WO2020125166A1

WO2020125166A1 - Power grid anticipated fault set prediction method and apparatus, and electronic device and storage medium

Info

Publication number: WO2020125166A1
Application number: PCT/CN2019/110941
Authority: WO
Inventors: 张越; 张佳楠; 李如意; 单连飞; 吕宏伟; 余建明; 刘艳; 卓峻峰; 张连超
Original assignee: 北京科东电力控制系统有限责任公司
Priority date: 2018-12-20
Filing date: 2019-10-14
Publication date: 2020-06-25
Also published as: CN109711450A

Abstract

A power grid anticipated fault set prediction method and apparatus, and an electronic device and a storage medium. The method comprises: obtaining a target section characteristic variable set of a preset characteristic variable type of a target power grid section (S101); determining a sample characteristic variable set matching the target section characteristic variable set in a preset sample library (S102), wherein the sample library comprises a correspondence between a sample fault set and a sample characteristic variable set, the characteristic variable type of the sample characteristic variable set is the same as the preset characteristic variable type; and finally determining that the sample fault set corresponding to the matched sample characteristic variable set is a predicted fault set of the target power grid section (S103). The fault set corresponding to the current section is obtained by comparing a characteristic variable of a sample in the sample library with a characteristic variable of the target power grid section; a problem that the current section depends on manual experience is solved; the possibility of incomplete prediction of the fault set is reduced; the ultra-real-time prediction of the fault set is implemented; and the working efficiency of a worker is improved.

Description

Method, device, electronic equipment and storage medium for predicting fault set of power grid

Technical field

The present invention relates to the technical field of power grid analysis, and in particular, to a method, device, electronic equipment, and storage medium for predicting a fault set of a power grid.

Background technique

The predicted fault set generation technology in static safety analysis is a necessary means to prevent power grid faults and simulation drills; at present, the predicted fault set generation mainly depends on the experience of the dispatcher to determine, but this method takes a long time and is blind. A large number of actual faults are missed. When the operation mode of the power grid is changed and new equipment is connected, it is necessary to manually adjust the expected fault set. The fault set cannot be adaptively added, deleted, changed, or checked, nor can the influence of bad weather on the power grid be considered. Therefore, there is an urgent need to quickly and accurately predict the fault set generation method, get rid of the dependence of the expected fault set generation on the dispatcher's experience, and improve work efficiency.

Summary of the invention

In order to overcome the above deficiencies in the prior art, the present invention provides a method for automatically generating a fault set, an electronic device and a storage medium to improve the above problems.

In order to achieve the above objective, the technical solutions provided by the embodiments of the present invention are as follows:

In a first aspect, an embodiment of the present invention provides a method for predicting a predicted fault set of a power grid, including: acquiring a target cross-section feature variable set of a preset feature variable type of a target power grid cross-section; and determining the target in a preset sample library A sample feature variable set matched with a cross-section feature variable set; wherein, the sample library includes a correspondence between a sample fault set and a sample feature variable set; the feature variable type of the sample feature variable set is the same as the preset feature variable type; It is determined that the sample fault set corresponding to the matched sample feature variable set is the predicted fault set of the target power grid cross section.

With reference to the first aspect, in some possible implementation manners, before acquiring the target cross-section feature variable set of the preset feature variable type of the target grid cross-section, the method further includes: acquiring multiple grid variables as the initial feature variable set; A preset feature variable evaluation rule filters the initial feature variable set; it is determined that the type of feature variable included in the filtered initial feature variable set is the preset feature variable type.

With reference to the first aspect, in some possible implementation manners, the filtering of the initial feature variable set according to a preset feature variable evaluation rule includes: obtaining all feature variables of each sample feature variable set in the sample library; based on the Relief algorithm To determine the weight value of each feature variable in the initial feature variable set; delete the feature variables in the initial feature variable set whose weight value is lower than the preset threshold.

With reference to the first aspect, in some possible implementation manners, the method further includes: acquiring at least one initial sample library by at least one acquisition method, and combining at least one initial sample library to obtain a static sample library, and the initial sample library includes samples of each sample The fault feature set and the sample feature variable set corresponding to the sample fault set; the static sample library is filtered according to the preset sample evaluation rules; and the filtered static sample library is determined as the sample library.

