CN110321425A - A kind of judgment method and device of grounding grid defect type - Google Patents
A kind of judgment method and device of grounding grid defect type Download PDFInfo
- Publication number
- CN110321425A CN110321425A CN201910624829.1A CN201910624829A CN110321425A CN 110321425 A CN110321425 A CN 110321425A CN 201910624829 A CN201910624829 A CN 201910624829A CN 110321425 A CN110321425 A CN 110321425A
- Authority
- CN
- China
- Prior art keywords
- defect
- participle
- keyword
- grounding grid
- text
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application discloses the judgment method and device of a kind of grounding grid defect type, and method includes: to obtain grounding grid defect to describe text, describes text classification to grounding grid defect according to grounding grid defect type；It describes text to every kind of grounding grid defect respectively to segment, corresponding defect segments library；The keyword in each defect participle library is extracted respectively, forms corresponding defect keywords database；Monitoring data are converted to operation of power networks and describe text by the power network monitoring data for obtaining power grid in operation；Text is described to operation of power networks to segment, and obtains operation participle collection；The keyword that operation participle is concentrated is extracted, operation keyword set is obtained；Operation keyword set is matched with each defect keywords database respectively；Determine that the corresponding grounding grid defect type of the highest defect keywords database of matching degree is the defect type of power grid in operation.This method and device can more accurately judge the defect type of power grid.
Description
Technical field
This application involves the technical fields of grounding grid defect type judgement, are specifically related to a kind of judgement of grounding grid defect type
Method and device.
Background technique
Power grid construction increasingly tends to intelligentize and informatization, but the defect record of power grid mainly relies on manual record,
The grounding grid defect of manual record describes in text containing many valuable information.
Currently, in technical field of electric power, the analysis efficiency for describing text for grounding grid defect is lower and accuracy
It varies with each individual.Therefore, text is described by analyzing grounding grid defect, judges that the method for grounding grid defect type is not mature enough, accurately
Property is lower.The judgment method and device for providing a kind of higher grounding grid defect type of accuracy become particularly important.
So how to realize to the accurate judgement of grounding grid defect type, it is urgently to be resolved to have become those skilled in the art
The technical issues of.
Summary of the invention
This application provides the prediction techniques and device of a kind of sulfur hexafluoride electrical equipment Air Leakage Defect, to solve existing skill
In art, the lower problem of the judgment method accuracy of grounding grid defect type.
In a first aspect, the application provides a kind of judgment method of grounding grid defect type, comprising:
It obtains grounding grid defect and describes text, text classification is described to the grounding grid defect according to grounding grid defect type；
Text is described to grounding grid defect described in every kind respectively to segment, and it is corresponding to obtain every kind of grounding grid defect type
Defect segments library；
The keyword in each defect participle library is extracted respectively, is formed corresponding with grounding grid defect type described in every kind
Defect keywords database；
The monitoring data are converted to operation of power networks and describe text by the power network monitoring data for obtaining power grid in operation；
Text is described to the operation of power networks to segment, and obtains operation participle collection；
The keyword that the operation participle is concentrated is extracted, operation keyword set is obtained；
The operation keyword set is matched with each defect keywords database respectively, determines the operation keyword set
With the matching degree of each defect keywords database；
Determine that the corresponding grounding grid defect type of the highest defect keywords database of matching degree is the defect of power grid in operation
Type.
Optionally, the acquisition grounding grid defect describes text, describes text to the grounding grid defect according to grounding grid defect type
This classification, comprising:
Obtain power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment checking experiment
Protocol information；
The power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment are overhauled and tried
Protocol information is tested, grounding grid defect is converted into and describes text, the grounding grid defect describes to include for describing at least one in text
The text of grounding grid defect；
Text classification is described to the grounding grid defect according to grounding grid defect type, it is corresponding to obtain every kind of grounding grid defect type
Grounding grid defect describes text.
Optionally, the keyword extracted in each defect participle library respectively, comprising:
For each participle in each defect participle library, word frequency of the participle in the participle library is calculated
TF；
For each participle in each defect participle library, defect participle library sum is calculated divided by containing the participle
Defect segment dictionary number, obtain anti-dictionary number frequency, denary logarithm taken to the anti-dictionary number frequency, obtains IDF value；
The word frequency TF is multiplied with the IDF value, obtains TF-IDF value；
The participle for determining that the TF-IDF value meets preset threshold is the keyword.
Optionally, the keyword for extracting operation participle and concentrating, comprising:
For each participle that the operation participle is concentrated, the word frequency TF ' that the participle is concentrated in the participle is calculated；
For each participle that the operation participle is concentrated, defect participle library sum is calculated divided by containing the participle
Defect segment library number, obtain anti-dictionary number frequency, denary logarithm taken to the anti-dictionary number frequency, obtains IDF ' value；
The word frequency TF is multiplied with the IDF ' value, obtains TF '-IDF ' value；
The participle for determining that the TF '-IDF ' value meets preset threshold is the keyword.
