CN110309134A - The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community - Google Patents
The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community Download PDFInfo
- Publication number
- CN110309134A CN110309134A CN201910473298.0A CN201910473298A CN110309134A CN 110309134 A CN110309134 A CN 110309134A CN 201910473298 A CN201910473298 A CN 201910473298A CN 110309134 A CN110309134 A CN 110309134A
- Authority
- CN
- China
- Prior art keywords
- period
- electricity consumption
- user
- community
- behavior
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 74
- 230000005856 abnormality Effects 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 238000012546 transfer Methods 0.000 title claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims abstract description 48
- 230000006399 behavior Effects 0.000 claims abstract description 47
- 230000011273 social behavior Effects 0.000 claims abstract description 8
- 238000004140 cleaning Methods 0.000 claims abstract description 6
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000000034 method Methods 0.000 claims description 24
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000005012 migration Effects 0.000 claims description 10
- 238000013508 migration Methods 0.000 claims description 10
- 238000009825 accumulation Methods 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 2
- 238000005192 partition Methods 0.000 claims description 2
- 239000004744 fabric Substances 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 2
- 230000003542 behavioural effect Effects 0.000 description 10
- 230000008859 change Effects 0.000 description 10
- 230000010429 evolutionary process Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000007423 decrease Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000000638 stimulation Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000001594 aberrant effect Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 206010016256 fatigue Diseases 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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
- G06F16/2455—Query execution
- G06F16/24564—Applying rules; Deductive 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
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Human Resources & Organizations (AREA)
- Water Supply & Treatment (AREA)
- Fuzzy Systems (AREA)
- Biophysics (AREA)
- Public Health (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of power distribution network multiplexing electric abnormality detection methods to be developed based on electricity consumption transfer of behavior and community, detection method includes the following steps for this: step 1: acquisition user power consumption data simultaneously carry out data cleansing, carry out feature extraction to the data after cleaning and obtain electricity consumption data sequence;Step 2: being matched to after electricity consumption data sequence division period progress trend cluster with individual and social behavior's feature, and carry out community evolution and migrate to estimate according to matching result;Step 3: being developed according to community and migrate estimated result calculating cycle interval exceptional value and local buildup Abnormal Evolution value;Step 4: period distances exceptional value and local buildup Abnormal Evolution value being ranked up and compared using ranking results and history abnormal user exception electricity consumption behavior corresponding data obtain user power utilization abnormality detection result.Compared with prior art, the present invention has many advantages, such as detection accuracy height, with strong points.
Description
Technical field
The present invention relates to power distribution network multiplexing electric abnormality detection technique fields, are based on electricity consumption transfer of behavior more particularly, to one kind
With the power distribution network multiplexing electric abnormality detection method of community fireworks.
Background technique
Distribution network users exception electricity consumption behavior contains power stealing, electric leakage, stealing and load nature of electricity consumed change, user's change
Behavior, it is grinding in electricity consumption behavioral analysis technology which, which can reduce demand side management efficiency and policy making validity,
Study carefully one of hot spot.
It is less for country's power stealing electric leakage sample data, abnormal electricity consumption behavior can not be carried out using supervised learning to be recognized
The status of model learning and building, one kind differentiating user's exception electricity consumption degree by comprehensive abnormal index and subitem abnormal index
With type, assists user management unit to carry out power utility check, user management etc. work, promote relevant departments, Utilities Electric Co.
The power distribution network multiplexing electric abnormality detection method of working efficiency is urgently developed.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind to be based on electricity consumption behavior
The power distribution network multiplexing electric abnormality detection method of migration and community fireworks.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community, the detection method packet
Include following steps:
Step 1: acquisition user power consumption data simultaneously carry out data cleansing, carry out feature extraction to the data after cleaning and obtain
To electricity consumption data sequence;
Step 2: being carried out to after electricity consumption data sequence division period progress trend cluster with individual and social behavior's feature
Matching, and community evolution is carried out according to matching result and is estimated with migration;
Step 3: being developed according to community and migrate estimated result calculating cycle interval exceptional value and local buildup Abnormal Evolution
Value;
Step 4: by period distances exceptional value and local buildup Abnormal Evolution value be ranked up and using ranking results with go through
The comparison of history abnormal user exception electricity consumption behavior corresponding data obtains user power utilization abnormality detection result.
