CN113886469B - Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity - Google Patents
Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity Download PDFInfo
- Publication number
- CN113886469B CN113886469B CN202111461264.3A CN202111461264A CN113886469B CN 113886469 B CN113886469 B CN 113886469B CN 202111461264 A CN202111461264 A CN 202111461264A CN 113886469 B CN113886469 B CN 113886469B
- Authority
- CN
- China
- Prior art keywords
- project
- engineering
- equipment
- library
- matching
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 230000000694 effects Effects 0.000 title claims abstract description 25
- 238000005065 mining Methods 0.000 title claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims abstract description 24
- 238000013507 mapping Methods 0.000 claims abstract description 17
- 238000002372 labelling Methods 0.000 claims abstract description 7
- 238000010276 construction Methods 0.000 claims description 23
- 230000000875 corresponding effect Effects 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000000354 decomposition reaction Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000011835 investigation Methods 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000003058 natural language processing Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000002596 correlated effect Effects 0.000 claims description 3
- 230000001788 irregular Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000013179 statistical model Methods 0.000 claims description 2
- 238000009412 basement excavation Methods 0.000 abstract description 2
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000009466 transformation Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2468—Fuzzy queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
- G06F16/90344—Query processing by using string matching techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention belongs to the technical field of power engineering, and relates to a method and a system for automatically mining a power distribution network engineering effect abnormity based on multi-source data, which comprises the steps of collecting power distribution network engineering data, equipment ledger information and problem events; extracting equipment information according to the project name in the project data, carrying out fuzzy matching with an actual equipment ledger, and establishing an incidence relation between the project and the problem by taking the equipment as a link; establishing a mapping relation between project categories and problem types, realizing one-to-one correspondence between specific projects and problems through project labeling, taking the intersection of all problems occurring before and after the project is implemented and the problems with the corresponding relation, obtaining the associated problems before and after the project is implemented, calculating the variable quantity of the problems, and studying and judging the project with abnormal effect; comparing the equipment ledger condition after the engineering is implemented with the engineering quantity, and studying and judging the project operation condition; the automatic excavation of the project effect abnormity of the power distribution network is realized through study and judgment of the project effect abnormity and the operation condition.
Description
Technical Field
The invention belongs to the technical field of power distribution and utilization, and particularly relates to the technical field of evaluation after construction success of power distribution network engineering, in particular to a method and a system for automatically mining power distribution network engineering performance abnormity based on multi-source data.
Background
The power distribution network construction and transformation project has the characteristics of small investment amount of single projects and large integral volume, the whole project construction effect is difficult to evaluate after the project construction is implemented, and a mode of combining integral data and typical project evaluation is often adopted. However, the existing method can only analyze the overall effect macroscopically, and the project with abnormal effect cannot be found out in the evaluation process, so that a lot of engineering construction effects still do not reach the expectation, and the investment is wasted.
At the present stage, because the project grading power grid project construction effect evaluation can not be carried out and the project implementation problem can not be found actively, partial projects have the phenomena of poor construction process, repeated establishment, false completion and the like, and the investment effect is not expected. The main problems in the whole engineering result evaluation are that the data source is complex, the authenticity and objectivity of manual collection are difficult to guarantee, system data cannot be interconnected and communicated, and various data are difficult to correspond to engineering.
In order to improve the current situation that the evaluation mode of the power distribution network engineering construction success is rough, dig the problems existing in the engineering construction, effectively guide the next work to be carried out and reduce the low-efficiency investment, the method and the system for automatically digging the abnormal power distribution network engineering success of the multi-source data provided by the invention utilize big data means based on the data stored in each business system and by comparing the problem solution conditions before and after the engineering is implemented, dig the problems existing in the construction process and actively judge whether the engineering has the abnormal construction success, thereby realizing the scanning of the entire engineering construction success.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides a method and a system for mining the abnormal success of the power distribution network engineering based on multi-source data, manual offline reporting is not needed, the online data based on PMS2.0, a power distribution automation system, a power utilization information acquisition system, an intelligent power supply service command system and the like automatically complete a fund collection link, multi-source unstructured data processing and association are carried out by using a big data means and a natural language processing technology, abnormal phenomena such as unsolved problems after engineering construction, unfinished project operation and the like are actively researched and judged, and the intelligent and automatic mining of the abnormal success of engineering implementation is realized.
