CN110321425B - Method and device for judging defect type of power grid - Google Patents

Abstract

The application discloses a method and a device for judging a power grid defect type, wherein the method comprises the following steps: acquiring a power grid defect description text, and classifying the power grid defect description text according to the type of the power grid defect; respectively segmenting each power grid defect description text, and correspondingly segmenting a word library by using the defects; extracting keywords in each defect word stock respectively to form a corresponding defect keyword stock; acquiring power grid monitoring data of a power grid in operation, and converting the monitoring data into power grid operation description text; performing word segmentation on the power grid operation description text to obtain an operation word segmentation set; extracting keywords in the operation word segmentation set to obtain an operation keyword set; matching the operation keyword set with each defect keyword library respectively; and determining the defect type of the power grid corresponding to the defect keyword library with the highest matching degree as the defect type of the power grid in operation. The method and the device can judge the defect type of the power grid more accurately.

Description

Method and device for judging defect type of power grid

Technical Field

The application relates to the technical field of power grid defect type judgment, in particular to a method and a device for judging a power grid defect type.

Background

The power grid construction tends to be intelligent and informationized more and more, but the defect recording of the power grid mainly depends on manual recording, and the manually recorded power grid defect description text contains a lot of valuable information.

Currently, in the field of power technology, analysis efficiency of a grid defect description text is low and accuracy varies from person to person. Therefore, the method for judging the type of the power grid defect is not mature enough and has low accuracy by analyzing the description text of the power grid defect. It becomes particularly important to provide a method and a device for judging the type of the power grid defect with higher accuracy.

Therefore, how to accurately judge the defect type of the power grid is a technical problem to be solved by those skilled in the art.

Disclosure of Invention

The application provides a prediction method and a prediction device for air leakage defects of sulfur hexafluoride electrical equipment, and aims to solve the problem that in the prior art, the accuracy of a judgment method for the type of the defects of a power grid is low.

In a first aspect, the present application provides a method for determining a type of a power grid defect, including:

acquiring a power grid defect description text, and classifying the power grid defect description text according to the type of the power grid defect;

dividing words of each power grid defect description text respectively to obtain a defect word division library corresponding to each power grid defect type;

extracting keywords in each defect word stock respectively to form a defect keyword stock corresponding to each power grid defect type;

acquiring power grid monitoring data of a power grid in operation, and converting the monitoring data into power grid operation description text;

performing word segmentation on the power grid operation description text to obtain an operation word segmentation set;

extracting keywords in the operation word segmentation set to obtain an operation keyword set;

matching the operation keyword set with each defect keyword library respectively, and determining the matching degree of the operation keyword set and each defect keyword library;

and determining the defect type of the power grid corresponding to the defect keyword library with the highest matching degree as the defect type of the power grid in operation.

Optionally, the obtaining the power grid defect description text classifies the power grid defect description text according to the power grid defect type, including:

acquiring historical defect description information of a power grid, defect standard information of first equipment of a power transformation and overhaul test procedure information of the power grid equipment;

converting the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment into a description text of the power grid defects, wherein the description text of the power grid defects comprises a text for describing at least one power grid defect;

classifying the grid defect description text according to the grid defect types to obtain the grid defect description text corresponding to each grid defect type.

Optionally, the extracting keywords in each defect word stock includes:

for each word in each defect word segmentation library, calculating word frequency TF of the word in the word segmentation library;

for each word in each defect word stock, calculating the total number of the defect word stocks divided by the number of the defect word stocks containing the word stocks to obtain a word stock frequency, and taking the logarithm taking 10 as the base for the word stock frequency to obtain an IDF value;

multiplying the word frequency TF with the IDF value to obtain a TF-IDF value;

and determining the word segmentation of which the TF-IDF value meets a preset threshold as the keyword.

Optionally, the extracting the keywords in the operation word segmentation set includes:

for each word in the running word segmentation set, calculating word frequency TF' of the word in the word segmentation set;

for each word segmentation in the operation word segmentation set, calculating the total number of the defect word segmentation libraries divided by the number of the defect word segmentation libraries containing the word segmentation to obtain a word-reversing library number frequency, and taking the logarithm taking 10 as the base for the word-reversing library number frequency to obtain an IDF' value;

multiplying the word frequency TF with the IDF ' value to obtain a TF ' -IDF ' value;

and determining that the TF '-IDF' value satisfies a preset threshold as the keyword.

