby achieving the purpose of uninterrupted voltage transformer detection in continuous industrial production, meeting the flexibility requirements for effectively using voltage transformers in different industrial productions, and greatly reducing the monitoring resource consumption of the voltage transformer, avoiding missed detection and false detection of the operating status of the voltage transformer, thereby improving the monitoring effectiveness of the operating status of the voltage transformer.

[0079] In another embodiment of the present invention, for further definition and explanation, the step of classifying the voltage data of the abnormal operating state to obtain the abnormal voltage phase, and determining the monitoring result of the abnormal operating state of the voltage transformer includes:

[0080] The Minkowski distance of the voltage data of the abnormal operation state is calculated based on a k-nearest neighbor algorithm; and an abnormal voltage phase is determined based on the Minkowski distance to determine the abnormal voltage phase as a monitoring result of a normal operation state.

[0081] In order to better monitor the voltage transformer in abnormal operation state, when the current execution end classifies the voltage data in abnormal operation state, specifically, firstly calculates the Minkowski distance of the voltage data based on the k-nearest neighbor algorithm KNN. At this time, after the Minkowski distance is calculated, it is compared with the calculated Minkowski distance based on the pre-configured distance threshold, and the abnormal voltage phase is determined based on the comparison result. In the embodiment of the present invention, it can be selected based on the size of the configured distance threshold whether to determine the voltage corresponding to the Minkowski distance greater than the distance threshold as the abnormal voltage phase, or to determine the voltage corresponding to the Minkowski distance less than or equal to the distance threshold as the abnormal voltage phase. In the embodiment of the present invention, the specific distance threshold configuration can be configured based on different high-voltage power transmission equipment or power transmission requirements, without specific limitation. Among them, the Minkowski distance formula is:

[0082]

[0083] Among them, d is the dimension and p is the norm.

[0084] In another embodiment of the present invention, for further definition and description, the steps further include:

[0085] When it is detected that the voltage sample data is updated, the data sliding window width is determined, and the original voltage sample data is selected based on the data sliding window width, and the step of re-executing the model training of the kernel principal component analysis model is combined with the new voltage sample data, so as to re-perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed the model training, and dynamically update the fault assessment standard threshold.

[0086] In order to improve the effectiveness of model training and the accuracy of state classification based on the principal component analysis model, and to achieve real-time detection, the current execution end can update the voltage sample data. Specifically, when the current execution end detects that there is updated data in the voltage sample data, the data sliding window width is first determined. At this time, the data sliding window width is used to adaptively adjust the data update according to the specified time, that is, the original voltage sample data is selected based on the data sliding window width, and the steps of model training of the kernel principal component analysis model are re-executed in combination with the new voltage sample data, so that the sliding window width can be adaptively adjusted according to the data, and the voltage standard threshold can be dynamically adjusted accordingly. This can not only ensure that the voltage monitoring process is real-time and effective, but also pay attention to more detailed voltage changes and detect abnormal voltage states.

[0087] In a specific embodiment, after the sliding window width L is calculated at time t, the updated data at time t is first added to the sample data, and then the old sample data is removed or added to adjust the sample data width to the same as L, and the adjusted sample data is used as the model training data at time t+1, such as Figure 2 The data update diagram is shown.

[0088] In another embodiment of the present invention, for further definition and explanation, the step of determining the data sliding window width includes:

[0089] The data sliding window width is calculated based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2.

[0090] In order to effectively determine the data sliding window width, and thus achieve the purpose of model training based on data update, the current execution end calculates the data sliding window width based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2. At this time, the method for calculating the data sliding window width L is:

[0091] Among them, Lw is the initial window width, Lmin and Lmax are the minimum and maximum widths; α, Jβ and γ are three functional parameters, which mainly adjust the sensitivity of L to process variables; ||Δa k |=||a k -ak-1 ||,||Δb k ||=||b k -b k-1 || is the mean vector difference and covariance matrix 2 norm; ||Δa k ||,||Δb k || is the change in the mean vector of the normal voltage of CVT and the norm of the covariance matrix 2. Among them, the data sliding window width is negatively correlated with the residual space statistic, that is, when the residual space statistic Q of the voltage transformer CVT changes rapidly, the data sliding window width L will be appropriately reduced; and when the residual space statistic Q of the voltage transformer CVT changes slowly, the data sliding window width L will be appropriately increased. The specific value of increase or decrease can be configured based on the amount of sample data, and the embodiment of the present invention does not make specific limitations.