Further, the acquisition process of electricity consumption data sequence specifically includes in the step 1: will be to be checked after cleaning
N power consumer of survey, with T days for a period, i-th of user electricity consumption in jth day within r-th of period is combined, obtains
To electricity consumption data sequence
Further, the step 2 include it is following step by step:
Step 21: the division period is carried out to electricity consumption data sequence;
Step 22: the electricity consumption trend of the electricity consumption data sequence after the computation partition period simultaneously carries out cluster operation to it;
Step 23: the electricity consumption data sequence after progress cluster operation is matched with individual and social behavior's feature;
Step 24: community being carried out according to matching result and is developed and migration estimation.
Further, the calculation formula of the electricity consumption trend in the step 22 are as follows:
In formula,Indicate the electricity consumption trend of i-th of user kth day within r-th of period,Indicate i-th user
The electricity consumption sequential value of kth day within r-th of period,I-th of user is indicated within the R period kth+1 day use
Electricity sequential value, R and r are natural number.
Further, the calculation formula of the period distances exceptional value in the step 3 are as follows:
In formula,Indicate the group abnormality value of user n in r-th of period, ηn(r → r+1) indicates period r to period r
The core degree changing value of+1 process user n,Indicate kth1Subordinated-degree matrix of a group in period r+1,Indicate kth2Subordinated-degree matrix of a group in period r, M indicate group's total number,It indicates
Kth1A group and kth2Transition matrix of a group during period r to period r+1, k1、k2It is natural number with M.
Further, the calculation formula of the local buildup Abnormal Evolution value in the step 3 are as follows:
In formula, φn,rIndicating the local buildup Abnormal Evolution value of user n in r-th of period, d indicates abnormal accumulation coefficient,
λlIndicate the abnormal accumulation coefficient of user n in first of period,Indicate the group abnormality value of user n in first of period, l
For natural number.
Further, the calculation formula of the core degree changing value are as follows:
In formula,Indicate period r to period r+1 process user VnThe core degree changing value of k,Indicate user VnCore degree of the k in period r+1,Indicate user VnCore of the k in period r
Heart degree.
Further, the calculation formula of the core degree are as follows:
In formula,Indicate user n in period rkCore degree,Indicate user n in period rkDegree of membership
Matrix, minI=1:M pik(r) and maxI=1:M pik(r) i-th of user kth day within r-th of period in 1~M group is indicated
Distribution density minimum value and maximum value.
Compared with prior art, the invention has the following advantages that
(1) in the present invention, made with electricity consumption classification, trade classification in the electricity consumption data of user and user's account information data
It carries out period EVOLUTIONARY COMPUTATION using community Evolution Theory for input pointer and obtains community evolution abnormality degree, specific steps include
Step 1: acquisition user power consumption data simultaneously carry out data cleansing, carry out feature extraction to the data after cleaning and obtain electricity consumption
Data sequence;Step 2: to electricity consumption data sequence divide the period carry out trend cluster after with individual and social behavior's feature into
Row matching, and community evolution is carried out according to matching result and is estimated with migration;Step 3: being developed according to community and migrate estimated result
Calculating cycle interval exceptional value and local buildup Abnormal Evolution value;Step 4: period distances exceptional value and local buildup is abnormal
Evolution value, which is ranked up and is compared using ranking results and history abnormal user exception electricity consumption behavior corresponding data, show that user uses
Electrical anomaly testing result show that testing result accuracy is high.
(2) with strong points, community of the present invention develops, and development and change become at any time for community in main research community network
The identification of individual behavior, evolved behavior include in gesture, rule and its community: the increase and decrease of community quantity and the variation of structure divide
Split, merge, expanding, shrinking and group between migrate etc..In power domain, the electric power with similar electricity consumption behavioural characteristic can
To be known as a kind of community.Power consumer electricity consumption behavioural characteristic is formed after being integrated by itself power demand and objective factor in community,
Different type user all has more apparent group feature to behavior response caused by extraneous factor stimulation.Therefore, lead to
Electricity consumption behavioural characteristic matching relationship between individual and community is crossed in identification evolutionary process, can preferably detect that individual behavior changes
Whether logicality is met, for the with strong points of power domain.
Detailed description of the invention
Fig. 1 is the multiplexing electric abnormality behavior evolution schematic diagram in the present invention;
Fig. 2 is method flow schematic diagram of the invention;
Fig. 3 is that the evolutionary process community quantity of the embodiment of the present invention changes schematic diagram;
Fig. 4 is No. 229 user's Abnormal Evolution result figures in the embodiment of the present invention;
Fig. 5 is No. 208 user's Abnormal Evolution result figures in the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiment is a part of the embodiments of the present invention, rather than whole embodiments.Based on this hair
Embodiment in bright, those of ordinary skill in the art are obtained every other without making creative work
Embodiment all should belong to the scope of protection of the invention.