Optionally, the matching the operation keyword set with each defect keyword library respectively, and determining the matching degree between the operation keyword set and each defect keyword library includes:

the keywords in the operation keyword set are respectively compared with the keywords in each defect keyword library, and the number of the same keywords in the operation keyword set and the defect keyword library is determined;

and determining the matching degree of the operation keyword set and the defect keyword library according to the number of the same keywords.

Optionally, the matching the operation keyword set with each defect keyword library respectively, and determining the matching degree between the operation keyword set and each defect keyword library includes:

comparing the keywords in the operation keyword set with the keywords in each defect keyword library respectively, determining the number of the same keywords in the operation keyword set and the defect keyword library, and determining the number of similar keywords in the operation keyword set and the defect keyword library;

and determining the matching degree of the operation keyword set and the defect keyword library according to the number of the same keywords and the number of the similar keywords.

In a second aspect, the present application provides a device for determining a type of a grid defect, including:

the defect data acquisition module is used for acquiring a power grid defect description text and classifying the power grid defect description text according to the power grid defect type;

the word segmentation module is used for respectively segmenting each power grid defect description text to obtain a defect word segmentation library corresponding to each power grid defect type;

the keyword extraction module is used for respectively extracting keywords in each defect word stock to form a defect keyword stock corresponding to each power grid defect type;

the monitoring data acquisition module is used for acquiring power grid monitoring data in operation and converting the monitoring data into power grid operation description text;

the word segmentation module is also used for segmenting the power grid operation description text to obtain an operation word segmentation set;

the keyword extraction module is further used for extracting keywords in the operation segmentation set to obtain an operation keyword set;

the keyword matching module is used for respectively matching the operation keyword set with each defect keyword library and determining the matching degree of the operation keyword set and each defect keyword library;

the defect type judging module is used for determining that the power grid defect type corresponding to the defect keyword library with the highest matching degree is the defect type of the power grid in operation.

Optionally, the defect data acquisition module includes:

the acquisition sub-module is used for acquiring historical defect description information of the power grid, defect standard information of the first power transformation equipment and overhaul test procedure information of the power grid equipment;

the data conversion sub-module is used for converting the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment into a power grid defect description text, wherein the power grid defect description text comprises a text for describing at least one power grid defect;

and the classification sub-module is used for classifying the power grid defect description text according to the power grid defect types to obtain power grid defect description text corresponding to each power grid defect type.

Optionally, the keyword extraction module includes:

the word frequency TF calculation sub-module is used for calculating the word frequency TF of each word in the word segmentation library for each word in each defect word segmentation library;

the IDF calculation sub-module is used for calculating the total word stock number divided by the word stock number containing the word segments for each word segment in each defect word stock to obtain a word stock number frequency, and taking the logarithm taking 10 as the base for the word stock number frequency to obtain an IDF value;

the TF-IDF calculation sub-module is used for multiplying the word frequency TF with the IDF value to obtain a TF-IDF value;

and the keyword determination submodule is used for determining that the word segmentation of the TF-IDF value meeting a preset threshold is the keyword.

As can be seen from the above technical solutions, the present application provides a method and an apparatus for determining a defect type of a power grid, where the method includes: acquiring a power grid defect description text, and classifying the power grid defect description text according to the type of the power grid defect; dividing words of each power grid defect description text respectively to obtain a defect word division library corresponding to each power grid defect type; extracting keywords in each defect word stock respectively to form a defect keyword stock corresponding to each power grid defect type; acquiring power grid monitoring data of a power grid in operation, and converting the monitoring data into power grid operation description text; performing word segmentation on the power grid operation description text to obtain an operation word segmentation set; extracting keywords in the operation word segmentation set to obtain an operation keyword set; matching the operation keyword set with each defect keyword library respectively, and determining the matching degree of the operation keyword set and each defect keyword library; and determining the defect type of the power grid corresponding to the defect keyword library with the highest matching degree as the defect type of the power grid in operation. Compared with the prior art, the defect type of the power grid can be judged more accurately through the method.

Drawings

In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.