The invention is realized by the following technical scheme. A multi-source data-based automatic mining method for power distribution network engineering performance abnormity comprises the following specific steps:
s1, data acquisition: acquiring project data issued by a plan from a distribution network engineering management and control system, and storing the project data in a project library; acquiring equipment ledger information from an equipment asset operation and maintenance lean management system, acquiring line switch ledgers and online conditions from a power distribution automation system, acquiring voltage and current data of a transformer area from a power utilization information acquisition system, and storing the voltage and current data in an equipment library; acquiring problem events from an intelligent power supply service command system, and storing the problem events in a problem library;
s2, multi-source data association based on equipment: extracting equipment names from project name fields stored in a project library by using a regular matching method, and decomposing a project into a specific 10kV line and a specific platform area; matching the equipment name in the project library with the account of the equipment library by using a character string fuzzy matching method, and storing account equipment successfully matched with the project into the project library; the account equipment stored in the project library is correlated to the problem events stored in the problem library, and the problem types s are respectively counted according to the year before and the year after the project is implemented1、s2、s3... sj...snAnd the number of times, where n is the total number of problem types, sjThe question is of the jth question type and is stored in a project library;
s3, engineering labeling: classifying all projects according to the category of newly-built and improved equipment to obtain the project category r1、r2、r3... ri...rmM is the total number of engineering categories, riEstablishing corresponding keyword sequences for the ith engineering category and traversing each engineering categoryAll projects in the project library carry out keyword retrieval on project implementation contents by utilizing a regular matching method, and if keywords are matched, the project corresponds to the project category riA label of (1), then ri=1, otherwise ri=0, r in each projectiEngineering class r of =1iThe corresponding project category labels are stored in a project library;
s4, establishing a mapping relation between the engineering and the problem: setting up a project class set { r1、r2、r3... ri...rm}; classifying all the question types to obtain a question type set { s }1、s2、s3...sj...sn}; establishing a two-dimensional mapping table, wherein the row label is an engineering class label, the column label is a problem type label, and if the engineering class r isiCorresponding construction content capable of solving problem type sjThen the mapping relation (r)i,sj) =1, otherwise (r)i,sj) = 0; all questions before and after the project is performed have been obtained in step S2, if there is a question type S before or after the project is performedjI.e. sjIf not, deleting the problem type label in the project; if the project type labels are obtained in step S3, the mapping relationship between the ith project type label and the jth question type label is present, that is, (r)i,sj) =1, then the problem type s is retainedjElse delete problem type sjTraversing all projects, and storing the problems before and after implementation obtained by calculation into a project library;
s5, engineering quantity decomposition: extracting the engineering quantity in the engineering content by using a regular method, decomposing the engineering quantity into a line, a platform area and a switch, and converting the engineering quantity into structured data; comparing the engineering quantity decomposition result of each project with the current situation information of the equipment library, studying and judging the equipment commissioning result, and storing the equipment commissioning result in the project library;
s6, abnormal effect mining: in step S4, the problem types before and after the project is performed are obtained, and it is assumed that the problem type set before the project is performed is Σ Si(i=1,2,…n1)，n1To implement the total number of problem types before, set of problem types afterIs Σ sj(j=1,2,…n2)，n2To implement the total number of problem types, the set of problem types to be solved by the project should be Δ s = Σ si-ΣsjLet the number of the problem types in the problem type set solved by the project be a, if a>If a is less than or equal to 0, the engineering effect is abnormal; in step S5, a result of the investigation and judgment of the operation of the equipment is obtained, if the result of the investigation and judgment is "yes", the equipment is normally operated, and if the result of the investigation and judgment is "no", the equipment is suspected to be not operated; and (4) combining the two research and judgment results, analyzing the causes of the abnormal engineering effect, and feeding back to the owner unit for checking.
Further preferably, the specific process of step 2 is as follows:
s21, item library data normalization: project establishment data are stored in the project library, project naming is subjected to standardization processing, and characters with irregular naming are replaced with standard contents according to project classification;
s22, extracting the name of the project library equipment: and (3) dividing the line and the transformer area part through a first character line from a normalized project name field, extracting equipment names by respectively utilizing a regular matching method, decomposing the project into a specific 10kV line and a specific transformer area, and storing the specific 10kV line and the specific transformer area in a project library.
S23, fuzzy matching of the project library and the equipment library: matching the equipment names in the project library with the ledgers of the equipment library one by one according to the county company-line-district sequence;
s24, associating a question bank: and associating the line and the platform area equipment obtained by calculation in the project library with the question library, counting all the types and times of the problems of the equipment in the previous year of engineering implementation and the next year of engineering implementation according to the problem events in the question library, and storing the types and times into the project library.
Further preferably, in step S23, the route matching process is as follows:
s231, extracting an engineering project information table from the project library, extracting an equipment ledger information table from the equipment library, grouping the information tables according to county companies, and reserving relevant fields of engineering names, lines and areas;
s232, matching lines of the same county company, and removing project establishment information tablesThe fixed character of the middle line name only retains the character string directly describing the corresponding equipment as the input character string xliWhere li is the index of the input line, li ∈ L1，L1Grouping a set of lower lines for the county company in the project establishment information table;
s233, taking the equipment library line account information as an index character string yljGo through the index string yljLj is the index of line account, lj belongs to L2，L2For the set of line ledger information grouped down by the county company, x is calculatedliAnd yljThe matching degree score is calculated according to the ratio of the editing distance to the length of the input character string, and a non-complete matching mode is adopted, namely, if xli∈yljIf yes, the matching degree score is 100;
s234, determining a threshold k according to the actual matching condition1Returning a matching score cljHighest index string yljIf the character string matches the score clj≥k1Then finish the line matching and return to yljAs input string xliThe matching result of (1);
and S235, traversing all county company groups, returning the matching results of the line parts of all projects, and storing the matching results in a project library.
