Power transmission line fault database construction method and device, terminal and storage medium

Abstract

The present application relates to power failure database construction technical field, especially in a kind of power transmission line fault database construction method, device, terminal and storage medium, the present application embodiment first obtains the abnormal data set of power transmission line and multiple observation data sets;Then abnormal data set and multiple observation data sets are carried out influence analysis and obtain multiple factor data sets;Then according to the characteristic data of multiple factor data sets and abnormal data set constructs fault label;Finally, fault label, multiple abnormal data sets and multiple factor data sets are associated and added to database.The present application embodiment extracts the features of observation data set by feature extraction queue, carries out influence analysis based on the extracted features on the formation of fault, can retain the observation data that has relatively obvious influence on the formation of fault, irrelevant data is deleted, can analyze the reason of the formation of fault by factor data, and data volume is smaller, will not mislead the person who consults database.

Description

Technical Field

[0001] This invention relates to the field of power fault database construction technology, and in particular to a method, apparatus, terminal and storage medium for constructing a transmission line fault database. Background Technology

[0002] Transmission line faults are closely related to the area, type, season, and climate of the transmission line. Common transmission line faults include wind-induced deviation faults, pollution flashover faults, and icing faults.

[0003] The causes, prevention, and maintenance methods of transmission line faults are of great significance to the design and construction of transmission lines. Therefore, it is necessary to build a database of transmission line faults to provide a reference for the renovation and maintenance of existing transmission lines.

[0004] Because transmission line faults are related to a variety of potential factors, it is often difficult to determine the correlation between influencing factors and transmission line faults when building a database. When the true cause of the fault cannot be provided, the database data can mislead technical personnel. Furthermore, due to the wide variety of fault scenarios, database construction must also address the issue of how to classify faults for easy retrieval.

[0005] Therefore, it is necessary to develop a method for constructing a transmission line fault database. Summary of the Invention

[0006] The present invention provides a method, apparatus, terminal and storage medium for constructing a transmission line fault database, which is used to solve the problem of poor accuracy of fault causes in the prior art.

[0007] In a first aspect, embodiments of the present invention provide a method for constructing a transmission line fault database, comprising:

[0008] Obtain an anomaly dataset of the transmission line and multiple observation datasets corresponding to the anomaly dataset, wherein the anomaly dataset includes sections of the transmission line, fault data of the transmission line, and anomaly data before the fault occurred.

[0009] An impact analysis was performed on the abnormal dataset and the multiple observation datasets, and irrelevant observation datasets were removed to obtain multiple factor datasets.

[0010] Based on the feature data of the multiple factor datasets and the anomaly dataset, fault labels are constructed;

[0011] The fault labels, the multiple abnormal datasets, and the multiple factor datasets are associated and added to the fault database of the section to which the transmission line belongs.

[0012] In one possible implementation, the impact analysis of the abnormal dataset and the multiple observation datasets is performed, and irrelevant observation datasets are removed to obtain multiple factor datasets, including:

[0013] Obtain multiple data feature extraction queues;

[0014] For each of the multiple observation datasets, perform the following steps:

[0015] Multiple feature vectors of the dataset are extracted using the multiple data feature extraction queues;

[0016] The impact analysis of the abnormal datasets in the corresponding datasets is performed using the multiple feature vectors to obtain multiple impact coefficients;

[0017] Whether to retain the observation dataset is determined based on the multiple influence coefficients.

[0018] In one possible implementation, the data feature extraction queue includes at least one of the following: a difference queue, an accumulation queue, and a fuzzy queue, wherein extracting multiple feature vectors from the dataset using the multiple data feature extraction queues includes:

[0019] For each data feature extraction queue, perform the following steps:

[0020] Initialize the extraction location;

[0021] Data extraction steps: Extract multiple data points from the dataset based on the extraction location, the number of which is the same as the number of data feature queues.