Optionally, described to match operation keyword set with each defect keywords database respectively, determine the operation
The matching degree of keyword set and each defect keywords database, comprising:
Keyword in the operation keyword set is carried out with the keyword in each defect keywords database respectively
It compares, determines the quantity of same keyword in the operation keyword set and the defect keywords database；
According to the quantity of the same keyword, the matching of the operation keyword set and the defect keywords database is determined
Degree.
Optionally, described to match operation keyword set with each defect keywords database respectively, determine the operation
The matching degree of keyword set and each defect keywords database, comprising:
Keyword in the operation keyword set is carried out with the keyword in each defect keywords database respectively
It compares, determines the quantity of same keyword in the operation keyword set and the defect keywords database, and determine the fortune
The quantity of close keyword in row keyword set and the defect keywords database；
According to the quantity of the quantity of the same keyword and the close keyword, determine the operation keyword set with
The matching degree of the defect keywords database.
Second aspect, the application provide a kind of judgment means of grounding grid defect type, comprising:
Defective data module is obtained, text is described for obtaining grounding grid defect, according to grounding grid defect type to the power grid
Defect describes text classification；
Word segmentation module is segmented for describing text to grounding grid defect described in every kind respectively, obtains every kind of power grid
The corresponding defect of defect type segments library；
Keyword extracting module is formed and every kind of institute for extracting the keyword in each defect participle library respectively
State the corresponding defect keywords database of grounding grid defect type；
It obtains monitoring data module and the monitoring data is converted into electricity for obtaining running power network monitoring data
Net operation description text；
The word segmentation module is also used to describe text to the operation of power networks to segment, and obtains operation participle collection；
The keyword extracting module is also used to extract the keyword that the operation participle is concentrated, obtains operation keyword
Collection；
Keywords matching module, for the operation keyword set to be matched with each defect keywords database respectively,
Determine the matching degree of the operation keyword set and each defect keywords database；
Defect type judgment module, for determining the corresponding grounding grid defect class of the highest defect keywords database of matching degree
Type is the defect type of power grid in operation.
Optionally, the acquisition defective data module, comprising:
Acquisition submodule, for obtain power grid historic defects description information, power transformation first time equipment deficiency standard information and
Grid equipment checking experiment protocol information；
Data conversion submodule is used for the power grid historic defects description information, power transformation first time equipment deficiency standard
Information and grid equipment checking experiment protocol information are converted into grounding grid defect and describe text, and the grounding grid defect describes in text
Including the text for describing at least one grounding grid defect；
Classification submodule obtains every kind of electricity for describing text classification to the grounding grid defect according to grounding grid defect type
The corresponding grounding grid defect of net defect type describes text.
Optionally, keyword extracting module, comprising:
Word frequency TF computational submodule, for calculating the participle for each participle in each defect participle library
Word frequency TF in the participle library；
IDF computational submodule, for for each participle in each defect participle library, calculate total dictionary number divided by
Dictionary number containing the participle obtains anti-dictionary number frequency, takes denary logarithm to the anti-dictionary number frequency, obtains
IDF value；
TF-IDF computational submodule obtains TF-IDF value for the word frequency TF to be multiplied with the IDF value；
Keyword determines submodule, and the participle for determining that the TF-IDF value meets preset threshold is the keyword.
From the above technical scheme, the application provides the judgment method and device of a kind of grounding grid defect type, the side
Method includes: to obtain grounding grid defect to describe text, describes text classification to the grounding grid defect according to grounding grid defect type；It is right respectively
Every kind of grounding grid defect describes text and is segmented, and obtains every kind of grounding grid defect type corresponding defect participle library；Point
The keyword in each defect participle library is indescribably taken, defect keyword corresponding with grounding grid defect type described in every kind is formed
Library；The monitoring data are converted to operation of power networks and describe text by the power network monitoring data for obtaining power grid in operation；To the electricity
Net operation description text is segmented, and operation participle collection is obtained；The keyword that the operation participle is concentrated is extracted, operation is obtained and closes
Keyword collection；The operation keyword set is matched with each defect keywords database respectively, determines the operation keyword set
With the matching degree of each defect keywords database；Determine the corresponding grounding grid defect of the highest defect keywords database of matching degree
Type is the defect type of power grid in operation.Compared with prior art, by the present processes, can more accurately judge
The defect type of power grid.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without any creative labor,
It is also possible to obtain other drawings based on these drawings.