Further preferably, in step S23, the station area matching process is as follows:
s236, if the platform area project is the platform area project, namely the project information contains platform area related fields, continuing to further match the platform areas matched with the lines, removing fixed characters of the platform area names in the project item information table, and only reserving character strings directly describing corresponding equipment as input character strings qtiTi is the index of the input station area, ti ∈ T1，T1Grouping a set of the next district for the county company in the project establishment information table;
s237, taking account information of the equipment library area as an index character string wtjTj is the index of the table ledger of the platform area, and tj belongs to T2，T2Grouping sets of standing book information of the next district for the county company when q istiAnd wtjWhen the lines are the same, calculating qtiAnd wtjThe editing distance between the two characters, and calculating the matching degree score according to the editing distance in the ratio of the length of the input character string, and adopting a non-complete matching mode, namely if q is equal to qti∈wtjIf yes, the matching degree score is 100;
s238, determining a threshold k according to the actual matching condition2Returning a matching score etjHighest index string wtjIf the character string matches the score etj≥k2Then completing the region matching and returning to wtjAs an input character string qtiThe matching result of (1);
and S239, traversing all county company groups, returning the matching results of all the platform area projects, and storing the matching results in a project library.
The invention also provides a power distribution network engineering abnormal effect automatic mining system based on the multi-source data, which comprises the following steps: the system comprises a data acquisition module, a project library module, a problem library module, a equipment library module and a result output module; the data acquisition module is used for collecting engineering information, problem information and equipment ledger information required by evaluation and respectively pushing the engineering information, the problem information and the equipment ledger information to the project library module, the problem library module and the equipment library module according to the information source and the type; the problem library module is used for establishing a problem statistical model and classifying and counting problem events; the equipment library module is used for establishing an equipment matching model, matching the engineering equipment, correlating various information through the equipment and pushing an engineering equipment matching result and a problem correlation result to the project library module; the project library module is used for establishing an engineering analysis model, analyzing engineering equipment and engineering quantity by using a natural language processing method through an engineering name and an engineering content field, calculating problem variable quantity before and after engineering implementation, judging engineering success and judging the real project condition of the engineering; and the result output module is used for outputting the project effect and the operation condition research and judgment result.
The invention provides a method and a system for automatically mining a power distribution network project effect abnormity based on multi-source data. Extracting equipment information according to the project name in the project data, decomposing the project into a 10kV line and a station area, carrying out fuzzy matching with an actual equipment account, establishing an incidence relation between the project and the problem by taking the equipment as a link, and counting the type and the number of the problem before and after the project is implemented; establishing a mapping relation between project categories and problem types, realizing one-to-one correspondence between specific projects and problems through project labeling, taking the intersection of all problems occurring before and after the project is implemented and the problems with the corresponding relation, obtaining the associated problems before and after the project is implemented, calculating the variable quantity of the problems, and studying and judging the project with abnormal effect; the method comprises the steps of utilizing a natural language processing method to realize structured decomposition of engineering quantity, comparing the equipment ledger condition after engineering implementation with the engineering quantity, and studying and judging the project operation condition; the automatic excavation of the project effect abnormity of the power distribution network is realized through study and judgment of the project effect abnormity and the operation condition.
The invention solves the problems that data of power distribution network engineering, problems and the like are multi-source and cannot be associated, and the engineering effect analysis is transferred from off-line to on-line, thereby ensuring the authenticity, accuracy and real-time of the data, being beneficial to reducing the workload of personnel for filling tables, greatly improving the analysis efficiency and realizing the effect analysis of the whole engineering.
Drawings
FIG. 1 is a flow chart of a power distribution network engineering achievement anomaly mining method based on multi-source data.
FIG. 2 is a framework diagram of a power distribution network engineering performance anomaly mining system based on multi-source data.
In the figure: 10-a data acquisition module, 20-an item library module, 30-a device library module, 40-a problem library module and 50-a result output module.
Detailed Description
The invention is explained in more detail below with reference to the figures and examples.
A multi-source data-based automatic mining method for power distribution network engineering performance abnormity comprises the following specific steps:
s1. data acquisition
And S11, acquiring standing and commissioning data of the evaluation object from the distribution network engineering management and control system, and storing the standing and commissioning data in a project library.
And S12, acquiring the machine account information of main equipment such as a 10kV line, a distribution transformer, a switch and the like from an equipment asset operation and maintenance lean management system (PMS 2.0 system), acquiring the DTU, the FTU machine account and the online condition of the line from a power distribution automation system, acquiring the voltage and current data of a transformer area from a power utilization information acquisition system, and storing the voltage and current data in an equipment library.
And S13, obtaining problem events such as line tripping, line heavy overload, power failure complaints, district outlet low voltage, district heavy overload, user low voltage complaints and the like from the intelligent power supply service command system, and storing the problem events in a problem library.
