CN116468426A - Power grid fault determining method and storage medium based on Apriori algorithm - Google Patents
Power grid fault determining method and storage medium based on Apriori algorithm Download PDFInfo
- Publication number
- CN116468426A CN116468426A CN202310477952.1A CN202310477952A CN116468426A CN 116468426 A CN116468426 A CN 116468426A CN 202310477952 A CN202310477952 A CN 202310477952A CN 116468426 A CN116468426 A CN 116468426A
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- factor
- acquired
- attribute
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 33
- 238000003745 diagnosis Methods 0.000 claims abstract description 56
- 238000012545 processing Methods 0.000 claims abstract description 45
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000012163 sequencing technique Methods 0.000 claims abstract description 10
- 238000006243 chemical reaction Methods 0.000 claims description 22
- 238000007781 pre-processing Methods 0.000 claims description 12
- 230000001502 supplementing effect Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000005070 sampling Methods 0.000 claims description 7
- 230000007812 deficiency Effects 0.000 claims description 6
- 239000013589 supplement Substances 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 5
- 238000011524 similarity measure Methods 0.000 claims description 5
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 11
- 230000008569 process Effects 0.000 description 10
- 238000004364 calculation method Methods 0.000 description 6
- 238000005065 mining Methods 0.000 description 6
- 102000003712 Complement factor B Human genes 0.000 description 5
- 108090000056 Complement factor B Proteins 0.000 description 5
- 230000009471 action Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 238000007405 data analysis Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 238000012937 correction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000036039 immunity Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013524 data verification Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 238000013138 pruning Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Classifications
-
- 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
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/52—Testing for short-circuits, leakage current or ground faults
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/54—Testing for continuity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/001—Methods to deal with contingencies, e.g. abnormalities, faults or failures
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Power Engineering (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention discloses a power grid fault determining method and a storage medium based on an Apriori algorithm, wherein standardized fault data are input into a pre-constructed fault diagnosis model to obtain a fault type, and the method comprises the following steps: iterative screening is carried out by using an Apriori algorithm and a preset support degree in a fault diagnosis model to obtain a fault factor frequent item set in standardized fault data; calculating preset confidence in the fault diagnosis model; and sequencing the frequent fault factor item set by utilizing a pre-established logic relation database between the frequent fault factor item set and the fault type and the preset confidence in the fault diagnosis model to obtain the fault type. The invention scans the historical fault data, performs fault tolerance processing and generalization, establishes a fault diagnosis model, explores the fault type from a macroscopic angle, and realizes the comprehensive analysis of fault factors after line faults.
Description
Technical Field
The invention relates to a power grid fault determining method and a storage medium based on an Apriori algorithm, and belongs to the technical field of power grid fault determination.
Background
With the improvement of management and operation and maintenance levels of the current power company, the operation mode of single type data analysis cannot well meet the current production requirements of the power system, and in view of the fact that the types of fault information systems are more and the precision requirements are higher, the power company is more hopeful to be able to adopt an operation management mode based on a big data mode. During the long-term operation of the power grid, a large amount of fault data are accumulated, wherein the fault data comprise fault-related voltage states, current states, relay protection device actions, operation information and the like, and when faults occur, a series of electrical reactions can be generated, so that the protection device actions are triggered, and the fault processing process is started. The series of secondary processes have a certain strong correlation, and the running mode of single type data analysis cannot well meet the current production requirements of the power system. The conventional data analysis method lacks a fault diagnosis model for accurately describing the strong correlation rule on the data level, so that the existing information system can only judge the fault type from single type evidence, the immunity is poor, the complex fault is analyzed, the arrangement and the correction of the multi-source data can only depend on manual work, and the working efficiency is low. In the current power grid fault diagnosis process, power grid dispatching personnel need to synthesize data of a plurality of sets of information systems to conduct research and judge and finish dispatching decisions, and the following problems exist:
1) The data volume of the information system is large, the formats are not uniform, the internal connection of the data is not tight, and fusion analysis cannot be realized;
2) The number of information systems is increased, the error probability is increased, a data verification means is lacked, and the reliability of fault diagnosis is reduced;
3) The information system principles are different, and the difficulty, the efficiency and the accuracy of relying on manual analysis are high.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a power grid fault determining method and a storage medium based on an Apriori algorithm, scans historical fault data, performs fault tolerance processing and generalization, establishes a fault diagnosis model, explores fault types from a macroscopic angle by combining the change characteristics of fault factors such as current, voltage, switch variables and the like of the current fault, and realizes comprehensive analysis of the fault factors after line faults, so that the fault analysis result has better pertinence and selectivity.