Embodiment
The transfer of behavior of power consumer refers to that in period R, user's individual or user group's (community) electricity consumption rule are at any time
The process of development and variation.In evolutionary process, if individual VnIt is under the jurisdiction of different community k respectively1(r) and k2(r+1), k1≠
k2, then claim individual VnTransport phenomena has occurred relative to social behavior.
Community develops in main research community network community at any time in the trend, rule and its community of development and change
The identification of body behavior, evolved behavior include: the increase and decrease of community quantity and the variation of structure, that is, divide, merge, expansion, shrink with
And migration etc. between group.In power domain, the electric power with similar electricity consumption behavioural characteristic is properly termed as a kind of community.Community
Middle power consumer electricity consumption behavioural characteristic after itself power demand and objective factor synthesis by forming, and different type user is to the external world
Behavior response caused by factor stimulation all has more apparent group feature.Therefore, pass through in identification evolutionary process
Electricity consumption behavioural characteristic matching relationship between body and community, can preferably detect whether individual behavior variation meets logicality.
Abnormal Evolution detection is detected in Evolutionary Cycles " no by transfer of behavior feature between analysis individual and community
It is gregarious " individual, and labeled as Abnormal Evolution individual.Temporally span point, the abnormal behaviour of user can be divided into interval it is abnormal,
Localized accumulated exception and global free abnormal three classes.Interval refers to that a certain stage individual behavior mutates extremely, formation and institute
Belong to other most of individual different behavioural characteristics of community, is then returned to original behavior pattern again;Localized accumulated refers to extremely
In evolutionary process, individual occurs continuous multiple stages and is not consistent with original social behavior mode;Free exception refers to individual row
More significant otherness is all had with all community in evolution overall process for mode, behavior pattern is equal within the complete period
Outside migration Mr. Yu's community edge or all community, i.e. community degree of membership stabilization is held in reduced levels.
(of different transfer of behavior features is indicated with different shape respectively as shown in Figure 1 for three classes Abnormal Evolution user
Body) electricity consumption patterns of change path schematic diagram.Individual (interval is abnormal) is in the evolutionary phase 1 as former community some individuals divide
It is returned to and the consistent position of original state into a new community, and in the stage 2.The performance characteristic of such abnormal behavior
To occur behavior pattern jump in short-term in Evolutionary Cycles, then restores again to a certain community and behavior pattern is kept to be in steady
Determine state.Individual (free abnormal) dissociates outside community in all Evolutionary Cycles, without apparent community subordinate relation.It is a
Continuous migration is in different community in evolutionary process for body (localized accumulated abnormal), and duration change occurs for its behavior pattern
Change, role positioning in its behavior behavioral characteristics or community can not be recognized.
In community EVOLUTION ANALYSIS, for network users community rendezvous problem is matched, cluster efficiency is comprehensively considered, using improvement
Distributed K-means is clustered, and each cycle takes optimal K value using distribution K-means cluster automatic seeking.
It is illustrated in figure 2 method flow schematic diagram of the invention, this method uses community Evolution Theory, Abnormal Evolution inspection
The asynchronism feature by transfer of behavior between analysis individual and community is surveyed, " unsocial " individual in Evolutionary Cycles is detected, and
Labeled as Abnormal Evolution individual.To increase algorithm robustness, influence of the abnormal data to identification result is reduced, is tired out using part
It is abnormal that meter method identifies that community develops.
Steps are as follows for its key calculating:
(1) feature extraction
1. feature extraction: extracting n power consumer V to be detected1,V2,...Vn.With T days for a period,It indicates
The electricity consumption in i-th (i=1,2....n) a user jth day within r-th of period.I-th of power consumer ViIn the r period
Electricity consumption data sequence is
Define user ViElectricity consumption trendThen sequenceIt indicates
User ViThe electricity consumption trend sequence in r period, electricity consumption trendIt is calculated with formula.