S2, multi-source data association based on equipment
S21, item library data normalization: project establishment data are stored in a project library, the standard names are usually 10kVxx line new construction (modification) projects, 10kVxx line xx branch line new construction (modification) projects, 10kVxx line xx bench pole upper switch new construction (modification) projects, 10kVxx line xx bench area low-voltage line new construction (modification) projects and the like, the project names are subjected to standardization processing, and characters with irregular names are replaced by standard contents according to project classification.
S22, extracting the name of the project library equipment: and (3) dividing the line and the platform area part through a first character line from a normalized project name field, extracting the equipment name by respectively using a regular matching method, decomposing the project into a 10kVxx line and an xx platform area, and storing the 10kVxx line and the xx platform area in a project library.
S221. example 1: the name of a normalized project is '10 kV Gong's generation line upper transformer area low-voltage line transformation project ', the line and the transformer area in the project name are partially decomposed by a first character' line ', and the project name is' 10kV Gong's generation line' and 'board upper transformer area low-voltage line transformation project'. The line name of 10kV is obtained by matching the line part with a regular method of 10kV.
S222, example 2: a normalized project name is 10kV urban south line 04B transformer area reconstruction project, and lines and transformer areas in the project name are partially decomposed by a first character line, namely 10kV urban south line and 04B transformer area. The method comprises the steps of matching a line part by a regular method of ' 10kV '. line ' to obtain a line name ' 10kV urban south line ', judging a station area part by a regular method of ' line # \ d. station area ' to be a number with a first character, matching the number with a project name to obtain a station area part, and integrating the station area part and the line part to obtain a station area name ' 10kV urban south line 04B station area '.
S23, fuzzy matching of the project library and the equipment library: and matching the equipment names in the project library with the ledgers of the equipment library one by one according to the county company-line-district sequence.
The line matching process is as follows:
s231, extracting an engineering project information table from the project library, extracting an equipment ledger information table from the equipment library, grouping the information tables according to county companies, and reserving relevant fields of engineering names, lines and areas;
s232, matching the lines of the same county company, removing fixed characters such as '10 kV' and 'lines' of the line name in the project item information table, and only keeping the character string directly describing the corresponding equipment as an input character string xliWhere li is the index of the input line, li ∈ L1，L1Grouping a set of lower lines for the county company in the project establishment information table;
s233, taking the equipment library line account information as an index character string yljGo through the index string yljLj is the index of line account, lj belongs to L2，L2For the set of line ledger information grouped down by the county company, x is calculatedliAnd yljThe edit Distance between the two characters is calculated, the matching degree score is calculated according to the edit Distance in the length ratio of the input character string, the edit Distance is calculated by adopting Levenshtein Distance, and a non-complete matching mode is adopted in the process, namely if xli∈yljIf yes, the matching degree score is 100;
s234, determining a threshold k according to the actual matching condition1Returning a matching score cljHighest index string yljIf the character string matches the score clj≥k1Then finish the line matching and return to yljAs input string xliThe matching result of (1);
s234, determining a threshold k according to the actual matching condition1Returning a matching score cjHighest index string yjIf the character string matches the score cj≥k1Then finish the line matching and return to yjAs input string xiThe matching result of (1);
and S235, traversing all county company groups, returning the matching results of the line parts of all projects, and storing the matching results in a project library.
The platform area matching process is as follows:
s236, if the project is a platform area project, namely project information contains platform area related fields, further matching the platform areas matched with the lines is continued, fixed characters such as 'platform area', 'distribution transformation' and the like of the platform area names in the project item information table are removed, only character strings directly describing corresponding equipment are reserved and used as input character strings qtiTi is the index of the input station area, ti ∈ T1，T1Grouping a set of the next district for the county company in the project establishment information table;
s237, taking account information of the equipment library area as an index character string wtjTj is the index of the table ledger of the platform area, and tj belongs to T2，T2Grouping sets of standing book information of the next district for the county company when q istiAnd wtjWhen the lines are the same, calculating qtiAnd wtjThe edit Distance between the two characters is calculated, the matching degree score is calculated according to the edit Distance in the length ratio of the input character string, the edit Distance is calculated by adopting Levenshtein Distance, a non-complete matching mode is adopted in the process, namely if q is the caseti∈wtjIf yes, the matching degree score is 100;
s238, determining a threshold k according to the actual matching condition2Returning a matching score etjHighest index string wtjIf the character string matches the score etj≥k2Then completing the region matching and returning to wtjAs an input character string qtiThe matching result of (1); (ii) a
And S239, traversing all county company groups, returning the matching results of all the platform area projects, and storing the matching results in a project library.
S24. correlationQuestion bank: the line and platform area equipment obtained by calculation in the project library is associated with the question library, and all the problem types s of the equipment in the previous year of engineering implementation and the next year of engineering implementation are counted according to the problem events in the equipment statistics question library1、s2、s3... sj...snAnd the number of times, where n is the total number of problem types, sjAnd storing the problem in the project library for the jth problem type.