With reference to the first aspect, in some possible implementations, before a sample feature variable set matching the target cross-section feature variable set is determined in the preset sample library, the method further includes: setting the target cross-section feature variable set and the sample library The initial evaluation value of the sample feature variable set in; update the initial evaluation value of the sample in the static sample library according to the result of the target cross-section feature variable set on the predicted fault set, where the prediction result includes: correct, wrong and irrelevant; Delete the sample feature variable set corresponding to the evaluation value lower than the preset evaluation threshold and the sample fault set corresponding to the sample feature variable set in the static sample library to obtain the sample library.

With reference to the first aspect, in some possible implementation manners, the determining a sample feature variable set that matches the target cross-section feature variable set in a preset sample library includes: comparing the target cross-section feature variable set Obtain the similarity between the target grid section and each sample in the sample library with the sample feature variable set in the sample library; determine the sample feature variable set with the highest similarity as the target section feature The sample feature variable set matching the variable set.

With reference to the first aspect, in some possible implementations, a sample feature variable set matching the target cross-section feature variable set is determined in a preset sample library, including: normalizing the target cross-section feature variable set and the sample feature variable Each feature variable in the set; calculate the classification distance between the target cross-section feature variable set and each sample feature variable set in the sample library based on the k-proximity algorithm; correspondingly, determine the sample feature variable set with the highest similarity as the target cross-section The sample feature variable set matched by the feature variable set includes: determining the sample feature variable set with the shortest classification distance as the sample feature variable set with the highest similarity.

In a second aspect, an embodiment of the present invention further provides a device for predicting a predicted set of power grid faults, including: a receiving module for acquiring a target profile feature variable set of a preset feature variable type of a target grid profile; a processing module for predicting A sample feature variable set matching the target cross-section feature variable set is determined from the set sample database, and a sample fault set corresponding to the matched sample feature variable set is determined as the predicted fault set of the target power grid cross-section; the sending module , Used to send the predicted fault set to the staff.

In a third aspect, an embodiment of the present invention further provides an electronic device, the device including:

A memory and a processor. The memory is used to store a fault prediction instruction, and the processor is used to run the fault prediction instruction to execute the power grid expected fault set prediction method described in any one of the implementation manners of the first aspect.

According to a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, where the computer program is stored in the readable storage medium, and when the computer program runs on a computer, the computer is executed as in the first aspect The method for predicting the expected fault set of the power grid described in any implementation.

The beneficial effects of the present invention include:

By obtaining the characteristics of the preset feature variable types of the target power grid section, and performing a comparative analysis according to the sample feature variable set of each sample in the preset sample library, the sample failure corresponding to the matched sample feature variable set in the sample library is obtained Set, and then predict the type of failure that occurs in the current grid section, instead of relying on the experience of dispatchers to analyze based on the grid variables of the current grid section and sample failures in the sample library, on the one hand, improve the efficiency of the staff, on the other hand The analysis of the current grid section is more complete.

In order to make the above objects, features and advantages of the present invention more obvious and understandable, the following describes the embodiments of the present invention and the accompanying drawings for detailed description.

BRIEF DESCRIPTION

In order to more clearly explain the technical solutions of the embodiments of the present invention, the drawings required in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and therefore do not It should be regarded as a limitation on the scope. For those of ordinary skill in the art, without paying any creative labor, other related drawings can be obtained based on these drawings.

FIG. 1 is a schematic flowchart of a method for predicting an expected fault set of a power grid according to an embodiment of the present invention;

FIG. 2 is a schematic flowchart of a process of optimizing and filtering to obtain preset feature variable types according to an embodiment of the present invention; FIG.

3 is a schematic flowchart of a method for generating a sample library according to an embodiment of the present invention;

4 is a functional block diagram of a device for predicting a fault set of a power grid according to an embodiment of the present invention.

detailed description

To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described clearly and completely in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of the embodiments of the present invention, but not all the embodiments. The components of the embodiments of the present invention generally described and illustrated in the drawings herein can be arranged and designed in various configurations.

Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but only to represent selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by a person of ordinary skill in the art without making creative efforts fall within the protection scope of the present invention.

It should be noted that the embodiments of the present invention and the features in the embodiments can be combined with each other without conflict.