Indicate i-th of power consumer ViElectricity consumption sequence W in period ri rAverage value.Calculation formula are as follows:
Indicate i-th of power consumer ViElectricity consumption sequence W in period ri rStandard deviation, calculation formula are as follows:
Eigenmatrix in n power consumer period rFor
2. subordinated-degree matrix calculates: εi,jIndicate user ViAnd VjElectricity consumption characteristic similarity.Similarity calculation is using European
Distance calculates and normalized;User characteristics cluster to form user group;If cluster forms M in period rrA group.K
(K=1,2.....Mr) group cluster centre user be CK, then user V in period riTo the degree of membership of k-th group are as follows:
Then all user V in period rnUser is to MrThe subordinated-degree matrix of a group are as follows:
3. transition matrix and core degree transformation matrices
Division, merging, expansion of the transition matrix X (r) to group during the Changing period r to period r+1 of quantitative description
, the dynamic change characterization between groups such as shrink.The subordinated-degree matrix of period r and period r+1 is respectively L (r), L (r+1).Then
Transition matrix S (r → r+1) meets formula:
Wherein:
Work as Kr=Kr+1When:When occurring merging in community evolutionary process or divide
Phenomenon, so that Kr≠Kr+1When:I is unit matrix; C
For a sufficiently large real number.
User kernel degree characterizes status of user's individual in group.Core degree can effectively inhibit user distribution in community
Influence of the density to its individual significance level.Power consumer n is subordinate to degree series to M group in period r are as follows: Ln(r)
=(Ln1(r),Ln2(r),…,LnM(r)).Core degree of the user n to k-th of community in period r are as follows:
The core degree changing value η of user n during period r to period r+1nK (r → r+1):
1. group abnormality calculates: group abnormality of the power consumer n in period r are as follows:
Localized accumulated of the power consumer n in period r is abnormal are as follows:
In formula: d is abnormal accumulation coefficient;λ be abnormal accumulation coefficient, λ ∈ (0,1), and
For further detection algorithm accuracy, randomly selects 2000 and tested with network users.With power distribution network in 2015
2000 users are experimental subjects, comprising 156 confirmation power stealing users and it is several it is uncertain whether 1844 of abnormal electricity consumption.It is logical
The parameters such as input user's account information, geographical location, weather, electricity consumption behavioural characteristic are crossed, the validity of proposed method is verified.
Community EVOLUTION ANALYSIS, using distributed K-means clustering method.Following figure indicates community accumulation process (1-12
Period) it is formed by community quantity change curve.In evolution overall process, user's community quantity is stable between 13-20,
Community quantity is relatively stable.
As can be seen from FIG. 3, it during electricity consumption behavior evolution, there are fuctuation within a narrow range or is remained unchanged with network users community
The phenomenon that, but community quantity is totally relatively stable, it was demonstrated that and community Evolution Theory is special with network users timing electricity consumption behavior for disclosing
Sign has preferable applicability.Corresponding point with group in network users community evolutionary process of the variation of evolutionary process community quantity
It splits, merge and the presence of individual behavior transport phenomena.It is drilled according to distribution user behavior migration in 12 Evolutionary Cycles with community
Synchronism characteristics between change solve abnormal user interval, accumulation exception and global abnormal value respectively.
Fig. 4 is set by taking No. 229 certain industrial electrical user's Abnormal Evolution processes as an example in the 4th Evolutionary Cycles stage
Standby failure, and carried out overhaul of the equipments in the 6th period and more changed jobs.Since user's maintenance work belongs to emitted behavior,
Cause to form biggish asynchronism feature between the community belonging to power mode and original.When being determined using interval Abnormal Evolution, missed
It is judged to abnormal electricity consumption.And the judgement of localized accumulated Abnormal Evolution is used then preferably to compensate for the unexpected incidents to identification result
It influences, only sets it to abnormity early warning.After user equipment maintenance replacement, it is just gradually reduced abnormity early warning etc.
Grade, and finally re-flag as normal electricity consumption user.
Fig. 5, No. 208 users develop initial stage (1-2) occur interval abnormity early warning, but due to its aberrant continuation period compared with
Short and intensity of anomaly is lower, and returns in the 3rd period behavior evolution in normal, therefore be not marked as abnormal user.When
4,6 periods occur continuity severely subnormal early warning accumulation after, which starts to step up, and
8th period was determined as abnormal electricity consumption.Although hereafter being reduced in the 11st period distances intensity of anomaly, due to multiplexing electric abnormality before this
Behavior is excessive, which is still labeled as abnormity early warning state.From another angle, the localized accumulated in the 11st period is different
Often determining can be regarded as excessively causes credit grade to decline extremely because of history, makes user after Abnormality remove still in being seen
Examine the stage undetermined.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, appoints
What those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications
Or replacement, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention is answered
It is subject to the protection scope in claims.