Fig. 1 is a kind of judgment method flow chart of grounding grid defect type provided by the present application；
Fig. 2 is the detailed step schematic diagram of step S1 in Fig. 1；
Fig. 3 is the detailed step schematic diagram of step S3 in Fig. 1；
Fig. 4 is the positional relationship of a kind of judgment method flow chart of grounding grid defect type provided by the present application and supplement step
Schematic diagram；
Fig. 5 is a kind of judgment means composition schematic diagram for the grounding grid defect type that please be provided itself；
Fig. 6 is the composition schematic diagram that defective data module 100 is obtained in Fig. 5；
Fig. 7 is the composition schematic diagram of keyword extracting module 300 in Fig. 5；
Fig. 8 is the composition schematic diagram of Keywords matching module 500 in Fig. 5；
Fig. 9 is the composition schematic diagram of Keywords matching module 500 ' in Fig. 5.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete
Site preparation description, it is clear that described embodiment is only a part of the embodiment of the application, instead of all the embodiments.Base
Embodiment in the application, those of ordinary skill in the art are obtained all without making creative work
Other embodiments shall fall in the protection scope of this application.
Power grid construction increasingly tends to intelligentize and informatization, but the defect record of power grid mainly relies on manual record,
The grounding grid defect of manual record describes in text containing many valuable information.
Currently, in technical field of electric power, the analysis efficiency for describing text for grounding grid defect is lower and accuracy
It varies with each individual.Therefore, text is described by analyzing grounding grid defect, judges that the method for grounding grid defect type is not mature enough, accurately
Property is lower.The judgment method and device for providing a kind of higher grounding grid defect type of accuracy become particularly important.
So how to realize to the accurate judgement of grounding grid defect type, it is urgently to be resolved to have become those skilled in the art
The technical issues of.
In view of this, the application provides the judgment method and device of a kind of grounding grid defect type.
In a first aspect, Fig. 1 is a kind of judgment method flow chart of grounding grid defect type provided by the present application, as shown in Figure 1,
A kind of judgment method of grounding grid defect type, comprising:
S1: it obtains grounding grid defect and describes text, text classification is described to grounding grid defect according to grounding grid defect type；
Optionally, Fig. 2 is the detailed step schematic diagram of step S1 in Fig. 1, as shown in Fig. 2, step S1: obtaining grounding grid defect
Text is described, text classification is described to the grounding grid defect according to grounding grid defect type, comprising:
S11: power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment maintenance are obtained
Testing regulations information；
S12: power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment are overhauled and tried
Protocol information is tested, grounding grid defect is converted into and describes text, grounding grid defect describes to include for describing at least one power grid in text
The text of defect；
S13: describing text classification to grounding grid defect according to grounding grid defect type, and it is corresponding to obtain every kind of grounding grid defect type
Grounding grid defect describes text.
S2: describing text to every kind of grounding grid defect respectively and segment, and obtains the corresponding defect of every kind of grounding grid defect type
Segment library；
S3: the keyword in each defect participle library is extracted respectively, forms defect corresponding with every kind of grounding grid defect type
Keywords database；
S4: the power network monitoring data of power grid in operation are obtained, monitoring data are converted into operation of power networks and describe text；
S5: describing text to operation of power networks and segment, and obtains operation participle collection；
S6: the keyword that operation participle is concentrated is extracted, operation keyword set is obtained；
S7: operation keyword set is matched with each defect keywords database respectively, determines operation keyword set and every
The matching degree of a defect keywords database；
S8: determine that the corresponding grounding grid defect type of the highest defect keywords database of matching degree is the defect class of power grid in operation
Type.
It should be noted that the power grid historic defects information in step S11 refer to power grid once occurred it is various lack
The time of defect generation, position, the parameter of power grid relevant device when defect occurs for sunken record information, such as the type of defect,
The geography information of power grid etc. of defect occurs.The parameter of power grid relevant device can be the electricity of defective locations when defect occurs
Stream, voltage and temperature etc..Power transformation first time equipment deficiency standard information and grid equipment checking experiment protocol information are power grid skills
The standard criterion class file information in art field.Herein, it is only exemplary and illustrates power grid historic defects description information, power transformation
First time equipment deficiency standard information and grid equipment checking experiment protocol information, the application are not specifically limited.
Method provided in this embodiment describes text by the grounding grid defect to history and carries out detailed analysis, uses analysis
Technology and keyword extraction techniques obtain defect keywords database.Equally using participle technique and keyword extraction techniques to operation
In operation of power networks describe text carry out word segmentation processing, obtain operation keyword set.It will run crucial with defect in keyword set
Dictionary is matched, and determines that the corresponding grounding grid defect type of the highest keywords database of matching degree is the defect class of power grid in operation
Type.This method can more accurately judge the defect type that power grid is likely to occur in operation.