S3, engineering labeling
S31, engineering classification: respectively classifying the line and the platform area engineering according to the category of newly-built and improved equipment to obtain the engineering category r1、r2、r3... ri...rmM is the total number of engineering categories, riIs the i-th engineering category.
S32, establishing a keyword sequence: for each project category, a corresponding keyword sequence is established, for example, a switch replacement project, and the keywords may be set to [ 'new/replacement/modification', '(on-pole) switch/(on-pole) breaker/ring main unit' ].
S33, keyword search: traversing all projects in the project library, taking out project contents in project establishment data, performing keyword retrieval on project implementation contents by using a regular matching method, and if keywords are matched, corresponding to the project category riA label of (1), then ri=1, otherwise ri=0, r in each projectiEngineering class r of =1iIs stored in the project library.
Example (c): setting a line engineering class r1 If the project content of the project 1 is 'newly reconstructed 10kV line 3km and pole-mounted switch 1 is replaced', the project content of the project 2 is 'reconstructed 10kV line 1.2km and LGJ-70 conducting wire is adopted', the label matching result of the project 1 is judged to be r1=1, label matching result of engineering 2 is judged as r1=0。
S34, typical line engineering categories comprise newly-built 10kV overhead lines, 10kV cable lines, switches, ring main units, lightning arresters and the like, and typical station engineering categories comprise newly-built distribution transformers, replacement distribution transformers, newly-built and modified low-voltage lines, replacement low-voltage cable branch boxes, transfer station areas, JP cabinets and the like.
And S35, completing labeling of all projects, and storing project category labels into a project library.
S4, establishing a mapping relation between engineering and problems
S41, engineering and problem classification: establishing a project category set { r1、r2、r3... ri...rmThe classification mode is the same as the project labeling, and the line and platform project is typically classified as the step S3; classifying all questions in the question bank according to lines and transformer areas respectively to obtain a question type set { s }1、s2、s3... sj...snN is the total number of problem types; typical line problems include main line tripping, branch line shutdown, frequent power failure complaints, line overload and the like, and typical transformer area problems include distribution transformer outlet low voltage, high loss distribution transformer, distribution transformer overload, user low voltage, low voltage complaints and the like.
S42, establishing a two-dimensional mapping table, wherein the row label is an engineering type label, the column label is a problem type label, and if the ith type of engineering riThe corresponding construction content can solve the corresponding jth problem type sjThen the mapping relation (r)i,sj) =1, otherwise (r)i,sj) And = 0. The specific mapping relationships are shown in tables 1 and 2.
TABLE 1
TABLE 2
S43, all problem types before and after the engineering is implemented are obtained in the step S2, and if the problem types S exist before or after the engineering is implementedjI.e. sj=1, then the worker is in the same placeKeeping the problem type label in the process, otherwise deleting the problem type label in the project; the project category is obtained in step S3, and the i-th project category label has a mapping relation with the j-th problem type label, that is, (r)i,sj) =1, then the problem type s is retainedjElse delete problem type sj。
And S44, traversing all projects, and storing the correlation problem obtained by calculation before and after implementation into a project library.
S5, decomposing engineering quantity
And S51, determining the engineering quantity target to be extracted, such as main equipment of lines, transformer areas, switches, lightning arresters and the like.
S52, aiming at each type of equipment in the extracted target, a keyword set describing the type of equipment is established, corresponding keywords are matched by a regular method, the engineering quantity of the equipment in the engineering content is extracted, the engineering quantity is decomposed into the equipment, the structured data is converted into and stored in a project library. The engineering quantity extracted by the line engineering should include the line model and length, the number of switches, the number of lightning arresters and the like, and the engineering quantity extracted by the district engineering should include the distribution transformer model and capacity and the like.
S53, comparing the project quantity decomposition result of each project with the current state information of the project associated equipment in the equipment library, if the project quantity decomposition result of each project is consistent with the current state information of the project associated equipment in the equipment library, indicating that the equipment is put into operation and updated in the system, judging that the equipment operation result is yes, otherwise judging that the equipment operation result is no, and storing the judgment result in the project library.
S6, abnormal effect mining
S61, in step S4, the problem types related to the project before and after the project is implemented are obtained, and the problem set before the project is assumed to be Σ Si(i=1,2,…n1)，n1To implement the total number of question types before, the problem set after implementation is Σ sj(j=1,2,…n2)，n2To implement the total number of problem types, the set of problem types to be solved by the engineering is Δ s = Σ si-ΣsjLet the number of the problem types in the problem type set solved by the project be a, if a>If a is less than or equal to 0, the engineering effect is abnormal.
S62, obtaining the equipment commissioning judgment result in the step S5, if the equipment commissioning judgment result is 'yes', the equipment is normally commissioned, and if the equipment commissioning judgment result is 'no', the equipment is suspected to be not commissioned.