Claims (8)
1. a kind of power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community, which is characterized in that the inspection
Survey method the following steps are included:
Step 1: acquisition user power consumption data simultaneously carry out data cleansing, carry out feature extraction to the data after cleaning and obtain electricity consumption
Measure data sequence;
Step 2: it is matched to after electricity consumption data sequence division period progress trend cluster with individual and social behavior's feature,
And community is carried out according to matching result and is developed and migration estimation;
Step 3: being developed according to community and migrate estimated result calculating cycle interval exceptional value and local buildup Abnormal Evolution value;
Step 4: period distances exceptional value and local buildup Abnormal Evolution value being ranked up and utilize ranking results different with history
Common family exception electricity consumption behavior corresponding data comparison obtains user power utilization abnormality detection result.
2. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 1
Method, which is characterized in that the acquisition process of electricity consumption data sequence specifically includes in the step 1: by the n to be detected after cleaning
A power consumer, with T days for a period, i-th of user electricity consumption in jth day within r-th of period is combined, and obtains electricity consumption
Measure data sequence
3. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 1
Method, which is characterized in that the step 2 include it is following step by step:
Step 21: the division period is carried out to electricity consumption data sequence;
Step 22: the electricity consumption trend of the electricity consumption data sequence after the computation partition period simultaneously carries out cluster operation to it;
Step 23: the electricity consumption data sequence after progress cluster operation is matched with individual and social behavior's feature;
Step 24: community being carried out according to matching result and is developed and migration estimation.
4. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 3
Method, which is characterized in that the calculation formula of the electricity consumption trend in the step 22 are as follows:
In formula,Indicate the electricity consumption trend of i-th of user kth day within r-th of period,Indicate i-th of user in r
The electricity consumption sequential value of kth day in a period,I-th of user is indicated within the R period kth+1 day electricity consumption sequence
Train value, R and r are natural number.
5. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 1
Method, which is characterized in that the calculation formula of the period distances exceptional value in the step 3 are as follows:
In formula,Indicate the group abnormality value of user n in r-th of period, ηn(r → r+1) indicates period r to period r+1 mistake
The core degree changing value of journey user n,Indicate kth1Subordinated-degree matrix of a group in period r+1,
Indicate kth2Subordinated-degree matrix of a group in period r, M indicate group's total number,Indicate kth1A group
Body and kth2Transition matrix of a group during period r to period r+1, k1、k2It is natural number with M.
6. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 1
Method, which is characterized in that the calculation formula of the local buildup Abnormal Evolution value in the step 3 are as follows:
In formula, φn,rIndicate the local buildup Abnormal Evolution value of user n in r-th of period, d indicates abnormal accumulation coefficient, λlIt indicates
The abnormal accumulation coefficient of user n in first of period,Indicate the group abnormality value of user n in first of period, l is nature
Number.
7. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 5
Method, which is characterized in that the calculation formula of the core degree changing value are as follows:
In formula,Indicate period r to period r+1 process user VnThe core degree changing value of k,
Indicate user VnCore degree of the k in period r+1,Indicate user VnCore degree of the k in period r.
8. a kind of power distribution network multiplexing electric abnormality detection side to be developed based on electricity consumption transfer of behavior and community according to claim 7
Method, which is characterized in that the calculation formula of the core degree are as follows:
In formula,Indicate user n in period rkCore degree,Indicate user n in period rkDegree of membership square
Battle array, minI=1:Mpik(r) and maxI=1:Mpik(r) point of i-th of user kth day within r-th of period in 1~M group is indicated
Cloth density minimum value and maximum value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910473298.0A CN110309134A (en) | 2019-05-31 | 2019-05-31 | The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910473298.0A CN110309134A (en) | 2019-05-31 | 2019-05-31 | The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110309134A true CN110309134A (en) | 2019-10-08 |
Family
ID=68074974
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910473298.0A Pending CN110309134A (en) | 2019-05-31 | 2019-05-31 | The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110309134A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275576A (en) * | 2020-01-19 | 2020-06-12 | 烟台海颐软件股份有限公司 | Identification method and identification system for abnormal electricity price execution user |