In a first aspect, the present invention provides a method for determining a power grid fault based on an Apriori algorithm, including:
preprocessing pre-acquired fault data to obtain standardized fault data;
inputting standardized fault data into a pre-constructed fault diagnosis model to obtain a fault type;
the method for obtaining the fault type comprises the steps of inputting standardized fault data into a pre-constructed fault diagnosis model to obtain the fault type, wherein the method comprises the following steps of:
iterative screening is carried out by using an Apriori algorithm and a preset support degree in a fault diagnosis model to obtain a fault factor frequent item set in standardized fault data;
calculating preset confidence in the fault diagnosis model;
and sequencing the frequent fault factor item set by utilizing a pre-established logic relation database between the frequent fault factor item set and the fault type and the preset confidence in the fault diagnosis model to obtain the fault type.
In combination with the first aspect, preprocessing the pre-acquired fault data to obtain standardized fault data, including:
performing data fault tolerance processing on the pre-acquired fault data to obtain fault data after the data fault tolerance processing;
and carrying out data standardization conversion on the fault data subjected to the data fault tolerance processing to obtain standardized fault data.
In combination with the first aspect, performing data fault tolerance processing on pre-acquired fault data to obtain fault data after the data fault tolerance processing, where the data fault tolerance processing includes:
if the attribute of the pre-acquired fault data is missing, querying an attribute library, wherein the attribute library comprises attributes, attribute types and compensation information, and supplementing the missing attributes by using the compensation information;
and carrying out fuzzy measurement and sequencing classification on the pre-acquired fault data or the fault data after supplement without attribute deletion to obtain the fault data after the fault tolerance processing of the data.
In combination with the first aspect, performing data fault tolerance processing on pre-acquired fault data to obtain fault data after the data fault tolerance processing, where the data fault tolerance processing includes:
if the number of missing fault factor data in the pre-acquired fault data is higher than a set fault factor data upper threshold, supplementing the missing fault factor data by utilizing Lagrange interpolation;
if the number of missing fault factor data in the pre-acquired fault data is lower than a set fault factor data lower limit threshold value, linearly calculating an average value to supplement the missing fault factor data;
if redundant pre-acquired fault data exist, deleting the redundant pre-acquired fault data;
performing similarity measure conversion and granularity conversion on fault data without fault factor data deficiency, pre-acquired fault data without redundancy, fault data with fault factor data deficiency in a filling way or pre-acquired fault data with redundancy deleted to obtain first preprocessing fault data;
and performing fixed-distance classification or fixed-ratio classification on the first preprocessing fault data to obtain fault data after the data fault tolerance processing.
In combination with the first aspect, performing data fault tolerance processing on pre-acquired fault data to obtain fault data after the data fault tolerance processing, where the data fault tolerance processing includes:
if the number of missing fault factor data in the pre-acquired fault data is higher than a set fault factor data upper threshold, supplementing the missing fault factor data by utilizing Lagrange interpolation;
if the number of missing fault factor data in the pre-acquired fault data is lower than a set fault factor data lower limit threshold value, linearly calculating an average value to supplement the missing fault factor data;
if redundant pre-acquired fault data exist, deleting the redundant pre-acquired fault data to obtain fault data after the data fault tolerance processing.
In combination with the first aspect, performing data standardization conversion on fault data after the data fault tolerance processing to obtain standardized fault data, including:
performing numerical processing on fault data after non-numerical data fault tolerance processing to obtain enumeration data;
if the enumerated data has the attribute isomerism, detecting the enumerated data by using the fuzzy measure;
and if the result of detecting the enumeration type data by using the fuzzy measure is qualified, performing granularity conversion on the enumeration type data to obtain standardized fault data.
With reference to the first aspect, if the enumerated data has heterogeneous attributes, detecting the enumerated data by using the fuzzy measure includes:
defining attribute combination space Ω= { C 1 ，C 2 ，…，C k Each attribute C i The departments and definitions of the departments are recorded in a pre-constructed attribute library, i epsilon [1, k]；
Acquiring attribute set F= { F of enumeration data 1 ，f 2 ，…，f k }，f k The attribute of the kth enumeration data;
if any i (1.ltoreq.i.ltoreq.k), f is present i ∈C i And judging the enumeration type data to be qualified.