[0022] Treat the multiple data points as vectors and calculate the vector product of the multiple data points and the data feature queue;

[0023] Add the vector product to the feature vector;

[0024] If the extraction position does not reach the last position of the dataset, the extraction position is moved according to the preset movement step and the process jumps to the data extraction step.

[0025] In one possible implementation, the impact analysis of the abnormal datasets in the corresponding datasets is performed using the multiple feature vectors to obtain multiple impact coefficients, including:

[0026] For each of the plurality of feature vectors, perform the following steps:

[0027] The influence coefficient is calculated based on the first formula, the feature vector, and the anomaly dataset of the corresponding dataset, where the first formula is:

[0028]

[0029] In the formula, INF is the influence coefficient, feat(Nn) is the (Nn)th element of the feature vector, EXC(Mn) is the (Mn)th element of the abnormal dataset, the feature vector contains N+1 elements, and the abnormal dataset contains M+1 data.

[0030] In one possible implementation, constructing fault labels based on the feature data of the plurality of factor datasets and the anomaly dataset includes:

[0031] Based on the aforementioned multiple factor datasets, multiple feature vectors are obtained;

[0032] Pooling is performed on the abnormal dataset and the multiple feature vectors to obtain abnormal feature values ​​and multiple factor feature values ​​of the multiple feature vectors;

[0033] A fault label vector is constructed based on the abnormal feature values ​​and the multiple factor feature values;

[0034] The fault label vector is added to a label class, wherein the label class includes multiple label vectors, and the label class is obtained based on the similarity clustering of the label vectors;

[0035] The label of a class is determined based on the class center of the label class.

[0036] In one possible implementation, the label class is obtained based on similarity clustering of label vectors, including:

[0037] Obtain multiple label vectors and the separation threshold;

[0038] Steps for selecting the vector to be classified: Randomly select one label vector from the multiple label vectors that have not been clustered as the vector to be classified;

[0039] Target vector finding steps: Find the target vector among multiple unclustered label vectors, wherein the similarity difference between the target vector and the vector to be classified is less than the separation threshold.

[0040] If a target vector is found, the target vector is added to the class to which the vector to be classified belongs, the target vector is used as the vector to be classified, and the process jumps to the target vector search step.

[0041] Otherwise, proceed to the step of selecting the vector to be classified.

[0042] In one possible implementation, the process of finding the target vector among multiple unclustered label vectors includes:

[0043] The target vector is found from the multiple label vectors that have not been clustered, according to the second formula, where the second formula is:

[0044]

[0045] In the formula, SIM represents the similarity score, LBL(j) is the j-th element of the unclustered label vector, and LBL... tbs (j) is the j-th element of the vector to be classified, SPC TH J is the separation threshold, and J is the total number of elements in the vector to be classified.

[0046] Secondly, embodiments of the present invention provide a transmission line fault database construction apparatus for implementing the transmission line fault database construction method as described in the first aspect or any possible implementation thereof, the transmission line fault database construction apparatus comprising:

[0047] The observation data acquisition module is used to acquire the abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset, wherein the abnormal dataset includes the section of the transmission line, the fault data of the transmission line, and the abnormal data before the fault occurred.

[0048] The data reduction module is used to perform impact analysis on the abnormal dataset and the multiple observation datasets, and to reduce irrelevant observation datasets to obtain multiple factor datasets.

[0049] The fault label construction module is used to construct fault labels based on the feature data of the multiple factor datasets and the abnormal datasets;

[0050] as well as,

[0051] The database construction module is used to associate the fault labels, the multiple abnormal datasets, and the multiple factor datasets and add them to the fault database of the section to which the transmission line belongs.

[0052] Thirdly, embodiments of the present invention provide a terminal, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the steps of the method as described in the first aspect or any possible implementation of the first aspect.

[0053] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any possible implementation thereof.