Optionally, Fig. 3 is the detailed step schematic diagram of step S3 in Fig. 1, as shown in figure 3, extracting each defect point respectively
Keyword in dictionary, comprising:
S31: for each participle in each defect participle library, word frequency TF, TF calculating of the participle in participle library is calculated
Formula is as follows:
TF=n/N,
Wherein, n is that each participle segments the number occurred in library in corresponding defect, and N is the corresponding participle library of each participle
The sum of middle participle；
S32: for each participle in each defect participle library, defect participle library sum is calculated divided by lacking containing participle
Participle dictionary number is fallen into, anti-dictionary number frequency is obtained, denary logarithm is taken to anti-dictionary number frequency, obtains IDF value, IDF value
Calculation formula is as follows:
IDF=lg (L/l),
Wherein, L is that defect segments library sum, and l is that the defect containing a certain participle segments library number；
S33: word frequency TF is multiplied with the IDF value, obtains TF-IDF value, TF-IDF value calculation formula is as follows:
TF-IDF=TF*IDF=(n/N) * lg (L/l)；
S34: the participle for determining that TF-IDF value meets preset threshold is keyword.
It should be noted that preset threshold is set according to the actual conditions of power grid historical defect data, different electricity
Net can correspond to the preset threshold being ignorant of, and the application is not specifically limited.
Optionally, the keyword that operation participle is concentrated is extracted, comprising:
For running each participle for segmenting and concentrating, calculates participle and segmenting word frequency TF ', the TF calculation formula of concentration such as
Under:
TF '=n '/N ',
Wherein, wherein n ' is the number that each participle occurs in corresponding operation participle concentration, and N ' is each participle correspondence
Segment the sum for the participle concentrated；
For each participle that operation participle is concentrated, defect participle library sum is calculated divided by the defect containing participle and segments library
Number obtains anti-dictionary number frequency, takes denary logarithm to anti-dictionary number frequency, obtains IDF ' value, IDF ' value calculation formula is such as
Under:
IDF '=lg (L '/l '),
L ' is operation participle collection sum, and l ' is the operation participle collection number containing a certain participle；
Word frequency TF ' is multiplied with IDF ' value, obtains TF '-IDF ' value, calculation formula is as follows:
TF '-IDF '=TF ' * IDF '=(n '/N ') * lg (L '/l ')；
The participle for determining that TF '-IDF ' value meets preset threshold is the keyword.
It should be noted that the method extracted the method and step for the keyword that operation participle is concentrated and extract defect analysis library
Step is identical, omits attached drawing herein.
Optionally, after grounding grid defect describes text by previously mentioned participle and keyword extraction step, still
Extractive technique when having segmental defect to describe the keyword of text could not extract, that is, by intelligence can still miss portion
When the defect divided describes the key word information of text, two supplement steps are needed.
Fig. 4 is the positional relationship of a kind of judgment method flow chart of grounding grid defect type provided by the present application and supplement step
Schematic diagram, as shown in figure 4, supplement step includes that the first supplement step S23 and second supplement step S56,
S23: text is described for the grounding grid defect missed by participle and keyword extraction step respectively and is manually mentioned
Keyword is taken, and the keyword extracted is merged into defect keywords database；
S56: text is described for the operation of power networks missed by participle and keyword extraction step respectively and is manually mentioned
Keyword is taken, and the keyword extracted is merged into operation keyword set；
Optionally, operation keyword set is matched with each defect keywords database respectively, determines operation keyword set
With the matching degree of each defect keywords database, comprising:
The keyword run in keyword set is compared with the keyword in each defect keywords database respectively, is determined
Run the quantity of same keyword in keyword set and defect keywords database；
According to the quantity of same keyword, the matching degree of operation keyword set and defect keywords database is determined, matching degree
Calculation formula is as follows:
P=k*M/Y,
Wherein, P is the matching degree for running keyword set and defect keywords database, and M is that operation keyword set and defect are crucial
The quantity of same keyword in dictionary, Y are the sum of keyword in defect keywords database, and k is a coefficient.
It should be noted that coefficient k can set different values according to different power grids, different power grids can also be directed to
Defect type sets different values, k can also be set as to 1, the application is not specifically limited.
Optionally, operation keyword set is matched with each defect keywords database respectively, determines operation keyword set
With the matching degree of each defect keywords database, comprising:
The keyword run in keyword set is compared with the keyword in each defect keywords database respectively, is determined
The quantity of same keyword in keyword set and defect keywords database is run, and determines operation keyword set and defect keyword
The quantity of close keyword in library；
According to the quantity of the quantity of same keyword and close keyword, operation keyword set and defect keywords database are determined
Matching degree, the calculation formula of matching degree is as follows:
P '=k ' * (M+H)/Y,
Wherein, P ' is the matching degree for running keyword set and defect keywords database, and M is that operation keyword set and defect are crucial
The quantity of same keyword in dictionary, H are the quantity for running close keyword in keyword set and defect keywords database, and Y is scarce
The sum of keyword in keywords database is fallen into, k ' is a coefficient.
It should be noted that coefficient k ' different values can be set according to different power grids, different electricity can also be directed to
Net defect type sets different values, k ' can also be set as to 1, the application is not specifically limited.