S63, aiming at the project with abnormal effect and a no operation research and judgment result, indicating that the project is not operated to cause poor effect, feeding back to an owner unit for checking, and taking the result as the basis of an evaluation construction unit; aiming at the project with abnormal effect and the operation, study and judgment result of 'yes', the project implementation quality is poor or the design is unreasonable, the result is fed back to an owner unit for checking, and the result is used as the basis of evaluation construction and design units; aiming at the condition that the project has success and the operation and study judgment result is 'no', the result is fed back to the owner unit for checking and is used as the basis of evaluating the construction unit; aiming at the condition that the project has success and the operation and study result is 'yes', the method does not need manual check and can be used as evaluation materials to be provided for the owner.
In order to realize the method, the invention provides an automatic mining system for abnormal effect of power distribution network engineering based on multi-source data, as shown in fig. 2, the system comprises a data acquisition module 10, a project library module 20, a device library module 30, a question library module 40 and a result output module 50.
The data acquisition module 10 is used for collecting engineering information, problem information and equipment ledger information required by evaluation, and respectively pushes the engineering information, the problem information and the equipment ledger information to the project library module 20, the problem library module 40 and the equipment library module 30 according to information sources and types.
The question bank module 40 is used for establishing a question statistic model, and classifying and counting the question events.
The project library module 20 establishes a device extraction model based on the project name, decomposes the project into a line and a platform area, and inputs the result into the device library module 30; establishing an equipment matching model in an equipment library, carrying out fuzzy matching on input equipment and stored equipment standing book information, and returning a result to a project library; the device matching result is correlated to the question bank module 40, information such as the type and the times of the problems on the device before and after engineering implementation is counted, and the result is stored in the project bank.
The project library module 20 establishes a project-problem mapping model, projects all problems before and after the project is implemented into the project-problem mapping model after the project content is labeled, obtains the problems related to the project implementation content, further calculates the problem solving conditions of the project implementation, and studies and judges the project implementation effect.
The project library module 20 establishes a project quantity decomposition model, converts the project content into the structured project quantity, compares the structured project quantity with the current equipment ledger in the equipment library module 30, calculates whether the project is finished and put into operation according to the project quantity, and studies and judges the project operation result.
The result output module 50 is used for outputting the project effect and the commissioning situation research and judgment result.
Claims (5)
1. A multi-source data-based automatic mining method for power distribution network engineering performance abnormity is characterized by comprising the following specific steps:
s1, data acquisition: acquiring project data issued by a plan from a distribution network engineering management and control system, and storing the project data in a project library; acquiring equipment ledger information from an equipment asset operation and maintenance lean management system, acquiring line switch ledgers and online conditions from a power distribution automation system, acquiring voltage and current data of a transformer area from a power utilization information acquisition system, and storing the voltage and current data in an equipment library; acquiring problem events from an intelligent power supply service command system, and storing the problem events in a problem library;
s2, multi-source data association based on equipment: extracting equipment names from project name fields stored in a project library by using a regular matching method, and decomposing a project into a specific 10kV line and a specific platform area; matching the equipment name in the project library with the account of the equipment library by using a character string fuzzy matching method, and storing account equipment successfully matched with the project into the project library; the account equipment stored in the project library is correlated to the problem events stored in the problem library, and the problem types s are respectively counted according to the year before and the year after the project is implemented1、s2、s3... sj...snAnd the number of times, where n is the total number of problem types, sjThe question is of the jth question type and is stored in a project library;
s3. the workerProgram labeling: classifying all projects according to the category of newly-built and improved equipment to obtain the project category r1、r2、r3... ri...rmM is the total number of engineering categories, riEstablishing a corresponding keyword sequence for each engineering category for the ith engineering category, traversing all the engineering in the project library, performing keyword retrieval on the engineering implementation content by using a regular matching method, and if the keywords are matched, determining the engineering category r corresponding to the projectiA label of (1), then ri=1, otherwise ri=0, r in each projectiEngineering class r of =1iThe corresponding project category labels are stored in a project library;
s4, establishing a mapping relation between the engineering and the problem: setting up a project class set { r1、r2、r3... ri...rm}; classifying all the question types to obtain a question type set { s }1、s2、s3...sj...sn}; establishing a two-dimensional mapping table, wherein the row label is an engineering class label, the column label is a problem type label, and if the engineering class r isiCorresponding construction content capable of solving problem type sjThen the mapping relation (r)i,sj) =1, otherwise (r)i,sj) = 0; all questions before and after the project is performed have been obtained in step S2, if there is a question type S before or after the project is performedjI.e. sjIf not, deleting the problem type label in the project; if the project type labels are obtained in step S3, the mapping relationship between the ith project type label and the jth question type label is present, that is, (r)i,sj) =1, then the problem type s is retainedjElse delete problem type sjTraversing all projects, and storing the problems before and after implementation obtained by calculation into a project library;
s5, engineering quantity decomposition: extracting the engineering quantity in the engineering content by using a regular method, decomposing the engineering quantity into a line, a platform area and a switch, and converting the engineering quantity into structured data; comparing the engineering quantity decomposition result of each project with the current situation information of the equipment library, studying and judging the equipment commissioning result, and storing the equipment commissioning result in the project library;
s6, abnormal effect mining: in step S4, the problem types before and after the project is performed are obtained, and it is assumed that the problem type set before the project is performed is Σ Si(i=1,2,…n1)，n1To implement the total number of problem types before, the set of problem types after implementation is Σ sj(j=1,2,…n2)，n2To implement the total number of problem types, the set of problem types to be solved by the project should be Δ s = Σ si-ΣsjLet the number of the problem types in the problem type set solved by the project be a, if a>If a is less than or equal to 0, the engineering effect is abnormal; in step S5, a result of the investigation and judgment of the operation of the equipment is obtained, if the result of the investigation and judgment is "yes", the equipment is normally operated, and if the result of the investigation and judgment is "no", the equipment is suspected to be not operated; and (4) combining the two research and judgment results, analyzing the causes of the abnormal engineering effect, and feeding back to the owner unit for checking.