It should be noted that similar reference numerals and letters indicate similar items in the following drawings, therefore, once an item is defined in one drawing, there is no need to further define and explain it in subsequent drawings.

In the description of the embodiments of the present invention, it should be noted that the indicated orientation or position relationship is based on the orientation or position relationship shown in the drawings, or the orientation or position relationship conventionally placed when the product of the invention is used, or this The orientation or positional relationship commonly understood by those skilled in the art, or the orientation or positional relationship normally placed when the product of the invention is used, is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element It has a specific orientation, is constructed and operated in a specific orientation, and therefore cannot be understood as a limitation of the present invention. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

In the description of the embodiments of the present invention, it should also be noted that, unless otherwise clearly specified and defined, the terms “setup”, “installation”, and “connection” should be understood in a broad sense, for example, it can be a fixed connection or a It is a detachable connection, or a one-piece connection; it can be a direct connection or an indirect connection through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood in specific situations.

This embodiment provides a method for automatically generating a predicted fault set of a power grid, which is used to predict a predicted fault set of a local power grid and verify the predicted expected fault set. According to the stored sample database and the prediction method, a fault set occurring in a certain power grid cross section is predicted. For details, please refer to FIG. 1. FIG. 1 is a schematic flowchart of a method for predicting a fault set of a power grid according to an embodiment of the present invention. The method includes steps S101-S103:

Step S101: Acquire a target profile feature variable set of a preset feature variable type of the target grid profile;

Wherein, the sample library includes the corresponding relationship between the sample fault set and the sample characteristic variable set, and the type of the characteristic variable of the sample characteristic variable set is the same as the type of the preset characteristic variable, and the characteristics of the target section characteristic variable set The variables include all variables in the power grid, such as bus voltage, branch power, etc., which is not limited in this embodiment.

Step S102: Determine a sample feature variable set matching the target cross-section feature variable set in a preset sample library;

It is understandable that the target cross-section feature variable set contains the same feature category as the sample feature variable set. The sample features are set in the preset sample library according to the preset matching rules. The variable set and the target cross-section feature variable set are compared to determine a sample feature variable set that matches the target cross-section feature variable set.

Step S103: Determine the sample fault set corresponding to the matched sample feature variable set as the predicted fault set of the target power grid cross section.

By using the sample library, the multiple target features of the target grid section and the sample features of each sample in the sample library are compared, and then according to the matching result, the sample fault set corresponding to the matched sample feature variable set in the sample library is determined to be the current section Predict the fault set; thus replacing the existing way of predicting the current grid section fault set through staff experience, on the one hand, it improves the efficiency of the staff, on the other hand, it can more comprehensively set the fault corresponding to the grid section Set to prevent the issue of major failures from being missed.

Alternatively, the grid variables may be optimized and screened to obtain the preset characteristic variable types, and the characteristic variables that can more accurately obtain the fault set can be selected and added to the preset characteristic variable types. Please refer to FIG. 2, which is a schematic flowchart of obtaining a preset feature variable type by optimized screening according to an embodiment of the present invention. The method for screening feature variable types includes:

Step S201: Acquire multiple grid variables as initial feature variable sets;

Among them, the multiple grid variables may include all grid variables that can be obtained or detected in the grid section, and then through screening, without worrying about the existence of many meaningless features, appropriate grid variables can be derived as the judgment characteristics.

Step S202: Filter the initial feature variable set according to a preset feature variable evaluation rule;

Wherein, the preset characteristic variable evaluation rule may adopt one or more of principal component analysis method, kernel principal component analysis method, partial least square method and Relief algorithm.

Optionally, based on the Relief algorithm, the method of acquiring the preset feature variable types includes: acquiring all feature variables of each sample feature variable set in the sample library; based on the Relief algorithm, determining the weight value of each feature variable in the initial feature variable set; Delete the feature variables in the initial feature variable set whose weight value is lower than the preset threshold.

An implementation method for obtaining the preset characteristic variable types is: setting the sample characteristic variable set in the sample library as (x _i , y _i ), i=1, 2, ..., N, and x _i are the grid section characteristics The variable, y _i is the sample fault set. Define the weight corresponding to the p-th feature variable as

Where N is the number of samples and x _i (p) is the p-th feature variable,

Is the same kind of sample with the closest Euclidean distance to x _i (p),

It is a heterogeneous sample with the closest Euclidean distance to x _i (p). The larger the feature variable weight w(p), the greater the contribution of the corresponding feature variable, and then use w(p)≥δ to filter the feature variable, and finally obtain the optimal feature variable that can characterize the sample feature variable set, Where δ is the threshold, in this example, the weight value of the feature variable with a large weight value and relatively stable is selected as the threshold value.