Fig. 1 is a flowchart of a method for determining a type of a power grid defect provided in the present application;

FIG. 2 is a detailed schematic diagram of step S1 in FIG. 1;

FIG. 3 is a detailed step diagram of step S3 in FIG. 1;

fig. 4 is a schematic diagram of a position relationship between a flowchart of a method for determining a type of a power grid defect and a complementary step provided in the present application;

FIG. 5 is a schematic diagram of a device for determining a type of a power grid defect;

FIG. 6 is a schematic diagram illustrating the composition of the defect data acquisition module 100 in FIG. 5;

fig. 7 is a schematic diagram illustrating the composition of the keyword extraction module 300 in fig. 5;

FIG. 8 is a schematic diagram illustrating the composition of the keyword matching module 500 in FIG. 5;

fig. 9 is a schematic diagram of the keyword matching module 500' in fig. 5.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present application based on the embodiments herein.

The power grid construction tends to be intelligent and informationized more and more, but the defect recording of the power grid mainly depends on manual recording, and the manually recorded power grid defect description text contains a lot of valuable information.

Currently, in the field of power technology, analysis efficiency of a grid defect description text is low and accuracy varies from person to person. Therefore, the method for judging the type of the power grid defect is not mature enough and has low accuracy by analyzing the description text of the power grid defect. It becomes particularly important to provide a method and a device for judging the type of the power grid defect with higher accuracy.

Therefore, how to accurately judge the defect type of the power grid is a technical problem to be solved by those skilled in the art.

In view of this, the present application provides a method and apparatus for determining a defect type of a power grid.

In a first aspect, fig. 1 is a flowchart of a method for determining a type of a power grid defect, where, as shown in fig. 1, the method includes:

s1: acquiring a power grid defect description text, and classifying the power grid defect description text according to the type of the power grid defect;

optionally, fig. 2 is a detailed step schematic diagram of step S1 in fig. 1, as shown in fig. 2, step S1: acquiring a power grid defect description text, classifying the power grid defect description text according to the power grid defect type, and comprising the following steps:

s11: acquiring historical defect description information of a power grid, defect standard information of first equipment of a power transformation and overhaul test procedure information of the power grid equipment;

s12: converting the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment into a description text of the power grid, wherein the description text of the power grid comprises a text for describing at least one power grid defect;

s13: classifying the grid defect description texts according to the grid defect types to obtain the grid defect description texts corresponding to each grid defect type.

S2: performing word segmentation on each power grid defect description text respectively to obtain a defect word segmentation library corresponding to each power grid defect type;

s3: extracting keywords in each defect word stock respectively to form a defect keyword stock corresponding to each grid defect type;

s4: acquiring power grid monitoring data of a power grid in operation, and converting the monitoring data into power grid operation description text;

s5: performing word segmentation on the power grid operation description text to obtain an operation word segmentation set;

s6: extracting keywords in the operation word segmentation set to obtain an operation keyword set;

s7: matching the operation keyword set with each defect keyword library respectively, and determining the matching degree of the operation keyword set and each defect keyword library;

s8: and determining the defect type of the power grid corresponding to the defect keyword library with the highest matching degree as the defect type of the power grid in operation.

It should be noted that, the historical defect information of the power grid in step S11 refers to recorded information of various defects that have occurred on the power grid, such as a type of the defect, a time and a location of occurrence of the defect, parameters of related devices of the power grid when the defect occurs, geographical information of the power grid when the defect occurs, and the like. The parameters of the grid related devices when a defect occurs may be current, voltage, temperature, etc. at the location of the defect. The defect standard information of the first power transformation equipment and the overhaul test rule information of the power grid equipment are standard specification file information in the technical field of power grids. The description information of the historical defect of the power grid, the standard information of the defect of the first equipment of the power transformation and the overhaul test procedure information of the equipment of the power grid are only exemplified here, and the application is not particularly limited.

According to the method provided by the embodiment, the historical grid defect description text is analyzed in detail, and the defect keyword library is obtained by using an analysis technology and a keyword extraction technology. And performing word segmentation processing on the running power grid operation description text by using a word segmentation technology and a keyword extraction technology to obtain an operation keyword set. And matching the operation keyword set with the defect keyword library, and determining the power grid defect type corresponding to the keyword library with the highest matching degree as the defect type of the power grid in operation. The method can more accurately judge the possible defect types of the power grid in operation.