[0054] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:

[0055] This invention discloses an implementation method for constructing a transmission line fault database. First, it acquires an abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset. The abnormal dataset includes the transmission line section, fault data of the transmission line, and abnormal data prior to fault formation. Then, it performs an impact analysis on the abnormal dataset and the multiple observation datasets, removing irrelevant observation datasets to obtain multiple factor datasets. Next, it constructs fault labels based on the feature data of the multiple factor datasets and the abnormal datasets. Finally, it associates the fault labels, the multiple abnormal datasets, and the multiple factor datasets and adds them to the fault database of the section to which the transmission line belongs. This invention extracts features from the observation datasets using a feature extraction queue, and then performs an impact analysis on fault formation based on the extracted features. This retains observation data that has a significant impact on fault formation while deleting irrelevant data. Therefore, the causes of fault formation can be analyzed through factor data, and the data volume is relatively small, avoiding misleading those consulting the database. This invention constructs label vectors based on features and categorizes fault data using a classification method, facilitating the retrieval of faults and their causes, resulting in higher retrieval efficiency. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is a flowchart of the method for constructing a transmission line fault database provided in the embodiments of the present invention;

[0058] Figure 2 This is a flowchart of extracting feature vectors from a dataset using a data feature extraction queue, provided by an embodiment of the present invention.

[0059] Figure 3 This is a functional block diagram of the power transmission line fault database construction device provided in the embodiments of the present invention;

[0060] Figure 4 This is a terminal function block diagram provided by an embodiment of the present invention. Detailed Implementation

[0061] In the following description, specific details such as particular system structures and techniques are set forth for illustrative purposes and not for limitation, so as to provide a thorough understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0062] To make the objectives, technical solutions, and advantages of the present invention clearer, specific embodiments will be described below in conjunction with the accompanying drawings.

[0063] The embodiments of the present invention will be described in detail below. This example is implemented based on the technical solution of the present invention, and provides detailed implementation methods and specific operation processes. However, the protection scope of the present invention is not limited to the following embodiments.

[0064] Figure 1 A flowchart illustrating the method for constructing a power transmission line fault database as provided in an embodiment of the present invention.

[0065] like Figure 1 As shown, a flowchart illustrating the implementation of the transmission line fault database construction method provided by the embodiments of the present invention is illustrated below:

[0066] In step 101, an abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset are obtained. The abnormal dataset includes sections of the transmission line, fault data of the transmission line, and abnormal data before the fault occurred.

[0067] In step 102, an impact analysis is performed on the abnormal dataset and the multiple observation datasets, and irrelevant observation datasets are removed to obtain multiple factor datasets.

[0068] In some embodiments, step 102 includes:

[0069] Obtain multiple data feature extraction queues;

[0070] For each of the multiple observation datasets, perform the following steps:

[0071] Multiple feature vectors of the dataset are extracted using the multiple data feature extraction queues;

[0072] The impact analysis of the abnormal datasets in the corresponding datasets is performed using the multiple feature vectors to obtain multiple impact coefficients;

[0073] Whether to retain the observation dataset is determined based on the multiple influence coefficients.

[0074] In some implementations, the data feature extraction queue includes at least one of the following: a difference queue, an accumulation queue, and a fuzzy queue, wherein extracting multiple feature vectors from the dataset using the multiple data feature extraction queues includes:

[0075] For each data feature extraction queue, perform the following steps:

[0076] Initialize the extraction location;

[0077] Data extraction steps: Extract multiple data points from the dataset based on the extraction location, the number of which is the same as the number of data feature queues.

[0078] Treat the multiple data points as vectors and calculate the vector product of the multiple data points and the data feature queue;

[0079] Add the vector product to the feature vector;

[0080] If the extraction position does not reach the last position of the dataset, the extraction position is moved according to the preset movement step and the process jumps to the data extraction step.