Method provided by the present application describes text by the grounding grid defect to history and carries out detailed analysis, uses analysis skill
Art and keyword extraction techniques obtain defect keywords database.Equally using participle technique and keyword extraction techniques in operation
Operation of power networks describe text carry out word segmentation processing, obtain operation keyword set.Will operation keyword set in defect keyword
Library is matched, and determines that the corresponding grounding grid defect type of the highest keywords database of matching degree is the defect type of power grid in operation.
This method can more accurately judge the defect type that power grid is likely to occur in operation.
Second aspect, Fig. 5 are a kind of judgment means composition schematic diagram of grounding grid defect type provided by the present application, such as Fig. 5
It is shown, a kind of judgment means 0000 of grounding grid defect type, comprising:
Defective data module 100 is obtained, describes text for obtaining grounding grid defect, power grid is lacked according to grounding grid defect type
Fall into description text classification；
Word segmentation module 200 is segmented for describing text to every kind of grounding grid defect respectively, obtains every kind of grounding grid defect class
The corresponding defect of type segments library；
Keyword extracting module 300 is formed and every kind of power grid for extracting the keyword in each defect participle library respectively
The corresponding defect keywords database of defect type；
It obtains monitoring data module 400 and monitoring data is converted into power grid for obtaining running power network monitoring data
Operation description text；
Word segmentation module 200 is also used to describe text to operation of power networks to segment, and obtains operation participle collection；
Keyword extracting module 300 is also used to extract the keyword that operation participle is concentrated, obtains operation keyword set；
Keywords matching module 500 is matched with each defect keywords database, really respectively for that will run keyword set
Surely the matching degree of keyword set and each defect keywords database is run；
Defect type judgment module 600, for determining the corresponding grounding grid defect class of the highest defect keywords database of matching degree
Type is the defect type of power grid in operation.
Optionally, Fig. 6 is the composition schematic diagram that defective data module 100 is obtained in Fig. 5, as shown in fig. 6, obtaining defect number
According to module 100, comprising:
Acquisition submodule 110, for obtaining power grid historic defects description information, power transformation first time equipment deficiency standard information
And grid equipment checking experiment protocol information；
Data conversion submodule 120 is used for the power grid historic defects description information, power transformation first time equipment deficiency mark
Calibration information and grid equipment checking experiment protocol information are converted into grounding grid defect and describe text, and grounding grid defect describes to wrap in text
Include the text for describing at least one grounding grid defect；
Classification submodule 130 obtains every kind of power grid for describing text classification to grounding grid defect according to grounding grid defect type
The corresponding grounding grid defect of defect type describes text.
Optionally, Fig. 7 is the composition schematic diagram of keyword extracting module 300 in Fig. 5, as shown in fig. 7, keyword extraction mould
Block 300, comprising:
Word frequency TF computational submodule 310, for calculating participle and segmenting for each participle in each defect participle library
Word frequency TF, TF calculation formula in library is as follows:
TF=n/N,
Wherein, n is that each participle segments the number occurred in library in corresponding defect, and N is the corresponding participle library of each participle
In participle sum；
IDF computational submodule 320, for calculating total dictionary number for each participle in each defect participle library
Divided by the dictionary number for containing the participle, anti-dictionary number frequency is obtained, denary logarithm is taken to the anti-dictionary number frequency, is obtained
Calculation formula to IDF value, IDF value is as follows:
IDF=lg (L/l),
Wherein, L is that defect segments library sum, and l is that the defect containing a certain participle segments library number；
TF-IDF computational submodule 330 obtains TF-IDF value, TF- for the word frequency TF to be multiplied with the IDF value
IDF value calculation formula is as follows:
TF-IDF=TF*IDF=(n/N) * lg (L/l)；
Keyword determines submodule 340, and the participle for determining that the TF-IDF value meets preset threshold is the key
Word.
It is easily understood that keyword extracting module carries out keyword and mentions to defect participle library and operation participle collection respectively
Shi Douhui is taken to be applied to, the present embodiment is about the description of the calculation formula to TF-IDF value to extract keyword to defect participle library
Situation for.When segmenting collection extraction keyword to operation, the calculation formula of TF-IDF value is constant, and only input object is different,
All input objects are changed to operation participle collection by defect participle library, and details are not described herein again.
Optionally, Fig. 8 is the composition schematic diagram of Keywords matching module 500 in Fig. 5, as shown in figure 8, Keywords matching mould
Block 500, comprising:
Keyword compare submodule 510: for will run the keyword in keyword set respectively with each defect keyword
Keyword in library is compared, and determines the quantity of same keyword in operation keyword set and defect keywords database；
Matching degree computational submodule 520: according to the quantity of same keyword, operation keyword set and defect keyword are determined
The calculation formula of the matching degree in library, matching degree is as follows:
P=k*M/Y,
Wherein, P is the matching degree for running keyword set and defect keywords database, and M is that operation keyword set and defect are crucial
The quantity of same keyword in dictionary, Y are the sum of keyword in defect keywords database, and k is a coefficient.