Optionally, fig. 3 is a detailed step schematic diagram of step S3 in fig. 1, and as shown in fig. 3, extracting keywords in each defect word stock respectively includes:

s31: for each word in each defect word bank, calculating the word frequency TF of the word in the word bank, wherein the TF has the following calculation formula:

TF＝n/N，

wherein N is the number of times that each word is present in the corresponding defect word stock, and N is the total number of words in the corresponding word stock;

s32: for each word in each defect word bank, calculating the total number of the defect word banks divided by the number of the defect word banks containing the word, obtaining the frequency of the word bank, taking the logarithm of the frequency of the word bank based on 10, and obtaining an IDF value, wherein the calculation formula of the IDF value is as follows:

IDF＝lg(L/l)，

wherein L is the total number of defect word segmentation libraries, and L is the number of defect word segmentation libraries containing a certain segmentation word;

s33: multiplying the word frequency TF with the IDF value to obtain a TF-IDF value, wherein the TF-IDF value has the following calculation formula:

TF-IDF＝TF*IDF＝(n/N)*lg(L/l)；

s34: and determining the word segmentation of which the TF-IDF value meets a preset threshold value as a keyword.

It should be noted that, the preset threshold is set according to the actual situation of the historical defect data of the power grid, and different power grids can correspond to the preset threshold which is not understood, and the application is not limited specifically.

Optionally, extracting keywords in the running segmentation set includes:

for each word in the running word set, calculating word frequency TF' of the word in the word set, wherein the TF has the following calculation formula:

TF’＝n’/N’，

wherein N 'is the number of times each word occurs in the corresponding operation word segmentation set, and N' is the total number of words in the corresponding word segmentation set;

for each word in the operation word segmentation set, calculating the total number of the defect word segmentation libraries divided by the number of the defect word libraries containing the word segmentation to obtain the frequency of the word reversal library, taking the logarithm of the frequency of the word reversal library based on 10 to obtain an IDF 'value, wherein the IDF' value has the following calculation formula:

IDF’＝lg(L’/l’)，

l 'is the total number of running word segmentation sets, and L' is the number of running word segmentation sets containing a certain word segmentation;

multiplying the word frequency TF 'with the IDF' value to obtain a TF '-IDF' value, wherein the calculation formula is as follows:

TF’-IDF’＝TF’*IDF’＝(n’/N’)*lg(L’/l’)；

and determining the word segmentation of which the TF '-IDF' value meets a preset threshold as the keyword.

The steps of the method for extracting keywords in the operation word set are the same as those of the method for extracting the defect analysis library, and the drawings are omitted here.

Optionally, after the grid defect description text is subjected to the aforementioned word segmentation and keyword extraction steps, when a part of keywords of the defect description text still cannot be extracted, that is, when part of keyword information of the defect description text still is missed through an intelligent extraction technology, two additional steps are required.

Fig. 4 is a schematic diagram of a position relationship between a flowchart of a method for determining a grid defect type and a supplementing step provided in the present application, as shown in fig. 4, the supplementing step includes a first supplementing step S23 and a second supplementing step S56,

s23: manually extracting keywords from the power grid defect description text which is missed through the word segmentation and keyword extraction steps, and merging the extracted keywords into a defect keyword library;

s56: manually extracting keywords aiming at the power grid operation description text which is missed through the word segmentation and keyword extraction steps respectively, and merging the extracted keywords into an operation keyword set;

optionally, the matching of the operation keyword set with each defect keyword library is performed, and the determining the matching degree of the operation keyword set and each defect keyword library includes:

the keywords in the operation keyword set are respectively compared with the keywords in each defect keyword library, and the number of the same keywords in the operation keyword set and the defect keyword library is determined;

according to the number of the same keywords, determining the matching degree of the operation keyword set and the defect keyword library, wherein the matching degree has the following calculation formula:

P＝k*M/Y，

wherein P is the matching degree of the operation keyword set and the defect keyword library, M is the number of the same keywords in the operation keyword set and the defect keyword library, Y is the total number of the keywords in the defect keyword library, and k is a coefficient.

It should be noted that the coefficient k may be set to different values according to different power grids, may be set to different values for different power grid defect types, or may be set to 1, which is not specifically limited in this application.