[0081] In some implementations, the step of using the multiple feature vectors to perform impact analysis on the abnormal datasets of the corresponding datasets to obtain multiple impact coefficients includes:

[0082] For each of the plurality of feature vectors, perform the following steps:

[0083] The influence coefficient is calculated based on the first formula, the feature vector, and the anomaly dataset of the corresponding dataset, where the first formula is:

[0084]

[0085] In the formula, INF is the influence coefficient, feat(Nn) is the (Nn)th element of the feature vector, EXC(Mn) is the (Mn)th element of the abnormal dataset, the feature vector contains N+1 elements, and the abnormal dataset contains M+1 data.

[0086] For example, the anomaly dataset includes the state of the transmission line before the fault occurred and the state of the fault itself. In some implementations, this means acquiring data on the fault location during the time period in which the fault occurred. Correspondingly, the observation dataset is based on observation data acquired from multiple observation points within that time period. For instance, if the line fault is a wind deflection fault occurring in the time period N to N+1, then the abnormal wind deflection conditions of the line at multiple time points within this time period should be acquired. Furthermore, wind speed information at multiple time points within this time period should also be acquired as observation data.

[0087] The examples above only illustrate some faults whose causes are relatively easy to determine. In reality, some faults are related to multiple factors, and these factors are not obvious. This invention proposes to conduct impact analysis based on observed and abnormal data to identify the observed data that affect the fault, thereby determining the cause of the fault. Furthermore, irrelevant data is removed, retaining only the relevant data as factor data.

[0088] The observation dataset actually contains a variety of information. Analyzing it from multiple dimensions can determine the impact of the observation dataset on fault formation. One way to extract information is by using feature extraction queues, such as difference queues, accumulation queues, or fuzzy queues, which can extract difference features, accumulation features, or fuzzy features, respectively. (-1,1) is a difference queue. When two data points in the observation dataset are multiplied by the difference queue, the difference in the observation dataset is obtained, representing the change of the observation data over time. In fact, the difference operation for the entire dataset is performed by extracting two data points through vector product, then shifting them by a certain unit, and then performing vector product again until all the data in the dataset has been extracted. Figure 2 As shown in the figure, the operation of this vector product is as follows: First, the feature extraction queue 201 is used to perform a vector product operation with the data at the first position 202 of the dataset to obtain the first data 203, which is then added to the feature vector. Then, the first position 202 is shifted to the second position 204, and the vector product operation is performed again to obtain the second data 205, which is then added to the feature vector again. This process is repeated until the end of the dataset is reached. Similarly, (1,1,1) is a cumulative queue, and the vector product operation extracts the cumulative sum of the three data points. (1 / 3,1 / 3,1 / 3) is a fuzzy queue, and the vector product operation extracts the fuzzy data (the average of the three elements) of the three data points.

[0089] We can see that by using different data feature extraction queues, different features can be extracted. By performing impact analysis on these features, we can determine whether the observed data influenced the formation of the fault. One method of impact analysis is to calculate the impact coefficient using the first formula, which is:

[0090]

[0091] In the formula, INF is the influence coefficient, feat(Nn) is the (Nn)th element of the feature vector, EXC(Mn) is the (Mn)th element of the abnormal dataset, the feature vector contains N+1 elements, and the abnormal dataset contains M+1 data.

[0092] The calculated influence coefficient using this formula is in the range of -1 to 1. The closer it is to -1, the stronger the negative correlation with the cause of the fault. The closer it is to 1, the stronger the positive correlation with the cause of the fault. We should retain the observation datasets whose absolute values ​​are greater than the influence threshold and use these datasets as factor datasets.

[0093] In step 103, fault labels are constructed based on the feature data of the multiple factor datasets and the anomaly dataset.

[0094] In some embodiments, step 103 includes:

[0095] Based on the aforementioned multiple factor datasets, multiple feature vectors are obtained;

[0096] Pooling is performed on the abnormal dataset and the multiple feature vectors to obtain abnormal feature values ​​and multiple factor feature values ​​of the multiple feature vectors;

[0097] A fault label vector is constructed based on the abnormal feature values ​​and the multiple factor feature values;

[0098] The fault label vector is added to a label class, wherein the label class includes multiple label vectors, and the label class is obtained based on the similarity clustering of the label vectors;

[0099] The label of a class is determined based on the class center of the label class.