In combination with the first aspect, if the result of detecting the enumeration type data by using the fuzzy measure is qualified, performing granularity conversion on the enumeration type data to obtain standardized fault data, including:
and if the result of detecting the enumeration type data by using the fuzzy measure is qualified, carrying out fixed-distance classification or fixed-ratio classification on the enumeration type data to obtain standardized fault data.
With reference to the first aspect, pre-constructing a fault diagnosis model includes:
randomly sampling in the historical fault data to obtain historical sample data;
iterative screening is carried out by using an Apriori algorithm and a preset support degree to obtain a frequent item set of the historical fault factors in the historical sample data;
calculating a relevance value among the fault factors in the frequent item set of the historical fault factors by using the preset confidence;
and sequencing the frequent fault factor item sets by using the relevance value to obtain a fault diagnosis model.
In a second aspect, the invention provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the method of any one of the first aspects when the program is executed.
In a third aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method of any of the first aspects.
The invention has the beneficial effects that:
the invention provides a power grid fault determining method and a storage medium based on an Apriori algorithm, which are used for scanning historical fault data, carrying out fault tolerance processing and generalization, establishing a fault diagnosis model, combining the change characteristics of current, voltage, switching variable and other fault factors of the current fault, exploring the fault type from a macroscopic angle, realizing comprehensive analysis of the fault factors after line fault, and enabling the fault analysis result to have better pertinence and selectivity. The subsequent operation is only conducted on the new fault data, repeated operation of historical fault data is avoided, the body quantity of the fault diagnosis model is expanded, and the mining efficiency is improved.
Drawings
In order to more clearly illustrate the technical solutions of the present application, the drawings that are needed in the embodiments will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a schematic illustration of an event tree in some embodiments of the present application;
FIG. 2 is a flow chart of a data generalization process in some embodiments of the present application;
fig. 3 is a functional block diagram in some embodiments of the present application.
Detailed Description
In order to facilitate the technical solution of the application, some concepts related to the present application will be described below first.
A large amount of fault data is accumulated in the long-term running process of the power grid, the fault data comprise voltage states, current states, relay protection device actions and running information related to faults, a logic relation (strong correlation rule) reflecting the power grid fault mechanism exists between the fault data, but a fault diagnosis model accurately describing the strong correlation rule is lacking in a data layer, so that the existing information system can only judge the fault type from single type evidence, the immunity is poor, the analysis of complex faults is carried out, the arrangement and the correction of multi-source data can only depend on manual work, and the working efficiency is low.
The Apriori algorithm is a classical correlation algorithm, and can mine implicit, unknown, potentially interesting and potentially valuable and fault diagnosis models for decision making from a large amount of data, and then acquire a high-dimensional frequent item set through a low-dimensional frequent item set by adopting a layer-by-layer iteration method.
The method comprises the steps of introducing an Apriori algorithm into the multi-source data comprehensive diagnosis of power grid faults, applying a fault diagnosis model in data mining to correlation analysis of power grid faults, finding valuable fault diagnosis models among items in a large amount of fault data, finding knowledge from the large amount of fault data, calculating a fault frequent item set by utilizing various frequent item set mining algorithms, finally merging, and mining the practical application value of the fault frequent item set.
As shown in fig. 3, the present application provides a power grid fault determining method based on Apriori algorithm, including: preprocessing pre-acquired fault data to obtain standardized fault data; and inputting the standardized fault data into a pre-constructed fault diagnosis model to obtain the fault type.
In this embodiment, inputting standardized fault data into a pre-constructed fault diagnosis model to obtain a fault type includes: iterative screening is carried out by using an Apriori algorithm and a preset support degree in a fault diagnosis model to obtain a fault factor frequent item set in standardized fault data; calculating a relevance value among fault factors in the frequent fault factor item set by using the preset confidence in the fault diagnosis model; and sequencing the frequent fault factor item sets by using the relevance value in the fault diagnosis model to obtain the fault type.
In this embodiment, the standardized fault data may include fault factor a, fault factor B, fault factor C, fault factor D, and fault factor E, where the fault factor a, fault factor B, fault factor C, fault factor D, and fault factor E may be a combination of 5 selected randomly from the fault factor data such as voltage phase, voltage value, current phase, current value, fault device, fault cause, distance, current, voltage, switch deflection, reclosing, protection action, and waveform curve.