Optionally, the matching of the operation keyword set with each defect keyword library is performed, and the determining the matching degree of the operation keyword set and each defect keyword library includes:

the keywords in the operation keyword set are respectively compared with the keywords in each defect keyword library, the number of the same keywords in the operation keyword set and the defect keyword library is determined, and the number of similar keywords in the operation keyword set and the defect keyword library is determined;

according to the number of the same keywords and the number of similar keywords, determining the matching degree of the operation keyword set and the defect keyword library, wherein the matching degree has the following calculation formula:

P’＝k’*(M+H)/Y，

wherein P 'is the matching degree of the operation keyword set and the defect keyword library, M is the number of the same keywords in the operation keyword set and the defect keyword library, H is the number of similar keywords in the operation keyword set and the defect keyword library, Y is the total number of keywords in the defect keyword library, and k' is a coefficient.

It should be noted that the coefficient k' may be set to different values according to different power grids, may be set to different values for different power grid defect types, or may be set to 1, which is not specifically limited in this application.

According to the method, the historical grid defect description text is analyzed in detail, and an analysis technology and a keyword extraction technology are used to obtain a defect keyword library. And performing word segmentation processing on the running power grid operation description text by using a word segmentation technology and a keyword extraction technology to obtain an operation keyword set. And matching the operation keyword set with the defect keyword library, and determining the power grid defect type corresponding to the keyword library with the highest matching degree as the defect type of the power grid in operation. The method can more accurately judge the possible defect types of the power grid in operation.

In a second aspect, fig. 5 is a schematic diagram of a device for determining a type of a power grid defect, as shown in fig. 5, a device 0000 for determining a type of a power grid defect, including:

the defect data acquisition module 100 is used for acquiring a power grid defect description text and classifying the power grid defect description text according to the type of the power grid defect;

the word segmentation module 200 is configured to segment each grid defect description text to obtain a defect word segmentation library corresponding to each grid defect type;

the keyword extraction module 300 is configured to extract keywords in each defect word stock respectively, and form a defect keyword stock corresponding to each power grid defect type;

the monitoring data acquisition module 400 is used for acquiring power grid monitoring data in operation and converting the monitoring data into power grid operation description text;

the word segmentation module 200 is further configured to segment the power grid operation description text to obtain an operation word segmentation set;

the keyword extraction module 300 is further configured to extract keywords in the running segmentation set, so as to obtain a running keyword set;

the keyword matching module 500 is configured to match the running keyword set with each defect keyword library, and determine a matching degree between the running keyword set and each defect keyword library;

the defect type determining module 600 is configured to determine a defect type of the power grid corresponding to the defect keyword library with the highest matching degree as a defect type of the power grid in operation.

Optionally, fig. 6 is a schematic diagram illustrating the composition of the module for acquiring defect data 100 in fig. 5, and as shown in fig. 6, the module for acquiring defect data 100 includes:

the acquisition sub-module 110 is used for acquiring the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment;

the data conversion sub-module 120 is configured to convert the grid history defect description information, the substation first-time equipment defect standard information, and the grid equipment overhaul test procedure information into a grid defect description text, where the grid defect description text includes a text for describing at least one grid defect;

the classifying sub-module 130 is configured to classify the grid defect description text according to the grid defect types, and obtain a grid defect description text corresponding to each grid defect type.

Optionally, fig. 7 is a schematic diagram illustrating the composition of the keyword extraction module 300 in fig. 5, and as shown in fig. 7, the keyword extraction module 300 includes:

the word frequency TF calculation submodule 310 is configured to calculate, for each word in each defect word bank, a word frequency TF of the word in the word bank according to the following calculation formula:

TF＝n/N，

wherein N is the number of times each word is present in the corresponding defect word stock, and N is the total number of words in the corresponding word stock;

an IDF calculation sub-module 320, configured to calculate, for each word in each of the defect word banks, a total word bank number divided by the number of word banks containing the word, to obtain an inverse word bank number frequency, and taking a logarithm based on 10 for the inverse word bank number frequency, to obtain an IDF value, where a calculation formula of the IDF value is as follows:

IDF＝lg(L/l)，

wherein L is the total number of defect word segmentation libraries, and L is the number of defect word segmentation libraries containing a certain segmentation word;

the TF-IDF calculation submodule 330 is configured to multiply the word frequency TF with the IDF value to obtain a TF-IDF value, where a TF-IDF value calculation formula is as follows:

TF-IDF＝TF*IDF＝(n/N)*lg(L/l)；

and a keyword determining sub-module 340, configured to determine that the word segment of the TF-IDF value satisfying the preset threshold is the keyword.