CN112307435A (en) * | 2020-10-30 | 2021-02-02 | 三峡大学 | Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend |
CN113592533A (en) * | 2021-06-30 | 2021-11-02 | 国网上海市电力公司 | Abnormal electricity utilization detection method and system based on unsupervised learning |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140358838A1 (en) * | 2013-06-04 | 2014-12-04 | International Business Machines Corporation | Detecting electricity theft via meter tampering using statistical methods |
CN105608329A (en) * | 2016-01-26 | 2016-05-25 | 中国人民解放军国防科学技术大学 | Organizational behavior anomaly detection method based on community evolution |
CN107230013A (en) * | 2017-05-11 | 2017-10-03 | 华北电力大学 | With the abnormal electricity consumption identification of network users and timi requirement method under a kind of unsupervised learning |
CN109377409A (en) * | 2018-09-29 | 2019-02-22 | 重庆大学 | A kind of user power utilization anomaly detection method based on BP neural network |
-
2019
- 2019-05-31 CN CN201910473298.0A patent/CN110309134A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140358838A1 (en) * | 2013-06-04 | 2014-12-04 | International Business Machines Corporation | Detecting electricity theft via meter tampering using statistical methods |
CN105608329A (en) * | 2016-01-26 | 2016-05-25 | 中国人民解放军国防科学技术大学 | Organizational behavior anomaly detection method based on community evolution |
CN107230013A (en) * | 2017-05-11 | 2017-10-03 | 华北电力大学 | With the abnormal electricity consumption identification of network users and timi requirement method under a kind of unsupervised learning |
CN109377409A (en) * | 2018-09-29 | 2019-02-22 | 重庆大学 | A kind of user power utilization anomaly detection method based on BP neural network |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275576A (en) * | 2020-01-19 | 2020-06-12 | 烟台海颐软件股份有限公司 | Identification method and identification system for abnormal electricity price execution user |
CN112307435A (en) * | 2020-10-30 | 2021-02-02 | 三峡大学 | Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend |
CN112307435B (en) * | 2020-10-30 | 2024-05-31 | 三峡大学 | Method for judging and screening abnormal electricity consumption based on fuzzy clustering and trend |
CN113592533A (en) * | 2021-06-30 | 2021-11-02 | 国网上海市电力公司 | Abnormal electricity utilization detection method and system based on unsupervised learning |
CN113592533B (en) * | 2021-06-30 | 2023-09-12 | 国网上海市电力公司 | Abnormal electricity utilization detection method and system based on unsupervised learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
US11043808B2 (en) | Method for identifying pattern of load cycle | |
EP2529186B1 (en) | Robust automated determination of the hierarchical structure of utility monitoring systems | |
CN106780121B (en) | Power consumption abnormity identification method based on power consumption load mode analysis | |
CN109856299A (en) | A kind of transformer online monitoring differentiation threshold value dynamic setting method, system | |
Zhu et al. | Intelligent short-term voltage stability assessment via spatial attention rectified RNN learning | |
Chicco et al. | Emergent electricity customer classification | |
CN110309134A (en) | The power distribution network multiplexing electric abnormality detection method to be developed based on electricity consumption transfer of behavior and community | |
Zhang et al. | Power consumption predicting and anomaly detection based on transformer and K-means | |
CN106682079A (en) | Detection method of user's electricity consumption behavior of user based on clustering analysis | |
CN110288039B (en) | Electricity stealing detection method based on user electricity load characteristics | |
CN109165604A (en) | The recognition methods of non-intrusion type load and its test macro based on coorinated training | |
CN110045209A (en) | Detection method, device, equipment and the readable storage medium storing program for executing of electricity consumption data exception | |
CN108256559A (en) | A kind of low pressure stealing method for positioning user based on the local outlier factor | |
CN106203832A (en) | Intelligent electricity anti-theft analyzes system and the method for analysis | |
Zhu et al. | Networked time series shapelet learning for power system transient stability assessment | |
Batra et al. | Neighbourhood nilm: A big-data approach to household energy disaggregation | |
Miraftabzadeh et al. | K-means and alternative clustering methods in modern power systems | |
CN110298552A (en) | A kind of power distribution network individual power method for detecting abnormality of combination history electrical feature | |
Modarresi et al. | Robust phase detection in distribution systems | |
CN103018611A (en) | Non-invasive load monitoring method and system based on current decomposition | |
Sun et al. | Electricity theft detection method based on ensemble learning and prototype learning | |
CN104850918B (en) | A kind of node load Forecasting Methodology counted and power network topology constrains | |
CN113191409A (en) | Method for detecting abnormal electricity consumption behaviors of residents through tag data expansion and deep learning | |
Liu et al. | Bitcoin mining recognition based on community detection with electricity consumption data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191008 |
|
RJ01 | Rejection of invention patent application after publication |