[0100] In some implementations, the label class is obtained based on similarity clustering of label vectors, including:

[0101] Obtain multiple label vectors and the separation threshold;

[0102] Steps for selecting the vector to be classified: Randomly select one label vector from the multiple label vectors that have not been clustered as the vector to be classified;

[0103] Target vector finding steps: Find the target vector among multiple unclustered label vectors, wherein the similarity difference between the target vector and the vector to be classified is less than the separation threshold.

[0104] If a target vector is found, the target vector is added to the class to which the vector to be classified belongs, the target vector is used as the vector to be classified, and the process jumps to the target vector search step.

[0105] Otherwise, proceed to the step of selecting the vector to be classified.

[0106] In some implementations, finding the target vector among multiple unclustered label vectors includes:

[0107] The target vector is found from the multiple label vectors that have not been clustered, according to the second formula, where the second formula is:

[0108]

[0109] In the formula, SIM represents the similarity score, LBL(j) is the j-th element of the unclustered label vector, and LBL... tbs (j) is the j-th element of the vector to be classified, SPC TH J is the separation threshold, and J is the total number of elements in the vector to be classified.

[0110] For example, after identifying the factors influencing fault formation, it is necessary to construct labels for the combination of factor data and abnormal data to facilitate searching and classification. The feature vectors obtained during factor data analysis in the aforementioned steps are the basis for labeling. Since there is a large amount of feature vector data, pooling operations are generally required. For example, max pooling extracts the largest element in the feature vector, while mean pooling extracts the average of multiple elements in the feature vector. After the above steps, multiple feature values ​​corresponding to multiple abnormal feature vectors and abnormal feature values ​​corresponding to the abnormal dataset can be extracted. These feature values ​​are arranged in a predetermined order to obtain fault label vectors, which are then classified into label classes, and labels are determined based on the label classes.

[0111] In practice, label classes are obtained by clustering multiple label vectors. Each label vector in a class can always find a label whose distance is less than the separating threshold. In one implementation, the determination of multiple vectors in a class mainly relies on the second formula:

[0112]

[0113] In the formula, SIM represents the similarity score, LBL(j) is the j-th element of the unclustered label vector, and LBL... tbs (j) is the j-th element of the vector to be classified, SPC TH J is the separation threshold, and J is the total number of elements in the vector to be classified.

[0114] In fact, the clustering process is similar to a ring toss. Starting from a point, we find points that are no more than a threshold distance from the starting point. If we find such a point, we use it as the starting point and continue searching using the threshold until no new points are found. This means that the clustering of that class is complete. Then we proceed to the next class of clustering until all vectors are clustered.

[0115] Once our fault label vectors are categorized, we can select the label of the vector at the class center (the vector with the lowest similarity to other vectors in the class) as the label to identify the abnormal dataset and the factor dataset.

[0116] In step 104, the fault labels, the multiple abnormal datasets, and the multiple factor datasets are associated and added to the fault database of the section to which the transmission line belongs.

[0117] For example, the fault label identifiers for the anomaly dataset and factor dataset are then added to the fault database of the section to which the transmission line belongs. When a search is performed, data for that type of fault can be retrieved using the fault label.

[0118] This invention discloses an implementation method for constructing a transmission line fault database. First, it acquires an abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset. The abnormal dataset includes the transmission line section, fault data of the transmission line, and abnormal data prior to fault formation. Then, it performs an impact analysis on the abnormal dataset and the multiple observation datasets, removing irrelevant observation datasets to obtain multiple factor datasets. Next, it constructs fault labels based on the feature data of the multiple factor datasets and the abnormal datasets. Finally, it associates the fault labels, the multiple abnormal datasets, and the multiple factor datasets and adds them to the fault database of the section to which the transmission line belongs. This invention extracts features from the observation datasets using a feature extraction queue, and then performs an impact analysis on fault formation based on the extracted features. This retains observation data that has a significant impact on fault formation while deleting irrelevant data. Therefore, the causes of fault formation can be analyzed through factor data, and the data volume is relatively small, avoiding misleading those consulting the database. This invention constructs label vectors based on features and categorizes fault data using a classification method, facilitating the retrieval of faults and their causes, resulting in higher retrieval efficiency.