Calculate the weight value of each feature variable according to the above formula, and then delete the feature variable corresponding to the difference similarity in the initial feature variable set that is lower than the preset threshold according to a preset threshold to complete the initial feature For the selection of the variable set, in other embodiments, the principal component analysis method or the like may also be used to implement the selection of characteristic variables, which is not limited in this embodiment.

Step S203: Determine the filtered initial feature variable set as the determined feature variable set.

By selecting the grid variables that can be measured to better reflect the characteristics of the fault set, the influence of unrelated grid variables on the prediction of the target fault set can be reduced and the accuracy of the prediction can be improved.

Optionally, this embodiment also proposes a method for generating the sample library, which is used to collect existing fault set sample acquisition methods, and screen the samples to generate the sample library. Please refer to FIG. 3, which is a schematic flowchart of a method for generating the sample library according to this embodiment. The steps of the method for generating the sample library include steps S301-S303:

Step S301: At least one initial sample library is acquired by at least one collection method, and the at least one initial failure set sample set is combined to obtain a static sample library;

Among them, the at least one acquisition method includes: static safety analysis N-1, N-2 methods, typical expected fault sets, and a variety of power grid analysis methods combined with the experience of power grid dispatchers. In this embodiment, a variety of power grid analysis methods can be used Collection mode to more comprehensively cover the information in the grid operation mode, which is not limited in this embodiment; the initial sample library includes the initial sample failure set of each sample and the initial sample failure corresponding to the initial sample failure set Feature variable set.

Step S302: Filter the static sample library according to a preset sample evaluation rule;

Wherein, the preset sample evaluation rule is used to remove valueless samples in the static sample library, and screen out samples in the static sample library that are more capable of determining the predicted fault set. The preset sample evaluation rules include: a method based on credibility, a method based on the uniformity of the results of multiple classifiers, and a group-based incremental learning algorithm.

Optionally, an implementation method for screening the static sample library based on a group-based incremental learning algorithm is: setting an initial evaluation of the target cross-section feature variable set and the sample feature variable set in the sample library Value; and update the initial evaluation value of the sample in the static sample library according to the result of the target cross-section feature variable set on the predicted fault set, where the predicted result includes: correct, wrong, and irrelevant; And deleting the sample feature variable set corresponding to the evaluation value below the preset evaluation threshold and the sample fault set corresponding to the sample feature variable set in the static sample library, thereby obtaining the sample library.

Where the initial evaluation value of each sample before being screened is q ₀ , the learning rate parameter β is set, and 0<β<1, if sample i is one of the k nearest neighbors of the sample to be classified, and If the classification result is correct, the evaluation value of the sample is updated to: q _i1 =βq _i1 +1-β; if sample i is one of the k nearest neighbors of the sample to be classified, but the classification result is wrong, the evaluation value of the sample is updated Is: q _i2 = βq _i2 ; if sample i is not any of the k nearest samples of the sample to be classified, and the judgment relationship is not relevant, the evaluation value of the sample is updated to: q _i3 = βq _i3 + (1-β )q ₀ . It can be known that under the three relationships, the samples are updated, and when the threshold is set to q ₀ , the samples whose evaluation value is lower than q ₀ in the sample library are removed to reduce the valuelessness or reduce the prediction accuracy Of samples to improve prediction accuracy. In this example, k is 5, the learning rate parameter β is 0.9, and the initial evaluation value q ₀ is 0.5.

Step S303: Determine the screened static sample library as the sample library.

By filtering out the non-value samples in the static sample database, it is possible to prevent the non-value samples of this type from affecting the prediction fault set result of the current section determined in step S103, and improve the prediction accuracy.