It is easy to understand that the keyword extraction module is applied to extracting keywords from the defect word segmentation library and the operation word segmentation set, respectively, and description of the calculation formula of the TF-IDF value in this embodiment takes the case of extracting keywords from the defect word segmentation library as an example. When the keyword is extracted from the operation word segmentation set, the calculation formula of the TF-IDF value is unchanged, only the input objects are different, and all the input objects are changed from the defect word segmentation library to the operation word segmentation set, so that the description is omitted.

Optionally, fig. 8 is a schematic diagram illustrating the composition of the keyword matching module 500 in fig. 5, and as shown in fig. 8, the keyword matching module 500 includes:

keyword alignment sub-module 510: the method comprises the steps of comparing keywords in an operation keyword set with keywords in each defect keyword library respectively, and determining the number of identical keywords in the operation keyword set and the defect keyword library;

matching degree calculation submodule 520: according to the number of the same keywords, determining the matching degree of the operation keyword set and the defect keyword library, wherein the matching degree has the following calculation formula:

P＝k*M/Y，

wherein P is the matching degree of the operation keyword set and the defect keyword library, M is the number of the same keywords in the operation keyword set and the defect keyword library, Y is the total number of the keywords in the defect keyword library, and k is a coefficient.

It should be noted that the coefficient k may be set to different values according to different power grids, may be set to different values for different power grid defect types, or may be set to 1, which is not specifically limited in this application.

Optionally, fig. 9 is a schematic diagram of the composition of the keyword matching module 500 'in fig. 5, and as shown in fig. 9, the keyword matching module 500' includes:

keyword comparison unit 511: the method comprises the steps of comparing keywords in an operation keyword set with keywords in each defect keyword library, determining the number of identical keywords in the operation keyword set and the defect keyword library, and determining the number of similar keywords in the operation keyword set and the defect keyword library;

matching degree calculation unit 522: the matching degree of the operation keyword set and the defect keyword library is determined according to the number of the same keywords and the number of similar keywords, and the calculation formula of the matching degree is as follows:

P’＝k’*(M+H)/Y，

wherein P 'is the matching degree of the operation keyword set and the defect keyword library, M is the number of the same keywords in the operation keyword set and the defect keyword library, H is the number of similar keywords in the operation keyword set and the defect keyword library, Y is the total number of keywords in the defect keyword library, and k' is a coefficient.

It should be noted that the coefficient k' may be set to different values according to different power grids, may be set to different values for different power grid defect types, or may be set to 1, which is not specifically limited in this application.

According to the method and the device, the historical grid defect description text is analyzed in detail, and the defect keyword library is obtained by using an analysis technology and a keyword extraction technology. And performing word segmentation processing on the running power grid operation description text by using a word segmentation technology and a keyword extraction technology to obtain an operation keyword set. And matching the operation keyword set with the defect keyword library, and determining the power grid defect type corresponding to the keyword library with the highest matching degree as the defect type of the power grid in operation. The method and the device can more accurately judge the possible defect types of the power grid in operation.

It will be apparent to those skilled in the art that the techniques of embodiments of the present invention may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solutions in the embodiments of the present invention may be embodied in essence or what contributes to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.

The same or similar parts between the various embodiments in this specification are referred to each other. In particular, for the embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference should be made to the description of the method embodiments for the matters.

Claims (7)

1. The method for judging the type of the power grid defect is characterized by comprising the following steps of:

acquiring historical defect description information of a power grid, defect standard information of first equipment of a power transformation and overhaul test procedure information of the power grid equipment;

converting the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment into a description text of the power grid defects, wherein the description text of the power grid defects comprises a text for describing at least one power grid defect;

classifying the grid defect description text according to the grid defect types to obtain a grid defect description text corresponding to each grid defect type;

dividing words of each power grid defect description text respectively to obtain a defect word division library corresponding to each power grid defect type;

extracting keywords in each defect word stock respectively to form a defect keyword stock corresponding to each power grid defect type;

acquiring power grid monitoring data of a power grid in operation, and converting the monitoring data into power grid operation description text;

performing word segmentation on the power grid operation description text to obtain an operation word segmentation set;

extracting keywords in the operation word segmentation set to obtain an operation keyword set;

matching the operation keyword set with each defect keyword library respectively, and determining the matching degree of the operation keyword set and each defect keyword library;

and determining the defect type of the power grid corresponding to the defect keyword library with the highest matching degree as the defect type of the power grid in operation.