[0119] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0120] The following are embodiments of the apparatus of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0121] Figure 3 This is a functional block diagram of the transmission line fault database construction device provided in the embodiments of the present invention, with reference to... Figure 3 The transmission line fault database construction device 3 includes: an observation data acquisition module 301, a data deletion module 303, a fault tag construction module 303, and a database construction module 304, wherein:

[0122] The observation data acquisition module 301 is used to acquire the abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset, wherein the abnormal dataset includes the section of the transmission line, the fault data of the transmission line, and the abnormal data before the fault occurred.

[0123] The data reduction module 303 is used to perform an impact analysis on the abnormal dataset and the multiple observation datasets, and to reduce irrelevant observation datasets to obtain multiple factor datasets.

[0124] The fault label construction module 303 is used to construct fault labels based on the feature data of the multiple factor datasets and the abnormal dataset;

[0125] The database construction module 304 is used to associate the fault labels, the multiple abnormal datasets, and the multiple factor datasets and add them to the fault database of the section to which the transmission line belongs.

[0126] Figure 4 This is a functional block diagram of the terminal provided in an embodiment of the present invention. For example... Figure 4 As shown, the terminal 4 in this embodiment includes a processor 400 and a memory 401, wherein the memory 401 stores a computer program 402 that can run on the processor 400. When the processor 400 executes the computer program 402, it implements the steps in the above-described methods and embodiments for constructing a transmission line fault database, for example... Figure 1 Steps 101 to 104 are shown.

[0127] For example, the computer program 402 may be divided into one or more modules / units, which are stored in the memory 401 and executed by the processor 400 to complete the present invention.

[0128] The terminal 4 can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The terminal 4 may include, but is not limited to, a processor 400 and a memory 401. Those skilled in the art will understand that... Figure 4 This is merely an example of terminal 4 and does not constitute a limitation on terminal 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, terminal 4 may also include input / output devices, network access devices, buses, etc.

[0129] The processor 400 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0130] The memory 401 can be an internal storage unit of the terminal 4, such as a hard disk or memory of the terminal 4. The memory 401 can also be an external storage device of the terminal 4, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the terminal 4. Furthermore, the memory 401 can include both internal storage units and external storage devices of the terminal 4. The memory 401 is used to store the computer program 402 and other programs and data required by the terminal 4. The memory 401 can also be used to temporarily store data that has been output or will be output.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the aforementioned method embodiments, and will not be repeated here.

[0132] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0133] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0134] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0135] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0136] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0137] If the integrated module / unit is implemented as 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, all or part of the processes in the above-described embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various methods and apparatus embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0138] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for constructing a transmission line fault database, characterized in that, include: Obtain an anomaly dataset of the transmission line and multiple observation datasets corresponding to the anomaly dataset, wherein the anomaly dataset includes sections of the transmission line, fault data of the transmission line, and anomaly data before the fault occurred. An impact analysis was performed on the anomalous dataset and the multiple observation datasets, and irrelevant observation datasets were removed to obtain multiple factor datasets, including: Obtain multiple data feature extraction queues; For each of the multiple observation datasets, perform the following steps: Multiple feature vectors of the dataset are extracted using the multiple data feature extraction queues; The impact analysis of the abnormal datasets in the corresponding datasets is performed using the multiple feature vectors to obtain multiple impact coefficients; Determine whether to retain the observation dataset based on the multiple influence coefficients; Based on the feature data of the multiple factor datasets and the anomaly dataset, fault labels are constructed, including: Based on the aforementioned multiple factor datasets, multiple feature vectors are obtained; Pooling is performed on the abnormal dataset and the multiple feature vectors to obtain abnormal feature values ​​and multiple factor feature values ​​of the multiple feature vectors; A fault label vector is constructed based on the abnormal feature values ​​and the multiple factor feature values; The fault label vector is added to a label class, wherein the label class includes multiple label vectors, and the label class is obtained based on the similarity clustering of the label vectors; The label of the class is determined based on the class center of the label class; The fault labels, the multiple abnormal datasets, and the multiple factor datasets are associated and added to the fault database of the section to which the transmission line belongs.