Optionally, after the sample library is determined in step S303, the new sample failure set can be continuously added, and the upper limit of the sample quantity of the sample library can be set, and the sample can be eliminated according to the preset sample evaluation rule The most valuable samples in the library are used to continuously optimize the sample library. The method includes: acquiring and adding new samples to the sample library, and then obtaining the preset upper limit of the number of samples in the sample library; Whether the number of samples in the sample library is less than or equal to the upper limit of the number of samples, and if not, delete the samples in the sample library according to the preset sample evaluation rule until the number of samples in the sample library is less than or equal to the Maximum sample size.

Optionally, the new fault set sample may be automatically added to the sample library after each target grid section prediction is completed and the result of the prediction is correct or not, according to the preset sample evaluation rules in step S302 Screening to achieve dynamic optimization, the dispatcher may also input some new fault set samples based on experience, this embodiment does not limit this.

Optionally, the upper limit of the sample size can be set by setting multiple upper limit of the sample size. After multiple predictions, the prediction accuracy of each upper limit of sample size is compared. Finally, the upper limit of the sample size with the highest prediction accuracy is selected as the upper limit of the sample size Similarly, the dispatcher can also make a selection according to experience or according to the processing capacity of the application processor, which is not limited in this embodiment.

It can be further understood that multiple sample caps can be used to make simultaneous predictions. Based on the prediction results of the sample caps with multiple sample caps, such as selecting the most frequent fault set in the prediction results, this implementation Examples also do not limit this.

After obtaining the preset feature variable type and sample library, the step of determining the sample feature variable set matching the target cross-section feature variable set in step S102 may be performed.

Optionally, one way to determine the matching sample feature variable set is to compare the target cross-section feature variable set of the target power grid cross-section with the sample feature variable set in the sample library to obtain the target power grid cross-section The similarity between each sample in the sample library; determining that the sample fault set corresponding to the sample feature variable set with the highest similarity is the predicted fault set of the target power grid cross section. Among them, a neural network algorithm or deep learning algorithm can be used to compare each feature variable of the target cross-section feature variable set with each of the sample feature variable sets to obtain a sample feature variable set with the highest similarity It is the matching sample feature variable set.

Optionally, another implementation of determining the matching sample feature variable set may use the k-proximity algorithm, specifically: normalizing each feature variable in the target cross-section feature variable set and the sample feature variable set; and Calculate the classification distance between the target cross-section feature variable set and each sample feature variable set in the sample library based on the k-proximity algorithm; correspondingly, determine the sample feature variable set with the highest similarity as the sample feature variable matching the target cross-section feature variable set Sets include: determining the sample feature variable set with the shortest classification distance as the sample feature variable set with the highest similarity.

The specific steps of the above implementation are: set the sample feature variable set to (x _j , y _j ), where j=1, 2, ..., N, N is the number of samples, and x _j is the optimal cross-section of the power grid The characteristic variable, y _j is the sample fault set. First, normalize each feature variable of each sample, and the normalization formula is:

among them,

Is the value corresponding to the feature after normalization, x _j is the value corresponding to the feature before normalization, and max(x _j ) and min(x _j ) are the types of the feature variable The maximum and minimum values after conversion. Then, the k-proximity algorithm is used to calculate the classification distance between the target section feature variable set and each sample feature variable set in the sample library. The classification distance is calculated using the Euclidean distance. The formula is:

Where do _o,j is the classification distance between the target cross-section feature variable set and the jth sample, H is the number of feature variables of the preset feature variable type, x _on is the n-dimensional variable of the target feature variable set, x _jn is The nth dimension variable of the jth sample. By calculating the classification distance between the target cross-section feature variable set and each sample feature variable set in the sample library, the k sample feature variable sets closest to the target feature variable set in the sample library are obtained, and the k sample feature variable sets and the target are calculated The average value of the distance of the characteristic variable set

Determine the classification distance from k sample feature variable sets to the average

The sample fault set corresponding to the sample feature variable set with the smallest difference is used as the predicted fault set. Among them, k=5 in this example.

After determining the matched sample feature variable set, step S103 is executed, and according to the corresponding relationship between the sample fault set and the sample feature variable set in the sample library, the sample fault set corresponding to the matched sample feature variable set is determined as the predicted fault set, and completed Prediction of faults on the target power grid section.