It should be noted that coefficient k can set different values according to different power grids, different power grids can also be directed to
Defect type sets different values, k can also be set as to 1, the application is not specifically limited.
Optionally, Fig. 9 is the composition schematic diagram of Keywords matching module 500 ' in Fig. 5, as shown in figure 9, Keywords matching
Module 500 ', comprising:
Keyword comparing unit 511: for will run the keyword in keyword set respectively with each defect keywords database
In keyword be compared, determine operation keyword set and defect keywords database in same keyword quantity, and determine
Run the quantity of close keyword in keyword set and defect keywords database；
Matching degree computing unit 522: for determining operation according to the quantity of same keyword and the quantity of close keyword
The calculation formula of the matching degree of keyword set and defect keywords database, matching degree is as follows:
P '=k ' * (M+H)/Y,
Wherein, P ' is the matching degree for running keyword set and defect keywords database, and M is that operation keyword set and defect are crucial
The quantity of same keyword in dictionary, H are the quantity for running close keyword in keyword set and defect keywords database, and Y is scarce
The sum of keyword in keywords database is fallen into, k ' is a coefficient.
It should be noted that coefficient k ' different values can be set according to different power grids, different electricity can also be directed to
Net defect type sets different values, k ' can also be set as to 1, the application is not specifically limited.
Method and apparatus provided by the present application describe text by the grounding grid defect to history and carry out detailed analysis, use
Analytical technology and keyword extraction techniques obtain defect keywords database.Equally use participle technique and keyword extraction techniques pair
Running operation of power networks describes text and carries out word segmentation processing, obtains operation keyword set.Will operation keyword set in defect
Keywords database is matched, and determines that the corresponding grounding grid defect type of the highest keywords database of matching degree is the defect of power grid in operation
Type.This method and device can more accurately judge the defect type that power grid is likely to occur in operation.
It is required that those skilled in the art can be understood that the technology in the embodiment of the present invention can add by software
The mode of general hardware platform realize.Based on this understanding, the technical solution in the embodiment of the present invention substantially or
Say that the part that contributes to existing technology can be embodied in the form of software products, which can deposit
Storage is in storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) execute certain part institutes of each embodiment of the present invention or embodiment
The method stated.
Same and similar part may refer to each other between each embodiment in this specification.Especially for embodiment
Speech, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to saying in embodiment of the method
It is bright.
Claims (9)
1. a kind of judgment method of grounding grid defect type characterized by comprising
It obtains grounding grid defect and describes text, text classification is described to the grounding grid defect according to grounding grid defect type；
Text is described to grounding grid defect described in every kind respectively to segment, and obtains the corresponding defect of every kind of grounding grid defect type
Segment library；
The keyword in each defect participle library is extracted respectively, forms defect corresponding with grounding grid defect type described in every kind
Keywords database；
The monitoring data are converted to operation of power networks and describe text by the power network monitoring data for obtaining power grid in operation；
Text is described to the operation of power networks to segment, and obtains operation participle collection；
The keyword that the operation participle is concentrated is extracted, operation keyword set is obtained；
The operation keyword set is matched with each defect keywords database respectively, determines the operation keyword set and every
The matching degree of a defect keywords database；
Determine that the corresponding grounding grid defect type of the highest defect keywords database of matching degree is the defect type of power grid in operation.
2. being lacked the method according to claim 1, wherein the acquisition grounding grid defect describes text according to power grid
It falls into type and text classification is described to the grounding grid defect, comprising:
Obtain power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment checking experiment regulation
Information；
The power grid historic defects description information, power transformation first time equipment deficiency standard information and grid equipment checking experiment are advised
Journey information is converted into grounding grid defect and describes text, and the grounding grid defect describes to include for describing at least one power grid in text
The text of defect；
Text classification is described to the grounding grid defect according to grounding grid defect type, obtains the corresponding power grid of every kind of grounding grid defect type
Defect describes text.
3. the method according to claim 1, wherein the pass extracted in each defect participle library respectively
Keyword, comprising:
For each participle in each defect participle library, word frequency TF of the participle in the participle library is calculated；
For each participle in each defect participle library, defect participle library sum is calculated divided by lacking containing the participle
Participle dictionary number is fallen into, anti-dictionary number frequency is obtained, denary logarithm is taken to the anti-dictionary number frequency, obtains IDF value；
The word frequency TF is multiplied with the IDF value, obtains TF-IDF value；
The participle for determining that the TF-IDF value meets preset threshold is the keyword.