2. The method for constructing a transmission line fault database according to claim 1, characterized in that, The data feature extraction queue includes at least one of the following: a difference queue, an accumulation queue, and a fuzzy queue. The step of extracting multiple feature vectors from the dataset using the multiple data feature extraction queues includes: For each data feature extraction queue, perform the following steps: Initialize the extraction location; Data extraction steps: Extract multiple data points from the dataset based on the extraction location, the number of which is the same as the number of data feature queues. Treat the multiple data points as vectors and calculate the vector product of the multiple data points and the data feature queue; Add the vector product to the feature vector; If the extraction position does not reach the last position of the dataset, the extraction position is moved according to the preset movement step and the process jumps to the data extraction step.

3. The method for constructing a transmission line fault database according to claim 1, characterized in that, The method involves using the multiple feature vectors to perform impact analysis on the abnormal datasets of the corresponding datasets, obtaining multiple impact coefficients, including: For each of the plurality of feature vectors, perform the following steps: The influence coefficient is calculated based on the first formula, the feature vector, and the anomaly dataset of the corresponding dataset, where the first formula is: In the formula, The influence coefficient, The eigenvector of the eigenvector One element, For the first abnormal dataset The feature vector contains elements. The abnormal dataset contains [number] elements. Data.

4. The method for constructing a transmission line fault database according to claim 1, characterized in that, The label classes are obtained based on the similarity clustering of label vectors, including: Obtain multiple label vectors and the separation threshold; Steps for selecting the vector to be classified: Randomly select one label vector from the multiple label vectors that have not been clustered as the vector to be classified; Target vector finding steps: Find the target vector among multiple unclustered label vectors, wherein the similarity difference between the target vector and the vector to be classified is less than the separation threshold. If a target vector is found, the target vector is added to the class to which the vector to be classified belongs, the target vector is used as the vector to be classified, and the process jumps to the target vector search step. Otherwise, proceed to the step of selecting the vector to be classified.

5. The method for constructing a transmission line fault database according to claim 4, characterized in that, The process of finding the target vector from multiple unclustered label vectors includes: The target vector is found from the multiple label vectors that have not been clustered, according to the second formula, where the second formula is: In the formula, For similarity, The first unclustered label vector One element, The first vector to be classified One element, The threshold for separation. This represents the total number of elements in the vector to be classified.

6. A device for constructing a transmission line fault database, characterized in that, For implementing the transmission line fault database construction method as described in any one of claims 1-5, the transmission line fault database construction apparatus includes: The observation data acquisition module is used to acquire the abnormal dataset of the transmission line and multiple observation datasets corresponding to the abnormal dataset, wherein the abnormal dataset includes the section of the transmission line, the fault data of the transmission line, and the abnormal data before the fault occurred. The data reduction module is used to perform impact analysis on the abnormal dataset and the multiple observation datasets, and to reduce irrelevant observation datasets to obtain multiple factor datasets. The fault label construction module is used to construct fault labels based on the feature data of the multiple factor datasets and the abnormal datasets; as well as, The database construction module is used to associate the fault labels, the multiple abnormal datasets, and the multiple factor datasets and add them to the fault database of the section to which the transmission line belongs.

7. A terminal comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 5 above.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 5 above.

Images