2. The method of claim 1, wherein the extracting keywords in each of the defect word libraries includes:

for each word in each defect word segmentation library, calculating word frequency TF of the word in the word segmentation library;

for each word in each defect word stock, calculating the total number of the defect word stocks divided by the number of the defect word stocks containing the word stocks to obtain a word stock frequency, and taking the logarithm taking 10 as the base for the word stock frequency to obtain an IDF value;

multiplying the word frequency TF with the IDF value to obtain a TF-IDF value;

and determining the word segmentation of which the TF-IDF value meets a preset threshold as the keyword.

3. The method of claim 1, wherein the extracting keywords in the running segmentation set comprises:

for each word in the running word segmentation set, calculating word frequency TF' of the word in the word segmentation set;

for each word segmentation in the operation word segmentation set, calculating the total number of the defect word segmentation libraries divided by the number of the defect word segmentation libraries containing the word segmentation to obtain a word-reversing library number frequency, and taking the logarithm taking 10 as the base for the word-reversing library number frequency to obtain an IDF' value;

multiplying the word frequency TF 'with the IDF' value to obtain a TF '-IDF' value;

and determining that the TF '-IDF' value satisfies a preset threshold as the keyword.

4. The method of claim 1, wherein the matching the running keyword set with each defective keyword library respectively, and determining the matching degree of the running keyword set with each defective keyword library comprises:

the keywords in the operation keyword set are respectively compared with the keywords in each defect keyword library, and the number of the same keywords in the operation keyword set and the defect keyword library is determined;

and determining the matching degree of the operation keyword set and the defect keyword library according to the number of the same keywords.

5. The method of claim 1, wherein the matching the running keyword set with each defective keyword library respectively, and determining the matching degree of the running keyword set with each defective keyword library comprises:

comparing the keywords in the operation keyword set with the keywords in each defect keyword library respectively, determining the number of the same keywords in the operation keyword set and the defect keyword library, and determining the number of similar keywords in the operation keyword set and the defect keyword library;

and determining the matching degree of the operation keyword set and the defect keyword library according to the number of the same keywords and the number of the similar keywords.

6. The utility model provides a judging device of electric wire netting defect type which characterized in that includes:

the acquisition sub-module is used for acquiring historical defect description information of the power grid, defect standard information of the first power transformation equipment and overhaul test procedure information of the power grid equipment;

the data conversion sub-module is used for converting the historical defect description information of the power grid, the defect standard information of the first power transformation equipment and the overhaul test procedure information of the power grid equipment into a power grid defect description text, wherein the power grid defect description text comprises a text for describing at least one power grid defect;

the classifying sub-module is used for classifying the power grid defect description text according to the power grid defect types to obtain power grid defect description text corresponding to each power grid defect type;

the word segmentation module is used for respectively segmenting each power grid defect description text to obtain a defect word segmentation library corresponding to each power grid defect type;

the keyword extraction module is used for respectively extracting keywords in each defect word stock to form a defect keyword stock corresponding to each power grid defect type;

the monitoring data acquisition module is used for acquiring power grid monitoring data in operation and converting the monitoring data into power grid operation description text;

the word segmentation module is also used for segmenting the power grid operation description text to obtain an operation word segmentation set;

the keyword extraction module is further used for extracting keywords in the operation segmentation set to obtain an operation keyword set;

the keyword matching module is used for respectively matching the operation keyword set with each defect keyword library and determining the matching degree of the operation keyword set and each defect keyword library;

the defect type judging module is used for determining that the power grid defect type corresponding to the defect keyword library with the highest matching degree is the defect type of the power grid in operation.

7. The apparatus of claim 6, wherein the keyword extraction module comprises:

the word frequency TF calculation sub-module is used for calculating the word frequency TF of each word in the word segmentation library for each word in each defect word segmentation library;

the IDF calculation sub-module is used for calculating the total word stock number divided by the word stock number containing the word segments for each word segment in each defect word stock to obtain a word stock number frequency, and taking the logarithm taking 10 as the base for the word stock number frequency to obtain an IDF value;

the TF-IDF calculation sub-module is used for multiplying the word frequency TF with the IDF value to obtain a TF-IDF value;

and the keyword determination submodule is used for determining that the word segmentation of the TF-IDF value meeting a preset threshold is the keyword.