[0092] The embodiment of the present invention provides another method for monitoring the operating status of a voltage transformer. Compared with the prior art, it can achieve the purpose of voltage transformer detection without power outages in continuous industrial production, meet the flexibility requirements for effectively using voltage transformers in different industrial productions, and greatly reduce the consumption of monitoring resources for the voltage transformer, avoiding missed detection and false detection of the operating status of the voltage transformer, thereby improving the effectiveness of monitoring the operating status of the voltage transformer.

[0093] Furthermore, as a response to the above Figure 1 In order to realize the method shown in the figure, the embodiment of the present invention provides a voltage transformer operating status monitoring device, such as Figure 3 As shown, the device comprises:

[0094] An acquisition module 21 is used to acquire voltage data of a voltage transformer, and perform principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training, to obtain an analysis result;

[0095] A first determination module 22, configured to determine a fault assessment standard threshold value through the residual space statistics obtained from the analysis result, and to determine an abnormal operation state of the voltage transformer when the voltage data matches the fault assessment standard threshold value;

[0096] The second determination module 23 is used to classify the voltage data of the abnormal operation state, obtain the abnormal voltage phase, and determine it as the monitoring result of the abnormal operation state of the voltage transformer.

[0097] Furthermore, the device also includes:

[0098] A processing module, used for acquiring voltage sample data of a voltage transformer and performing normalization processing on the voltage sample data;

[0099] The training module is used to perform model training on the kernel principal component analysis model based on the voltage sample data that has completed the normalization process, so as to complete the model training.

[0100] Furthermore, the training module is specifically used to perform high-dimensional space mapping on the voltage sample data that has completed standard processing based on a kernel function, and calculate the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors; perform principal component decomposition on the eigenvalues ​​and the unit orthogonal eigenvectors to determine the principal component space matrix and the residual space matrix; when the cumulative variance percentage of the principal component number is greater than a preset expected value, the training of the kernel principal component analysis model is completed.

[0101] Furthermore, the processing module is specifically used to complete the voltage sample data by interpolation and delete abnormal sample data in the voltage sample data; based on the voltage sample data with the abnormal sample data deleted, the voltage sample data is operated with the data average and the data standard deviation to complete the normalization processing of the voltage sample data.

[0102] Furthermore, the device also includes:

[0103] An execution module is used to determine the data sliding window width when it is detected that the voltage sample data is updated, and select the original voltage sample data based on the data sliding window width, and re-execute the step of training the kernel principal component analysis model in combination with the new voltage sample data, so as to re-perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed the model training, and dynamically update the fault assessment standard threshold.

[0104] Furthermore, the execution module is specifically used to calculate the data sliding window width based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2, wherein the data sliding window width is negatively correlated with the residual space statistic.

[0105] Furthermore, the second determination module is specifically configured to calculate the Minkowski distance of the voltage data of the abnormal operation state based on a k-nearest neighbor algorithm; and determine the abnormal voltage phase based on the Minkowski distance, so as to determine the abnormal voltage phase as a monitoring result of the normal operation state.

[0106] The embodiment of the present invention provides a device for monitoring the operating status of a voltage transformer. Compared with the prior art, the embodiment of the present invention obtains voltage data of a voltage transformer, and performs principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result; determines a fault assessment standard threshold by using a residual space statistic obtained from the analysis result, and determines the abnormal operating status of the voltage transformer when the voltage data matches the fault assessment standard threshold; classifies the voltage data of the abnormal operating status to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operating status of the voltage transformer, thereby achieving the purpose of voltage transformer detection without power outages in continuous industrial production, meeting the flexibility requirements for effectively using voltage transformers in different industrial productions, and greatly reducing the monitoring resource consumption of the voltage transformer, avoiding missed detection and false detection of the operating status of the voltage transformer, thereby improving the monitoring effectiveness of the operating status of the voltage transformer.

[0107] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the voltage transformer operating status monitoring method in any of the above method embodiments.

[0108] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0109] like Figure 4 As shown, the terminal may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .

[0110] The processor 302 , the communication interface 304 , and the memory 306 communicate with each other via the communication bus 308 .

[0111] The communication interface 304 is used to communicate with other devices such as clients or other servers.

[0112] The processor 302 is used to execute the program 310, and specifically can execute the relevant steps in the above-mentioned voltage transformer operating status monitoring method embodiment.