2. The method for automatically mining the abnormal performance of the power distribution network project based on the multi-source data according to claim 1, wherein the specific process of the step 2 is as follows:
s21, item library data normalization: project establishment data are stored in the project library, project naming is subjected to standardization processing, and characters with irregular naming are replaced with standard contents according to project classification;
s22, extracting the name of the project library equipment: dividing lines and transformer areas from a normalized project name field through a first character line, extracting equipment names by using a regular matching method respectively, decomposing the project into specific 10kV lines and transformer areas, and storing the specific 10kV lines and transformer areas in a project library;
s23, fuzzy matching of the project library and the equipment library: matching the equipment names in the project library with the ledgers of the equipment library one by one according to the county company-line-district sequence;
s24, associating a question bank: and associating the line and the platform area equipment obtained by calculation in the project library with the question library, counting all the types and times of the problems of the equipment in the previous year of engineering implementation and the next year of engineering implementation according to the problem events in the question library, and storing the types and times into the project library.
3. The method for automatically mining the abnormal performance of the power distribution network project based on the multi-source data according to claim 2, wherein in the step S23, the line matching process is as follows:
s231, extracting an engineering project information table from the project library, extracting an equipment ledger information table from the equipment library, grouping the information tables according to county companies, and reserving relevant fields of engineering names, lines and areas;
s232, matching the lines of the same county company, removing fixed characters of line names in the project item information table, and only keeping character strings directly describing corresponding equipment as input character strings xliWhere li is the index of the input line, li ∈ L1，L1Grouping a set of lower lines for the county company in the project establishment information table;
s233, taking the equipment library line account information as an index character string yljGo through the index string yljLj is the index of line account, lj belongs to L2，L2For the set of line ledger information grouped down by the county company, x is calculatedliAnd yljThe matching degree score is calculated according to the ratio of the editing distance to the length of the input character string, and a non-complete matching mode is adopted, namely, if xli∈yljIf yes, the matching degree score is 100;
s234, determining a threshold k according to the actual matching condition1Returning a matching score cljHighest index string yljIf the character string matches the score clj≥k1Then finish the line matching and return to yljAs input string xliThe matching result of (1);
and S235, traversing all county company groups, returning the matching results of the line parts of all projects, and storing the matching results in a project library.
4. The method for automatically mining the abnormal performance of the power distribution network project based on the multi-source data according to claim 3, wherein in the step S23, the distribution room matching process is as follows:
s236, if the platform area project is the platform area project, namely the project information contains platform area related fields, continuing to further match the platform areas matched with the lines, removing fixed characters of the platform area names in the project item information table, and only reserving character strings directly describing corresponding equipment as input character strings qtiTi is the index of the input station area, ti ∈ T1，T1Grouping a set of the next district for the county company in the project establishment information table;
s237, taking account information of the equipment library area as an index character string wtjTj is the index of the table ledger of the platform area, and tj belongs to T2，T2Grouping sets of standing book information of the next district for the county company when q istiAnd wtjWhen the lines are the same, calculating qtiAnd wtjThe editing distance between the two characters, and calculating the matching degree score according to the editing distance in the ratio of the length of the input character string, and adopting a non-complete matching mode, namely if q is equal to qti∈wtjIf yes, the matching degree score is 100;
s238, determining a threshold k according to the actual matching condition2Returning a matching score etjHighest index string wtjIf the character string matches the score etj≥k2Then completing the region matching and returning to wtjAs an input character string qtiThe matching result of (1);
and S239, traversing all county company groups, returning the matching results of all the platform area projects, and storing the matching results in a project library.