In this embodiment, as shown in fig. 1, by using an Apriori algorithm and a preset support in a fault diagnosis model, a fault factor frequent item set in standardized fault data is obtained through iterative screening, which includes: referring to the self-connection relation between the fault factor a, the fault factor B, the fault factor C, the fault factor D and the fault factor E shown in fig. 1, an iterative method is adopted to search out a fault factor candidate 1 item set { a, B, C, D, E } and a corresponding preset support, pruning is carried out to remove the fault factor 1 item set lower than the preset support, and a fault factor frequent 1 item set greater than or equal to the preset support is obtained. And then, performing self-connection on the frequent 1-item sets of the fault factors with the support degree being greater than or equal to the preset support degree to obtain 2-item sets of the candidate fault factors, screening and removing the frequent 2-item sets of the candidate fault factors with the support degree being lower than the preset support degree to obtain a real frequent 2-item set { AC, AD, AE, BC, BD, BE, CD, CE, DE } of the fault factors with the support degree being greater than or equal to the preset support degree, and iterating until the frequent item sets of the fault factors cannot be found, wherein the corresponding frequent 3-item sets of the fault factors are the frequent item sets of the fault factors in the standardized fault data.
The fault diagnosis model of the reliability of the comprehensive power grid fault diagnosis mainly comprises a preset support degree and a preset confidence degree which are related to fault factors contained in the multi-source fault data:
1) Support degree: strong correlation rule A->The preset support=p (AB) of B refers to the probability that the fault factor a and the fault factor B occur in the total term set at the same time. The index can be used as a first threshold for establishing a strong association rule by setting a minimum support threshold S min Nonsensical fault diagnosis models with low occurrence probability are eliminated, and fault diagnosis models implied by more frequent item sets are reserved.
2) Confidence level: the confidence=p (b|a) =p (AB)/P (a) is preset, that is, the probability of containing the fault factor B in the term set containing the fault factor a, and the confidence may indicate the magnitude of the association degree between the fault factors.
For files containing a large number of non-digital languages in the accident handling process, fault data come from a plurality of professions and a plurality of systems, a large amount of scattered, heterogeneous and redundant fault data needing reasoning and correlation exist, the fault data of the original concepts cannot be identified by a computer and are not easy to use by an algorithm, and the fault data can be divided into two types: one type is enumeration type data, which is expressed by words, such as the identity, the fault device and the fault reason; the other is quantized data such as distance, current and voltage. Firstly, based on the fault information of the power transmission line, matching corresponding fault files, finding out a fault window, generalizing fault data in the fault window, and adopting the following steps as shown in fig. 2. In this embodiment, preprocessing pre-acquired fault data to obtain standardized fault data includes:
the generalization process of fault data (fault data that has not been fault-tolerant processed or normalized) can be divided into the following two parts:
performing data fault tolerance processing on the pre-acquired fault data to obtain fault data after the data fault tolerance processing; and carrying out data standardization conversion on the fault data subjected to the data fault tolerance processing to obtain standardized fault data.
In this embodiment, performing data fault-tolerant processing on pre-acquired fault data to obtain fault data after the data fault-tolerant processing includes: in this embodiment, if there is a missing or redundant situation in the fault data, and the fault data is processed repeatedly or cannot be interpreted correctly, the fault file is scanned first to perform initial judgment; if the number of missing fault factor data in the pre-acquired fault data is higher than a set fault factor data upper threshold, supplementing the missing fault factor data by utilizing Lagrange interpolation; if the number of missing fault factor data in the pre-acquired fault data is lower than a set fault factor data lower limit threshold value, linearly calculating an average value to supplement the missing fault factor data; if redundant pre-acquired fault data exist, deleting the redundant pre-acquired fault data to obtain fault data after the data fault tolerance processing. And then performing similarity measure conversion and granularity conversion on fault data without fault factor data deficiency, pre-acquired fault data without redundancy, fault data with fault factor data deficiency in a filling way or pre-acquired fault data with redundancy deleted to obtain first preprocessing fault data. And performing fixed-distance classification or fixed-ratio classification on the first preprocessing fault data to obtain generalized data (namely fault data after the fault tolerance processing of the data).
Acquiring a fault file, scanning the fault file, and acquiring pre-acquired fault data; if the attribute of the pre-acquired fault data is missing, inquiring a pre-constructed attribute library, wherein the attribute library comprises attribute types and corresponding compensation information, and supplementing the missing attribute by using the compensation information; performing fuzzy measurement and classification on pre-acquired fault data or supplemented fault data without attribute deletion, and supplementing the deleted attribute by using compensation information; if the pre-acquired fault data has no attribute missing, fuzzy measurement and classification are carried out on the pre-acquired fault data to obtain generalized data (namely fault data after data fault tolerance processing).