4. the method according to claim 1, wherein the keyword for extracting operation participle and concentrating, comprising:
For each participle that the operation participle is concentrated, the word frequency TF ' that the participle is concentrated in the participle is calculated；
For each participle that the operation participle is concentrated, the defect participle library sum is calculated divided by lacking containing the participle
Participle library number is fallen into, anti-dictionary number frequency is obtained, denary logarithm is taken to the anti-dictionary number frequency, obtains IDF ' value；
The word frequency TF ' is multiplied with the IDF ' value, obtains TF '-IDF ' value；
The participle for determining that the TF '-IDF ' value meets preset threshold is the keyword.
5. the method according to claim 1, wherein described that operation keyword set is crucial with each defect respectively
Dictionary is matched, and determines the matching degree of the operation keyword set and each defect keywords database, comprising:
Keyword in the operation keyword set is compared with the keyword in each defect keywords database respectively,
Determine the quantity of same keyword in the operation keyword set and the defect keywords database；
According to the quantity of the same keyword, the matching degree of the operation keyword set and the defect keywords database is determined.
6. the method according to claim 1, wherein described that operation keyword set is crucial with each defect respectively
Dictionary is matched, and determines the matching degree of the operation keyword set and each defect keywords database, comprising:
Keyword in the operation keyword set is compared with the keyword in each defect keywords database respectively,
It determines the quantity of same keyword in the operation keyword set and the defect keywords database, and determines that the operation is crucial
The quantity of close keyword in word set and the defect keywords database；
According to the quantity of the quantity of the same keyword and the close keyword, determine the operation keyword set with it is described
The matching degree of defect keywords database.
7. a kind of judgment means of grounding grid defect type characterized by comprising
Defective data module is obtained, text is described for obtaining grounding grid defect, according to grounding grid defect type to the grounding grid defect
Text classification is described；
Word segmentation module is segmented for describing text to grounding grid defect described in every kind respectively, obtains every kind of grounding grid defect
The corresponding defect of type segments library；
Keyword extracting module is formed and every kind of electricity for extracting the keyword in each defect participle library respectively
The corresponding defect keywords database of net defect type；
Monitoring data module is obtained, for obtaining running power network monitoring data, the monitoring data are converted into power grid fortune
Row description text；
The word segmentation module is also used to describe text to the operation of power networks to segment, and obtains operation participle collection；
The keyword extracting module is also used to extract the keyword that the operation participle is concentrated, obtains operation keyword set；
Keywords matching module is determined for matching the operation keyword set with each defect keywords database respectively
The matching degree of the operation keyword set and each defect keywords database；
Defect type judgment module, for determining that the corresponding grounding grid defect type of the highest defect keywords database of matching degree is
The defect type of power grid in operation.
8. device according to claim 7, which is characterized in that the acquisition defective data module, comprising:
Acquisition submodule, for obtaining power grid historic defects description information, power transformation first time equipment deficiency standard information and power grid
Overhaul of the equipments testing regulations information；
Data conversion submodule is used for the power grid historic defects description information, power transformation first time equipment deficiency standard information
And grid equipment checking experiment protocol information, be converted into grounding grid defect and describe text, the grounding grid defect describe include in text
For describing the text of at least one grounding grid defect；
Submodule of classifying obtains every kind of power grid and lacks for describing text classification to the grounding grid defect according to grounding grid defect type
It falls into the corresponding grounding grid defect of type and describes text.
9. device according to claim 7, which is characterized in that keyword extracting module, comprising:
Word frequency TF computational submodule, for calculating the participle in institute for each participle in each defect participle library
State the word frequency TF in participle library；
IDF computational submodule, for for each participle in each defect participle library, calculating total dictionary number divided by containing
The dictionary number of the participle obtains anti-dictionary number frequency, takes denary logarithm to the anti-dictionary number frequency, obtains IDF value；
TF-IDF computational submodule obtains TF-IDF value for the word frequency TF to be multiplied with the IDF value；
Keyword determines submodule, and the participle for determining that the TF-IDF value meets preset threshold is the keyword.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910624829.1A CN110321425B (en) | 2019-07-11 | 2019-07-11 | Method and device for judging defect type of power grid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910624829.1A CN110321425B (en) | 2019-07-11 | 2019-07-11 | Method and device for judging defect type of power grid |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110321425A true CN110321425A (en) | 2019-10-11 |
| CN110321425B CN110321425B (en) | 2023-07-21 |