Please refer to FIG. 4. FIG. 4 is a functional block diagram of a device 40 for predicting a grid fault set according to this embodiment. This embodiment also proposes a device 40 for predicting the expected set of power grid faults. The device includes: a receiving module 401, a processing module 402, and a sending module 403, wherein the receiving module 401 is used to obtain the preset characteristic variable types of the target power grid section The target cross-section feature variable set; the processing module 402 is used to determine a sample feature variable set matching the target cross-section feature variable set in a preset sample library, and determine a sample fault set corresponding to the matched sample feature variable set Is the predicted fault set of the target power grid section; the sending module 403 is used to send the predicted fault set to the staff.

Optionally, the receiving module 401 is also used to obtain multiple grid variables as an initial feature variable set; correspondingly, the processing module 402 is also used to filter the initial feature variable set according to a preset feature variable evaluation rule and determine the filtered The initial feature variable set is the preset feature variable type.

Optionally, the receiving module 401 is also used to obtain all feature variables of each sample feature variable set in the sample library; correspondingly, the processing module 402 is also used to determine the value of each feature variable in the initial feature variable set based on the Relief algorithm Weight value; delete the feature variables in the initial feature variable set whose weight value is lower than the preset threshold.

Optionally, the receiving module 401 is further used to obtain at least one initial sample library by using at least one collection method, and the processing module 402 is further used to combine the at least one initial sample set to obtain a static sample library and filter according to preset sample evaluation rules In the static sample library, it is determined that the filtered static sample library is the sample library.

Optionally, the processing module 402 is also used to set the initial evaluation value of the target cross-sectional feature variable set and the sample feature variable set in the sample library; and according to the results of the target cross-sectional feature variable set for predicting the fault set, the static sample library The initial evaluation value of the sample is updated, where the prediction results include: correct, wrong, and irrelevant; and delete the sample feature variable set corresponding to the evaluation value below the preset evaluation threshold in the static sample library, and the corresponding Sample failure set, thereby obtaining a sample library.

Optionally, the processing module 402 also compares the target cross-section feature variable set of the target power grid cross-section with the sample feature variable set in the sample library to obtain the target power grid cross-section and each sample in the sample library The similarity between them; determining that the sample fault set corresponding to the sample feature variable set with the highest similarity is the predicted fault set of the target power grid section.

Optionally, the processing module 402 is also used to normalize each feature variable in the target cross-section feature variable set and the sample feature variable set; and calculate the target cross-section feature variable set and each sample feature variable in the sample library based on the k-proximity algorithm The classification distance between the sets; correspondingly, the sample feature variable set with the highest similarity is determined as the sample feature variable set matching the target cross-section feature variable set, including: determining the sample feature variable set with the shortest classification distance as the sample with the highest similarity Feature variable set.

This embodiment also proposes an electronic device. The electronic device includes a memory and a processor. The memory is used to store a fault prediction instruction, and the processor is used to run the fault prediction instruction to perform the above-mentioned grid prediction fault set prediction method.

This embodiment also proposes a computer-readable storage medium that stores a computer program, and when the computer program runs on a computer, causes the computer to execute the power grid prediction described in this embodiment Fault set prediction method. This embodiment will not repeat them here.

In the several embodiments provided in this application, it should be understood that the disclosed device and method may also be implemented in other ways. The device embodiments described above are only schematic. For example, the flowcharts and block diagrams in the drawings show possible implementation architectures, functions, and functions of devices, methods, and computer program products according to multiple embodiments of the present invention. operating. In this regard, each block in the flowchart or block diagram may represent a module, program segment, or part of code that contains one or more of the Executable instructions. It should also be noted that in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two consecutive blocks can actually be executed substantially in parallel, and sometimes they can also be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented with a dedicated hardware-based system that performs specified functions or actions Or, it can be realized by a combination of dedicated hardware and computer instructions.

In addition, the functional modules in the embodiments of the present invention may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.