In this embodiment, since the fault recording devices come from multiple manufacturers, the fault files are different in specification, different in standard quantity, and the diversity brings about data inconsistency, for example, even if the same current or voltage is used, the difference of the CT/PT transformation ratio brings about a large deviation of the current or voltage, or has different dimensions and magnitude. The standardized conversion needs to be performed on the fault data after the fault tolerance processing, which comprises the following steps: performing fuzzy measurement and sequencing classification on enumeration data with data attribute semantic non-standard; and carrying out similarity measure conversion, distance determination or fixed ratio classification on the quantized data with different scales and magnitude order differences.
In this embodiment, if there is nonstandard semantics in the enumerated data attribute in the failure data, detecting the enumerated data by using the fuzzy measure includes: all attribute combination spaces are defined to be omega, wherein k attributes exist, and the attribute combination space omega= { C 1 ，C 2 ，…，C k Each attribute C i The departments and definitions of the departments are recorded in a pre-constructed attribute library, i epsilon [1, k]The purpose is to convert each heterogeneous attribute into standard semantics; acquiring attribute set F { F of enumeration data 1 ，f 2 ，…，f k }，f k The attribute of the kth enumeration data; if any i (1.ltoreq.i.ltoreq.k), f is present i ∈C i And judging the enumeration type data to be qualified.
In this embodiment, in the matching process, the situation that the data cannot be identified is unavoidable, manual intervention can be performed, and the conversion rate is improved along with accumulation of the attribute library. The enumerated data attributes in the fault data are then sequenced, such as fault classifications including power failure, circuit failure, device failure, and component failure.
In this embodiment, if the quantized data in the fault data have different dimension and magnitude differences, similarity measure conversion is performed on the quantized data, and the quantized data with different dimension and magnitude differences are converted into dimensionless pure values.
In the present embodiment, for the case of different dimensions: for example, the voltage amplitude of a certain group of fault data adopts primary value dimension of the power system, and most of fault data adopts secondary value dimension of the power system, or different choices of 1A or 5A exist in secondary rated current of each station in the power system according to different loads, at the moment, the amplitude is not adopted, but the amplitude ratio is adopted as a fault factor, such as a cycle current ratio before/after the fault moment, a maximum fault current/rated value, a reclosing current/rated value and the like, and the similarity calculation adopts a common Euclidean distance calculation method.
In this embodiment, for the case of different orders of magnitude: for example, the magnitude difference of sampling points in the cycle interval time is similar to that of the magnitude difference of sampling points, the magnitude of sampling points or sampling points is not used as a fault factor at the moment, the change rate of a cycle curve is used as the fault factor, and the similarity calculation adopts a common cosine distance calculation method.
In this embodiment, for nonstandard fault data, granularity conversion needs to be performed to make the fault data fall into a small specific interval, and although some details are discarded, the granularity fault data is more meaningful and effective features are easier to obtain. Granularity conversion is completed on the attributes such as current amplitude, voltage amplitude, cycle time interval and the like by using a classification method of fixed-distance classification or fixed-ratio classification, for example, fault-related current amplitude is classified into ultra-low, medium, high and ultra-high according to the fixed-distance classification method, and fault data such as other continuous voltage amplitude, cycle time interval and the like are converted and mapped to corresponding specific intervals.
In the present embodiment, the fault diagnosis model is constructed in advance, including: randomly sampling in the historical fault data to obtain historical sample data; iterative screening is carried out by using an Apriori algorithm and a preset support degree to obtain a frequent item set of the historical fault factors in the historical sample data; calculating a relevance value among the fault factors in the frequent item set of the historical fault factors by using the preset confidence; and sequencing the frequent fault factor item sets by using the relevance value to obtain a fault diagnosis model. The historical fault data may be a voltage phase, a voltage value, a current phase, a current value, a fault device, a fault cause, a distance, a current, a voltage, a switch deflection, a reclosing, a protection action, and a waveform profile of the historical record.