[0113] Specifically, the program 310 may include program codes, which include computer operation instructions.

[0114] The processor 302 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0115] The memory 306 is used to store the program 310. The memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0116] The program 310 may be specifically configured to enable the processor 302 to perform the following operations:

[0117] Acquire voltage data of the voltage transformer, and perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed model training to obtain an analysis result;

[0118] Determining a fault assessment standard threshold value through the residual space statistics obtained by the analysis result, and determining an abnormal operating state of the voltage transformer when the voltage data matches the fault assessment standard threshold value;

[0119] The voltage data of the abnormal operation state is classified to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operation state of the voltage transformer.

[0120] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0121] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring the operating status of a voltage transformer, characterized in that: include: Acquire voltage data of the voltage transformer, and perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed model training to obtain an analysis result; Determining a fault assessment standard threshold value through the residual space statistics obtained by the analysis result, and determining an abnormal operating state of the voltage transformer when the voltage data matches the fault assessment standard threshold value; The voltage data of the abnormal operation state is classified to obtain an abnormal voltage phase, which is determined as a monitoring result of the abnormal operation state of the voltage transformer.

2. The method according to claim 1, characterized in that Before performing principal component analysis on the voltage data based on the kernel principal component analysis model that has completed model training to obtain an analysis result, the method further includes: Acquiring voltage sample data of a voltage transformer, and performing normalization processing on the voltage sample data; The kernel principal component analysis model is trained based on the voltage sample data that has undergone normalization processing to complete the model training.

3. The method according to claim 1, characterized in that The performing model training on the kernel principal component analysis model based on the voltage sample data that has completed the normalization process, and completing the model training includes: Performing high-dimensional space mapping on the voltage sample data that has completed the standardization processing based on a kernel function, and calculating the eigenvalues ​​of the autocovariance matrix of the voltage sample data and the corresponding unit orthogonal eigenvectors; Performing principal component decomposition on the eigenvalues ​​and the unit orthogonal eigenvectors to determine a principal component space matrix and a residual space matrix; When the cumulative variance percentage of the principal component number is greater than a preset expected value, the training of the kernel principal component analysis model is completed.

4. The method according to claim 2, characterized in that: The normalizing process of the voltage sample data comprises: Performing data completion processing on the voltage sample data by interpolation, and deleting abnormal sample data in the voltage sample data; The voltage sample data after the abnormal sample data is deleted is operated with the data average value and the data standard deviation to complete the normalization processing of the voltage sample data.

5. The method according to claim 4, characterized in that The method further comprises: When it is detected that the voltage sample data is updated, the data sliding window width is determined, and the original voltage sample data is selected based on the data sliding window width, and the step of re-executing the model training of the kernel principal component analysis model is combined with the new voltage sample data, so as to re-perform principal component analysis on the voltage data based on the kernel principal component analysis model that has completed the model training, and dynamically update the fault assessment standard threshold.

6. The method according to claim 5, characterized in that Determining the data sliding window width includes: The data sliding window width is calculated based on the initial window width, variable sensitivity, mean vector difference, covariance matrix 2 norm, and covariance matrix 2, wherein the data sliding window width is negatively correlated with the residual space statistic.

7. The method according to claim 1, characterized in that The step of classifying the voltage data of the abnormal operation state to obtain the abnormal voltage phase and determining the monitoring result of the abnormal operation state of the voltage transformer includes: The Minkowski distance of the voltage data of the abnormal operation state is calculated based on a k-nearest neighbor algorithm; and an abnormal voltage phase is determined based on the Minkowski distance to determine the abnormal voltage phase as a monitoring result of a normal operation state.

8. A voltage transformer operating status monitoring device, characterized in that: include: An acquisition module is used to acquire voltage data of a voltage transformer, and perform principal component analysis on the voltage data based on a kernel principal component analysis model that has completed model training to obtain an analysis result; A first determination module, configured to determine a fault assessment standard threshold value through a residual space statistic obtained from the analysis result, and to determine an abnormal operation state of the voltage transformer when the voltage data matches the fault assessment standard threshold value; The second determination module is used to classify the voltage data of the abnormal operation state, obtain the abnormal voltage phase, and determine it as the monitoring result of the abnormal operation state of the voltage transformer.

9. A storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute an operation corresponding to the method for monitoring the operating status of a voltage transformer as described in any one of claims 1 to 7.

10. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the voltage transformer operating status monitoring method according to any one of claims 1-7.

Images