If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention essentially or part of the contribution to the existing technology or part of the technical solution can be embodied in the form of a software product, the computer software product is stored in a storage medium, including Several instructions are used to enable a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code . It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that these entities or operations There is any such actual relationship or order. Moreover, the terms "include", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device that includes a series of elements includes not only those elements, but also those not explicitly listed Or other elements that are inherent to this process, method, article, or equipment. Without more restrictions, the element defined by the sentence "include one..." does not exclude that there are other identical elements in the process, method, article or equipment that includes the element.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

A method for predicting an expected fault set of a power grid, characterized in that the method includes:

Obtain the target profile characteristic variable set of the preset characteristic variable type of the target grid section;

A sample feature variable set matching the target cross-section feature variable set is determined in a preset sample library; wherein, the sample library includes a correspondence between a sample fault set and a sample feature variable set, the sample feature variable set The type of characteristic variable is the same as the type of the preset characteristic variable;

It is determined that the sample fault set corresponding to the matched sample feature variable set is the predicted fault set of the target power grid cross section.
The method for predicting a predicted fault set of a power grid according to claim 1, wherein before acquiring the target cross-sectional feature variable set of the preset feature variable type of the target power grid cross-section, the method further comprises:

Obtain multiple grid variables as initial feature variable sets;

Filtering the initial feature variable set according to preset feature variable evaluation rules;

It is determined that the type of the characteristic variable included in the initial characteristic variable set after screening is the preset characteristic variable type.
The method for predicting an expected fault set of a power grid according to claim 2, wherein the filtering of the initial feature variable set according to a preset feature variable evaluation rule includes:

Acquiring all feature variables of each sample feature variable set in the sample library;

Based on the Relief algorithm, determine the weight value of each of the feature variables in the initial feature variable set;

Delete the feature variable in the initial feature variable set whose weight value is lower than a preset threshold.
The method for predicting an expected fault set of a power grid according to claim 1, wherein before the sample feature variable set matching the target cross-section feature variable set is determined in the preset sample library, the method further include:

At least one initial sample library is acquired by at least one acquisition method, and the at least one initial sample library is combined to obtain a static sample library, the initial sample library includes a sample failure set of each sample and a sample feature variable corresponding to the sample failure set set;

Filter the static sample library according to preset sample evaluation rules;

It is determined that the screened static sample library is the sample library.
The method for predicting a fault set of a power grid according to claim 4, wherein the filtering the static sample library according to a preset sample evaluation rule includes:

Setting an initial evaluation value of the target cross-sectional feature variable set and the sample feature variable set in the sample library;

Update the initial evaluation value of the sample in the static sample library according to the result of the target cross-section feature variable set on the predicted fault set, where the predicted result includes: correct, wrong, and irrelevant;

Delete the sample feature variable set corresponding to the evaluation value below the preset evaluation threshold and the sample fault set corresponding to the sample feature variable set in the static sample library to obtain the sample library.
The method for predicting a predicted fault set of a power grid according to claim 1, wherein the determining of a sample feature variable set matching the target cross-section feature variable set in a preset sample library includes:

Comparing the feature variable set of the target cross-section with the feature variable set of the sample in the sample library to obtain the similarity between the target grid cross-section and each sample in the sample library;

It is determined that the sample feature variable set with the highest similarity is a sample feature variable set that matches the target cross-sectional feature variable set.
The method for predicting a fault set of a power grid according to claim 6, wherein the comparison between the target cross-section feature variable set and the sample feature variable set in the sample library obtains the target power grid cross-section and the The similarity between samples in the sample library, including:

Normalize each feature variable in the target cross-section feature variable set and the sample feature variable set;

And calculate the classification distance between the target cross-section feature variable set and each sample feature variable set in the sample library based on the k-proximity algorithm;

Correspondingly, the sample feature variable set that determines the highest similarity is the sample feature variable set matching the target cross-section feature variable set, including:

It is determined that the sample feature variable set with the shortest classification distance is the sample feature variable set with the highest similarity.
A power grid prediction fault set prediction device, which is characterized by comprising:

The receiving module is used to obtain the target profile feature variable set of the preset feature variable type of the target grid profile;

The processing module is configured to determine a sample feature variable set matching the target cross-section feature variable set in a preset sample library, and determine the sample fault set corresponding to the matched sample cross-section feature variable set as the target power grid cross-section Set of predicted failures;

The sending module is used to send the predicted fault set to the staff.
An electronic device, characterized in that the electronic device includes:

Memory for storing fault prediction instructions; and

A processor, configured to execute the fault prediction instruction to execute the power grid expected fault set prediction method according to any one of claims 1-7.
A storage medium characterized in that a computer program is stored in the storage medium, and when the computer program is run on a computer, the computer is caused to execute the power grid prediction according to any one of claims 1-7 Fault set prediction method.