5. The automatic mining system for the abnormal effect of the power distribution network engineering based on the multi-source data for realizing the method of claim 1 is characterized by comprising a data acquisition module, a project library module, a problem library module, a device library module and a result output module; the data acquisition module is used for collecting engineering information, problem information and equipment ledger information required by evaluation and respectively pushing the engineering information, the problem information and the equipment ledger information to the project library module, the problem library module and the equipment library module according to the information source and the type; the problem library module is used for establishing a problem statistical model and classifying and counting problem events; the equipment library module is used for establishing an equipment matching model, matching the engineering equipment, correlating various information through the equipment and pushing an engineering equipment matching result and a problem correlation result to the project library module; the project library module is used for establishing an engineering analysis model, analyzing engineering equipment and engineering quantity by using a natural language processing method through an engineering name and an engineering content field, calculating problem variable quantity before and after engineering implementation, judging engineering success and judging the real project condition of the engineering; and the result output module is used for outputting the project effect and the operation condition research and judgment result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111461264.3A CN113886469B (en) | 2021-12-03 | 2021-12-03 | Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111461264.3A CN113886469B (en) | 2021-12-03 | 2021-12-03 | Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113886469A CN113886469A (en) | 2022-01-04 |
| CN113886469B true CN113886469B (en) | 2022-04-12 |
Family
ID=79016302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111461264.3A Active CN113886469B (en) | 2021-12-03 | 2021-12-03 | Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113886469B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298056A (en) * | 2019-03-27 | 2019-10-01 | 国网浙江海盐县供电有限公司 | A kind of power distribution network contact efficiency assessment method |
| CN112288240A (en) * | 2020-10-15 | 2021-01-29 | 国网江西省电力有限公司电力科学研究院 | Power distribution network construction project effect evaluation method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104820952B (en) * | 2015-05-12 | 2018-06-15 | 国家电网公司 | Appraisal procedure before a kind of power distribution network investment effect based on two type fuzzy theorys |
| US9785161B2 (en) * | 2015-08-24 | 2017-10-10 | University Of Rochester | Heterogeneous method for energy efficient distribution of on-chip power supplies and power network on-chip system for scalable power delivery |
| US11275768B2 (en) * | 2018-05-25 | 2022-03-15 | Salesforce.Com, Inc. | Differential support for frequent pattern analysis |
| CN110232490A (en) * | 2019-03-07 | 2019-09-13 | 中国电力科学研究院有限公司 | A kind of appraisal procedure and system of distribution network engineering investment effect |
-
2021
- 2021-12-03 CN CN202111461264.3A patent/CN113886469B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298056A (en) * | 2019-03-27 | 2019-10-01 | 国网浙江海盐县供电有限公司 | A kind of power distribution network contact efficiency assessment method |
| CN112288240A (en) * | 2020-10-15 | 2021-01-29 | 国网江西省电力有限公司电力科学研究院 | Power distribution network construction project effect evaluation method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113886469A (en) | 2022-01-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107402976B (en) | Power grid multi-source data fusion method and system based on multi-element heterogeneous model | |
| CN106815373B (en) | Distribution network first-aid repair big data display method and system based on BI analysis | |
| CN113159339A (en) | One-region one-index line loss management method and system based on big data | |
| CN105427039A (en) | Efficient processing method of distribution network repair work orders based on responsibility areas | |
| CN109980642A (en) | A kind of digital intelligent distribution network planning auxiliary system | |
| CN110705873A (en) | Novel power distribution network operation state portrait analysis method | |
| CN111832977A (en) | Maintenance application automatic ticketing method based on natural language parsing | |
| CN114201844A (en) | Hydropower station panoramic modeling method based on IEC61850 standard | |
| CN111241488A (en) | Distribution network protection fixed value setting system based on full information data flow | |
| CN111105218A (en) | Power distribution network operation monitoring method based on holographic image technology | |
| CN113902583A (en) | Distribution network side operation and maintenance method and system using low-voltage network equipment data | |
| CN107742162B (en) | Multidimensional feature association analysis method based on allocation monitoring information | |
| CN113886469B (en) | Multi-source data-based automatic mining method and system for power distribution network engineering effect abnormity | |
| CN113689079A (en) | Transformer area line loss prediction method and system based on multivariate linear regression and cluster analysis | |
| CN116826709A (en) | Automatic matching method, device and storage medium for adjustable load resource and scheduling master station model | |
| CN116108203A (en) | Method, system, storage medium and equipment for constructing power grid panoramic dispatching knowledge graph and managing power grid equipment | |
| CN111340253A (en) | Main network maintenance request form analysis method and system | |
| CN110765986A (en) | Method for automatically auditing monitoring signals of main station of power system | |
| CN115409264A (en) | Power distribution network emergency repair stagnation point position optimization method based on feeder line fault prediction | |
| CN108335231A (en) | A kind of power distribution network data diagnosis method of Auto-matching | |
| CN114677112A (en) | Power distribution network project large-scale evaluation method and system based on big data | |
| CN114662279A (en) | Relay protection information modeling method and system based on secondary equipment big data platform | |
| CN110298585B (en) | Hierarchical automatic auditing method for monitoring information of substation equipment | |
| CN102789603A (en) | Provincial dispatch active regulation method for considering load switch-off model | |
| CN109767062B (en) | Dynamic generation method of power grid task disposal scheme |
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 |