In this embodiment, for the establishment of the fault diagnosis model, it is necessary to randomly extract part of the historical fault data from the historical fault data, take the part of the historical fault data as a sample, and take a data set formed by the samples as a training set of the fault diagnosis model. Each sample may be represented as an attribute tuple, characterizing a fault state characteristic attribute. Digging generalized original fault data by using an Apriori algorithm, comparing characteristic quantities such as a data curve, switch deflection and duration of a historical event process, searching similar cases, referring to event reasons and a processing method of a historical fault record, and establishing a fault diagnosis model, wherein the steps are as follows:
1) Fault diagnosis model m=<Dia，State(value)>Dia is event cause D and corresponding fault processing method, D= { D 1 ，d 2 ，…，d n The number of fault types is n, d i For the ith fault type, state (value) is fault State information corresponding to the generalized fault type, and each fault type has a corresponding diagnosis result and a fault processing method. Definition c= { C 1 ，C 2 ，…，C m Fault state information in database, frequent fault factor item set C k ＝{c 1 ，c 2 ，…，c p A frequent item set of fault factors, k E [1, m ] in fault state information C in database]Fault factor frequent item set C k Support degree ofSetting a minimum support threshold S according to the running condition min Fault factor frequent item set C k Support degree is->The calculation formula of (2) is as follows: />Wherein num (C) k ) Frequent item set C for failure factor k The number of occurrences, W, is the number of occurrences of all faults.
2) Frequent item set C for fault factors in sequence k Performing support degree calculation and collecting frequent items of fault factors C k Support filtering is performed, if the fault factor frequent item setFault factor frequent item set C k Will be filtered, S min The value is not constant and depends on the number and quality of fault conditions in the database. Frequent item sets for fault factorsMay be the case with very few protective equipment failures or switch deactuations, as S min And filtering the support degree for the standard, and iteratively screening to obtain a fault factor frequent item set meeting the requirements by using an Apriori algorithm and a preset support degree in a fault diagnosis model.
3) Calculating fault type d i Confidence of (2), i.e. calculate the frequent fault factor term set C using (2) k For fault type d i Confidence Conf of (2)idence(C k ＝＞d)：
Confidence level is embodied in the frequent item set C of fault factors k Is diagnosed as failure type d i Pre-establishing a frequent fault factor item set C k And the logical relation database is used for inquiring the logical relation database to obtain the fault type corresponding to the frequent fault factor item set.
4) After the work is finished, a fault diagnosis model M is built, the fault diagnosis model M is built according to historical data, the faults are scientifically classified, a new fault diagnosis model is built according to the increase of data quantity, and the accuracy of fault judgment is improved along with the enrichment of the fault diagnosis model. Comparing the latest information acquired by the fault site, obtaining the confidence coefficient of the fault type according to the fault typical state analysis, and realizing the function of confirming the fault type, thereby improving the accuracy and efficiency of complex fault judgment and the intelligent level of the scheduling decision support.
According to experimental data generalization results, performing strong correlation rule mining by using an Apriori algorithm, and setting a minimum support degree S in view of diversity and complexity of faults min 0.05, minimum Confidence (C k ＝＞d i ) And (5) mining fault data with multi-source attributes to establish a fault diagnosis model. The method has the advantages that 50 times of typical fault cases are taken, the fault type is judged by using a fault diagnosis model, and the confidence coefficient under the corresponding state is obtained, wherein 43 times of judgment results are accurate, and the power grid fault is analyzed by using a strong correlation rule to obtain a good effect, so that the method has high practicability and reliability.
The partial results are shown in Table 1, limited to the table size, showing only the local state quantities.
TABLE 1 Fault case correlation fault factor analysis Table
Analysis of fault cases in table 1: the overtime phenomenon of the breaking time of the breaker in case 1 is compared with the similar accident before, and the fault type is judged to be the fault of the secondary control loop of the breaker; the case 2 has longer thermal stability duration, and has the similarity of overhigh short-circuit current, larger impact frequency, equivalent value and the like, and the insulation strength is reduced due to the deformation of the transformer winding caused by the short-circuit impact in the past year, so that the system judges the fault type as the short-circuit fault of the transformer; and according to the relevance of the change of the phase current and the voltage and the change quantity, the fault types are respectively judged to be single-phase grounding short circuit, two-phase short circuit and single-phase open-phase.
Example two
In this embodiment, the present invention provides an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the steps of any of the methods described above when the program is executed.
In this embodiment, the invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods described above.
According to the invention, by mining the strong correlation rule of the function, equipment and type fault information under the condition that the power grid breaks down, and realizing the modeling of the information processing process of the power grid fault including protection information, wave recording files and the like based on the strong correlation rule, the analysis and diagnosis of the equipment operation information are developed, and the establishment of a fault diagnosis model based on the Apriori algorithm is provided by combining the characteristics.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
The foregoing detailed description of the embodiments has further described the objects, technical solutions and advantageous effects of the present application, and it should be understood that the foregoing is only a detailed description of the present application and is not intended to limit the scope of the present application, and any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the present application should be included in the scope of protection of the present application.
Claims (10)
1. The power grid fault determining method based on the Apriori algorithm is characterized by comprising the following steps of:
preprocessing pre-acquired fault data to obtain standardized fault data;
inputting standardized fault data into a pre-constructed fault diagnosis model to obtain a fault type;
the method for obtaining the fault type comprises the steps of inputting standardized fault data into a pre-constructed fault diagnosis model to obtain the fault type, wherein the method comprises the following steps of:
iterative screening is carried out by using an Apriori algorithm and a preset support degree in a fault diagnosis model to obtain a fault factor frequent item set in standardized fault data;
calculating preset confidence in the fault diagnosis model;
and sequencing the frequent fault factor item set by utilizing a pre-established logic relation database between the frequent fault factor item set and the fault type and the preset confidence in the fault diagnosis model to obtain the fault type.
2. The method for determining a power grid fault based on the Apriori algorithm of claim 1,
preprocessing pre-acquired fault data to obtain standardized fault data, including:
performing data fault tolerance processing on the pre-acquired fault data to obtain fault data after the data fault tolerance processing;
and carrying out data standardization conversion on the fault data subjected to the data fault tolerance processing to obtain standardized fault data.
3. The method for determining a power grid fault based on the Apriori algorithm of claim 2,
performing data fault tolerance processing on pre-acquired fault data to obtain fault data after the data fault tolerance processing, wherein the data fault tolerance processing comprises the following steps:
if the attribute of the pre-acquired fault data is missing, querying an attribute library, wherein the attribute library comprises attributes, attribute types and compensation information, and supplementing the missing attributes by using the compensation information;
and carrying out fuzzy measurement and sequencing classification on the pre-acquired fault data or the fault data after supplement without attribute deletion to obtain the fault data after the fault tolerance processing of the data.
4. The method for determining a power grid fault based on the Apriori algorithm of claim 3,
performing data fault tolerance processing on pre-acquired fault data to obtain fault data after the data fault tolerance processing, wherein the data fault tolerance processing comprises the following steps:
if the number of missing fault factor data in the pre-acquired fault data is higher than a set fault factor data upper threshold, supplementing the missing fault factor data by utilizing Lagrange interpolation;
if the number of missing fault factor data in the pre-acquired fault data is lower than a set fault factor data lower limit threshold value, linearly calculating an average value to supplement the missing fault factor data;
if redundant pre-acquired fault data exist, deleting the redundant pre-acquired fault data;
performing similarity measure conversion and granularity conversion on fault data without fault factor data deficiency, pre-acquired fault data without redundancy, fault data with fault factor data deficiency in a filling way or pre-acquired fault data with redundancy deleted to obtain first preprocessing fault data;
and performing fixed-distance classification or fixed-ratio classification on the first preprocessing fault data to obtain fault data after the data fault tolerance processing.
5. The method for determining a power grid fault based on the Apriori algorithm of claim 2,
performing data standardization conversion on the fault data subjected to the data fault tolerance processing to obtain standardized fault data, wherein the method comprises the following steps:
performing numerical processing on fault data after non-numerical data fault tolerance processing to obtain enumeration data;
if the enumerated data has the attribute isomerism, detecting the enumerated data by using the fuzzy measure;
and if the result of detecting the enumeration type data by using the fuzzy measure is qualified, performing granularity conversion on the enumeration type data to obtain standardized fault data.
6. The method for determining a power grid fault based on the Apriori algorithm of claim 5,
if the enumerated data has heterogeneous attributes, detecting the enumerated data by using the fuzzy measure, including:
defining attribute combination space Ω= { C 1 ，C 2 ，…，C k Each attribute C i The departments and definitions of the departments are recorded in a pre-constructed attribute library, i epsilon [1, k]；
Acquiring attribute set F= { F of enumeration data 1 ，f 2 ，…，f k }，f k The attribute of the kth enumeration data;
if any i (1.ltoreq.i.ltoreq.k), f is present i ∈C i And judging the enumeration type data to be qualified.
7. The method for determining a power grid fault based on the Apriori algorithm of claim 5,
if the result of detecting the enumeration type data by using the fuzzy measure is qualified, performing granularity conversion on the enumeration type data to obtain standardized fault data, wherein the method comprises the following steps:
and if the result of detecting the enumeration type data by using the fuzzy measure is qualified, carrying out fixed-distance classification or fixed-ratio classification on the enumeration type data to obtain standardized fault data.
8. The method for determining a power grid fault based on the Apriori algorithm of claim 1,
pre-constructing a fault diagnosis model, comprising:
randomly sampling in the historical fault data to obtain historical sample data;
iterative screening is carried out by using an Apriori algorithm and a preset support degree to obtain a frequent item set of the historical fault factors in the historical sample data;
calculating a relevance value among the fault factors in the frequent item set of the historical fault factors by using the preset confidence;
and sequencing the frequent fault factor item sets by using the relevance value to obtain a fault diagnosis model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any one of claims 1 to 8 when the program is executed by the processor.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477952.1A CN116468426A (en) | 2023-04-28 | 2023-04-28 | Power grid fault determining method and storage medium based on Apriori algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310477952.1A CN116468426A (en) | 2023-04-28 | 2023-04-28 | Power grid fault determining method and storage medium based on Apriori algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116468426A true CN116468426A (en) | 2023-07-21 |
Family
ID=87176984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310477952.1A Pending CN116468426A (en) | 2023-04-28 | 2023-04-28 | Power grid fault determining method and storage medium based on Apriori algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116468426A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117132025A (en) * | 2023-10-26 | 2023-11-28 | 国网山东省电力公司泰安供电公司 | Power consumption monitoring and early warning system based on multisource data fusion |
| CN117310333A (en) * | 2023-10-10 | 2023-12-29 | 国网江苏省电力有限公司扬州供电分公司 | High-low voltage overall process fault studying and judging method based on key factor filtering method |
-
2023
- 2023-04-28 CN CN202310477952.1A patent/CN116468426A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117310333A (en) * | 2023-10-10 | 2023-12-29 | 国网江苏省电力有限公司扬州供电分公司 | High-low voltage overall process fault studying and judging method based on key factor filtering method |
| CN117132025A (en) * | 2023-10-26 | 2023-11-28 | 国网山东省电力公司泰安供电公司 | Power consumption monitoring and early warning system based on multisource data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106779505B (en) | Power transmission line fault early warning method and system based on big data driving | |
| CN116468426A (en) | Power grid fault determining method and storage medium based on Apriori algorithm | |
| US20120304008A1 (en) | Supervised fault learning using rule-generated samples for machine condition monitoring | |
| MX2013000577A (en) | Machine learning for power grids. | |
| CN116467674B (en) | Intelligent fault processing fusion updating system and method for power distribution network | |
| CN111027615A (en) | Middleware fault early warning method and system based on machine learning | |
| CN112464439A (en) | Three-layer data mining-based power system fault type analysis method and device | |
| CN112785108A (en) | Power grid operation data correlation analysis method and system based on regulation cloud | |
| CN114723285A (en) | Power grid equipment safety evaluation prediction method | |
| CN111044845B (en) | Power distribution network accident identification method and system based on Apriori algorithm | |
| CN113408548A (en) | Transformer abnormal data detection method and device, computer equipment and storage medium | |
| CN116933010A (en) | Load rate analysis and evaluation method and system based on multi-source data fusion and deep learning | |
| CN116433045A (en) | Accurate supply management method and system for electric power emergency materials | |
| CN117394529A (en) | SCADA-based auxiliary decision method and system for main distribution network loop-closing reverse power supply control conditions | |
| CN113626480A (en) | Direct current converter station SER event set diagnosis method based on improved association rule | |
| CN117034149A (en) | Fault processing strategy determining method and device, electronic equipment and storage medium | |
| CN110361609B (en) | Extra-high voltage equipment monitoring system and method | |
| CN116011158A (en) | Topology identification method, system and device for low-voltage transformer area | |
| CN114265837A (en) | Station-side interactive data feature extraction method | |
| CN114399177A (en) | Scheduling disposal rule mining and generating method and system based on Apriori | |
| CN113325266A (en) | Power grid fault diagnosis method and system based on fuzzy integral multi-source information fusion | |
| Khalyasmaa et al. | Fuzzy inference algorithms for power equipment state assessment | |
| CN117171548B (en) | Intelligent network security situation prediction method based on power grid big data | |
| Dunwen et al. | A trainsient voltage stability evaluation model based on morphological similarity distance online calculation | |
| Zhang et al. | Substation Operation Information Maintenance Based on Intelligent Data Mining |
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 |