Family
ID=68121837
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910624829.1A Active CN110321425B (en) | 2019-07-11 | 2019-07-11 | Method and device for judging defect type of power grid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110321425B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113377759A (en) * | 2021-07-01 | 2021-09-10 | 贵州电网有限责任公司 | Defect filling data management method based on expert system algorithm |
| CN113850077A (en) * | 2021-09-27 | 2021-12-28 | 未鲲(上海)科技服务有限公司 | Topic identification method, device, server and medium based on artificial intelligence |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837770A (en) * | 2014-02-26 | 2014-06-04 | 广东电网公司电力科学研究院 | Electrical equipment defect detection and maintenance method |
| CN104463708A (en) * | 2014-12-15 | 2015-03-25 | 广东电网有限责任公司电力科学研究院 | Power equipment defect fast grading method |
| CN105844416A (en) * | 2016-03-28 | 2016-08-10 | 国网江苏省电力公司电力科学研究院 | Integrated date management system for analyzing fault defects of substation equipment |
| CN105955960A (en) * | 2016-05-06 | 2016-09-21 | 浙江大学 | Semantic frame-based power grid defect text mining method |
| CN108182523A (en) * | 2017-12-26 | 2018-06-19 | 新疆金风科技股份有限公司 | The treating method and apparatus of fault data, computer readable storage medium |
| CN108664538A (en) * | 2017-11-30 | 2018-10-16 | 全球能源互联网研究院有限公司 | A kind of automatic identification method and system of the doubtful familial defect of power transmission and transforming equipment |
| CN108874984A (en) * | 2018-06-11 | 2018-11-23 | 浙江大学 | A kind of increased quality method to second-rate grid equipment defect text |
-
2019
- 2019-07-11 CN CN201910624829.1A patent/CN110321425B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103837770A (en) * | 2014-02-26 | 2014-06-04 | 广东电网公司电力科学研究院 | Electrical equipment defect detection and maintenance method |
| CN104463708A (en) * | 2014-12-15 | 2015-03-25 | 广东电网有限责任公司电力科学研究院 | Power equipment defect fast grading method |
| CN105844416A (en) * | 2016-03-28 | 2016-08-10 | 国网江苏省电力公司电力科学研究院 | Integrated date management system for analyzing fault defects of substation equipment |
| CN105955960A (en) * | 2016-05-06 | 2016-09-21 | 浙江大学 | Semantic frame-based power grid defect text mining method |
| CN108664538A (en) * | 2017-11-30 | 2018-10-16 | 全球能源互联网研究院有限公司 | A kind of automatic identification method and system of the doubtful familial defect of power transmission and transforming equipment |
| CN108182523A (en) * | 2017-12-26 | 2018-06-19 | 新疆金风科技股份有限公司 | The treating method and apparatus of fault data, computer readable storage medium |
| CN108874984A (en) * | 2018-06-11 | 2018-11-23 | 浙江大学 | A kind of increased quality method to second-rate grid equipment defect text |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113377759A (en) * | 2021-07-01 | 2021-09-10 | 贵州电网有限责任公司 | Defect filling data management method based on expert system algorithm |
| CN113850077A (en) * | 2021-09-27 | 2021-12-28 | 未鲲(上海)科技服务有限公司 | Topic identification method, device, server and medium based on artificial intelligence |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110321425B (en) | 2023-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109359271B (en) | Transformer winding deformation degree online detection method based on logistic regression | |
| CN110319982B (en) | Buried gas pipeline leakage judgment method based on machine learning | |
| CN109684673B (en) | Feature extraction and cluster analysis method for transient stability result of power system | |
| CN105279365A (en) | Method for learning exemplars for anomaly detection | |
| WO2020252574A1 (en) | Artificial intelligence-based process and system for visual inspection of infrastructure | |
| CN108562821B (en) | Method and system for determining single-phase earth fault line selection of power distribution network based on Softmax | |
| CN108304567B (en) | Method and system for identifying working condition mode and classifying data of high-voltage transformer | |
| CN101738998B (en) | System and method for monitoring industrial process based on local discriminatory analysis | |
| CN110601173B (en) | Distribution network topology identification method and device based on edge calculation | |
| CN110321425A (en) | A kind of judgment method and device of grounding grid defect type | |
| CN110334816A (en) | A kind of industrial equipment detection method, device, equipment and readable storage medium storing program for executing | |
| CN115936658A (en) | Power equipment abnormality detection method, system and readable storage medium | |
| CN110083542A (en) | Model test Method, device and electronic equipment in a kind of recommender system | |
| CN110197475A (en) | Insulator automatic recognition system, method and application in a kind of transmission line of electricity | |
| CN105279315B (en) | A kind of transformer online monitoring information fusion analysis method based on correlation analysis and mahalanobis distance | |
| CN114694130A (en) | Method and device for detecting telegraph poles and pole numbers along railway based on deep learning | |
| CN114441463A (en) | Full-spectrum water quality data analysis method | |
| CN111737993B (en) | Method for extracting equipment health state from fault defect text of power distribution network equipment | |
| CN116453437B (en) | Display screen module testing method, device, equipment and storage medium | |
| CN107328994A (en) | Insulation Resistance of Transformer experimental rig and method | |
| CN116975389A (en) | Transformer substation fault analysis method and device | |
| CN105741184A (en) | Transformer state evaluation method and apparatus | |
| CN115904955A (en) | Performance index diagnosis method and device, terminal equipment and storage medium | |
| Zhang et al. | Determining statistical process control baseline periods in long historical data streams | |
| CN112732773A (en) | Uniqueness checking method and system for relay protection defect data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |