CN109636667A - A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature - Google Patents
A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature Download PDFInfo
- Publication number
- CN109636667A CN109636667A CN201811493614.2A CN201811493614A CN109636667A CN 109636667 A CN109636667 A CN 109636667A CN 201811493614 A CN201811493614 A CN 201811493614A CN 109636667 A CN109636667 A CN 109636667A
- Authority
- CN
- China
- Prior art keywords
- user
- low
- electricity consumption
- voltage customer
- electric abnormality
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 24
- 230000005856 abnormality Effects 0.000 title claims abstract description 23
- 230000005611 electricity Effects 0.000 claims abstract description 56
- 230000002159 abnormal effect Effects 0.000 claims abstract description 17
- 239000011159 matrix material Substances 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 9
- 230000003542 behavioural effect Effects 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 241001269238 Data Species 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- 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
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Public Health (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature, is related to technical field of electric power detection.A kind of consideration electric power users week electricity consumption behavioral characteristic in multiplexing electric abnormality detection is provided, multiplexing electric abnormality detection complexity is reduced, promotes the low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature of the accuracy of multiplexing electric abnormality Data Detection.From all electricity consumption measure features of low-voltage customer, construct all electricity consumption indicatrixes of user, based on FCM clustering algorithm, obtain all electricity consumption curves of typical case of low-voltage electricity user, the low-voltage customer for marking abnormal electricity consumption using the method based on distance again, monitors multiplexing electric abnormality user finally by survey electric current, voltage, equilibrium data is actively called together.Present invention combination power information acquisition system obtains daily electricity data, and is marked according to abnormal user, calls together per hour to abnormal user and surveys electric current, voltage and equilibrium data, realizes the reliable monitoring of multiplexing electric abnormality behavior.
Description
Technical field
The invention belongs to technical field of electric power detection more particularly to a kind of electricity consumption based on low-voltage customer week electrical feature are different
Normal detection method.
Background technique
As power grid scale constantly expands, electric power information degree is continuously improved and Internet of Things, cloud computing, number
According to the control of a new generation such as excavation, the continuous development of measurement and data processing technique, big data application has been directed to electric power enterprise
Each business scope.
Currently, lack the abnormal electricity consumption behavioral study for being directed to low-voltage customer, this is because for low-voltage customer, electricity
Frequency acquisition generally once every hour, and hour electricity data quality cannot often ensure, for the authentic data of research
Only daily power consumption reduces detection reliability.
Summary of the invention
The present invention is in view of the above problems, provide a kind of consideration electric power users week electricity consumption behavior spy in multiplexing electric abnormality detection
Point reduces multiplexing electric abnormality detection complexity, promoted the accuracy of multiplexing electric abnormality Data Detection based on user's week electrical feature
Low-voltage customer multiplexing electric abnormality detection method.
The technical scheme adopted by the invention is that: the following steps are included:
1) all electricity consumption indicatrixes of low-voltage customer are established;It is used based on low pressure different types of in power information acquisition system
The daily electricity information at family, the typical all electricity consumption indicatrixes of building user;
2) all electricity consumption indicatrixes typical to low-voltage customer carry out fuzzy C-means clustering, obtain cluster centre;
3) all electricity consumption curves of low-voltage customer and cluster centre are subjected to distance metric, and multiplexing electric abnormality user is sieved
Choosing;
It is abnormal to think that data exist if threshold value is greater than constant η, marks abnormal user data;Otherwise it is assumed that being normal number
According to process terminates.
Step 1) the following steps are included:
1.1) all electricity consumption weighted averages;
1.2) all electricity consumption data normalizations.
Step 2) the following steps are included:
2.1) subordinated-degree matrix U is initialized;
2.2) c cluster centre Vi is calculated;
2.3) subordinated-degree matrix U is updated;
2.4) cost function J is calculated;
2.5) iterative step 2.2) and step 2.4) go to step 2.6) when J < threshold epsilon;
2.6) algorithm terminates, and exports c cluster centre V1, V2 ... Vc.
The present invention utilizes all electricity consumption indicatrixes of daily electricity basic data structuring user's, and poly- based on fuzzy C-mean algorithm
Class algorithm carries out clustering, obtains all electricity consumption curves of typical case of low-voltage electricity user, then use the method mark based on distance
The low-voltage customer for remembering abnormal electricity consumption monitors multiplexing electric abnormality user by actively calling survey electric current, voltage, equilibrium data together.
Compared with prior art, the present invention at least has the advantages that
1) compared with traditional multiplexing electric abnormality detection method, detection method of the invention goes out from all electricity consumption indicatrixes of user
Hair, the typical all electricity consumption indicatrixes of building user;
2) it after method of the invention is based on FCM clustering algorithm realization user's electric energy measurement data abnormality detection, can mark
Abnormal user realizes the reliable monitoring of abnormal user electricity consumption behavior.
Detailed description of the invention
Fig. 1 is the flow diagram of low-voltage customer multiplexing electric abnormality detection method of the invention.
Specific embodiment
The present invention is as shown in Figure 1, comprising the following steps:
1) all electricity consumption indicatrixes of low-voltage customer are established: being used based on low pressure different types of in power information acquisition system
The daily electricity information at family, the typical all electricity consumption indicatrixes of building user;
In, by the date of user's daily electricity, go out what day belongs to according to Cai Le formula to calculating, abscissa is 1-7, is indulged
Coordinate is daily electricity.Wherein, typical selection is chosen according to sample data.
The effect of step 1 is the sample clustered.
2) all electricity consumption indicatrixes typical to low-voltage customer carry out fuzzy C-means clustering, obtain cluster centre;
3) all electricity consumption curves of low-voltage customer and cluster centre are subjected to distance metric, and multiplexing electric abnormality user is sieved
Choosing.
Preferably, in step 1), the different type of the low-voltage customer includes that low pressure resident and low pressure are non-resident
User;
Preferably, step 1) the following steps are included:
1.1) all electricity consumption weighted averages:
The electricity consumption data for enabling user i-th week are Wi={ wi1, wi2, wi3, wi4, wi5, wi6, wi7, it extracts n weeks go through altogether
The Records of the Historian is recorded, then the weight coefficient in i-th week jth day are as follows:J=1,2 ... 7, then weighted average of the user at j days
Electricity consumption are as follows:
1.2) data set normalizes:
If data set X={ x1, x2... xn, a shared n data object, xi=(xi1, xi2... xim), share m category
Property.After then normalizingWherein max { xij, min { xijBe j-th attribute maximum value and most
Small value.
Preferably, in step 2), all electricity consumption data of different type user is clustered based on FCM algorithm, are obtained
To cluster centre collection, basis early period is established for user power utilization anomaly analysis;
Preferably, the number of cluster is c, cluster centre V1, V2... Vc;
Preferably, step 2) the following steps are included:
2.1) subordinated-degree matrix U is initialized, it is made to meet condition
2.2) c cluster centre v is calculatedi
Wherein fuzzy coefficient m value is 5;
2.3) subordinated-degree matrix U is updated,
Wherein fuzzy coefficient m value is 5;
2.4) cost function is calculated,
2.5) iterative step 2.2) and step 2.4), using the changing value J of cost function as clustering performance decision condition, when
When J < ε, approximate representation is that electricity consumption Model tying center Vn is no longer changed, and goes to step 2.6);Wherein, ε is in Limit
Middle representative be one greater than 0 very little number, can be arbitrarily small, as long as be not equal to zero;
Preferably, depending on the specific value of ε is with actually detected precision, the number of iterations becomes more or does not restrain if value is too small,
Value is excessive, causes the omission of certain abnormal datas;More preferably ε=0.15
2.6) algorithm terminates, export c cluster centre V1, V2... Vc。
Preferably, in step 3), abnormal electricity consumption user is marked with Euclidean distance;If min (dist (x,
Vi)) then to think that data exist abnormal by > η, mark user data;Otherwise it is assumed that being normal data, process terminates.Distance metric
Min (dist (x, Vi)) be x and Vi distance minimum value, selected distance measure it is simple, practical.
Preferably, depending on the specific value of η is with actually detected precision, abnormal data becomes more if value is too small, and value is excessive
Then cause the omission of certain abnormal datas;It is highly preferred that η=1.5.
As optimal technical scheme, described detection method includes the following steps:
1) all electricity consumption indicatrixes of low-voltage customer are established: being used based on low pressure different types of in power information acquisition system
The daily electricity information at family, the typical all electricity consumption indicatrixes of building user;The different type of low-voltage customer includes low pressure resident
With the non-resident user of low pressure (including general commercial user and general industry user).The following steps are included:
1.1) all electricity consumption weighted averages;All load curve data are weighted and averaged, available each user's
Typical week load curve.
1.2) all electricity consumption data normalizations.In electricity data, user is due to type difference, and there are larger differences for electricity consumption
It is different, if in cluster process, the big attribute of the order of magnitude will affect very greatly cluster result using electricity consumption as cluster feature,
Data are normalized so generally requiring, data are limited in [0,1] range.
2) it is based on fuzzy C-means clustering, obtains cluster centre;The number of cluster is c, cluster centre V1, V2... Vc;Packet
Include following steps:
2.1) subordinated-degree matrix U is initialized;
2.2) c cluster centre V is calculatedi;
2.3) subordinated-degree matrix U is updated;
2.4) cost function J is calculated;
2.5) iterative step 2.2) and step 2.4) go to step 2.6) as J < ε;
2.6) algorithm terminates, and exports c cluster centre V1, V2... Vc。
3) all electricity consumption curves of low-voltage customer and cluster centre are subjected to distance metric, if η is apart from detection threshold value min
It is abnormal that (dist (x, Vi)) > η then thinks that data exist, and marks user data;Otherwise it is assumed that being normal data, process terminates.Its
In, η=1.5;
It should be noted that and understand, in the feelings for not departing from the spirit and scope of the present invention required by appended claims
Under condition, various modifications and improvements can be made to the present invention of foregoing detailed description.It is therefore desirable to the model of the technical solution of protection
It encloses and is not limited by given any specific exemplary teachings.
The Applicant declares that the above is only a preferred embodiment of the present invention, it is noted that for the art
For those of ordinary skill, without departing from the inventive concept of the premise, several improvement and deformations can also be made, these improvement
It also should be regarded as protection scope of the present invention with deformation.
Claims (3)
1. a kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature, which is characterized in that including following step
It is rapid:
1) all electricity consumption indicatrixes of low-voltage customer are established;Based on low-voltage customer different types of in power information acquisition system
Daily electricity information, the typical all electricity consumption indicatrixes of building user;
2) all electricity consumption indicatrixes typical to low-voltage customer carry out fuzzy C-means clustering, obtain cluster centre;
3) all electricity consumption curves of low-voltage customer and cluster centre are subjected to distance metric, and multiplexing electric abnormality user is screened;
It is abnormal to think that data exist if threshold value is greater than constant η, marks abnormal user data;Otherwise it is assumed that be normal data, stream
Journey terminates.
2. a kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature according to claim 1,
Be characterized in that, step 1) the following steps are included:
1.1) all electricity consumption weighted averages;
1.2) all electricity consumption data normalizations.
3. a kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature according to claim 1,
Be characterized in that, step 2) the following steps are included:
2.1) subordinated-degree matrix U is initialized;
2.2) c cluster centre Vi is calculated;
2.3) subordinated-degree matrix U is updated;
2.4) cost function J is calculated;
2.5) iterative step 2.2) and step 2.4) go to step 2.6) when J < threshold epsilon;
2.6) algorithm terminates, and exports c cluster centre V1, V2 ... Vc.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811493614.2A CN109636667A (en) | 2018-12-07 | 2018-12-07 | A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811493614.2A CN109636667A (en) | 2018-12-07 | 2018-12-07 | A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109636667A true CN109636667A (en) | 2019-04-16 |
Family
ID=66071820
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811493614.2A Pending CN109636667A (en) | 2018-12-07 | 2018-12-07 | A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109636667A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110083986A (en) * | 2019-05-21 | 2019-08-02 | 国网湖南省电力有限公司 | Electrified energy-consuming device, which is opposed electricity-stealing, again simulates monitoring method, system, equipment and medium |
CN110321934A (en) * | 2019-06-12 | 2019-10-11 | 深圳供电局有限公司 | Method and system for detecting abnormal data of user electricity consumption |
CN110610121A (en) * | 2019-06-20 | 2019-12-24 | 国网重庆市电力公司 | Small-scale source load power abnormal data identification and restoration method based on curve clustering |
CN110991555A (en) * | 2019-12-16 | 2020-04-10 | 国网上海市电力公司 | Method for monitoring abnormal electricity consumption of user in typical industry |
CN111178556A (en) * | 2019-12-25 | 2020-05-19 | 深圳供电局有限公司 | Electric quantity abnormality detection method and device, computer equipment and readable storage medium |
CN112925827A (en) * | 2021-03-04 | 2021-06-08 | 南京怡晟安全技术研究院有限公司 | User property abnormity analysis method based on power acquisition Internet of things data |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104809255A (en) * | 2015-05-21 | 2015-07-29 | 国家电网公司 | Load shape acquisition method and system |
CN105678314A (en) * | 2015-10-15 | 2016-06-15 | 东南大学 | Typical demand-side user screening method based on fuzzy C clustering |
CN106204335A (en) * | 2016-07-21 | 2016-12-07 | 广东工业大学 | A kind of electricity price performs abnormality judgment method, Apparatus and system |
CN106682079A (en) * | 2016-11-21 | 2017-05-17 | 云南电网有限责任公司电力科学研究院 | Detection method of user's electricity consumption behavior of user based on clustering analysis |
CN107220906A (en) * | 2017-05-31 | 2017-09-29 | 国网上海市电力公司 | Multiple Time Scales multiplexing electric abnormality analysis method based on electricity consumption acquisition system |
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 |
CN107730395A (en) * | 2017-09-13 | 2018-02-23 | 国网天津市电力公司电力科学研究院 | A kind of multiplexing electric abnormality detection method based on power consumption deviation ratio for low-voltage customer |
-
2018
- 2018-12-07 CN CN201811493614.2A patent/CN109636667A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104809255A (en) * | 2015-05-21 | 2015-07-29 | 国家电网公司 | Load shape acquisition method and system |
CN105678314A (en) * | 2015-10-15 | 2016-06-15 | 东南大学 | Typical demand-side user screening method based on fuzzy C clustering |
CN106204335A (en) * | 2016-07-21 | 2016-12-07 | 广东工业大学 | A kind of electricity price performs abnormality judgment method, Apparatus and system |
CN106682079A (en) * | 2016-11-21 | 2017-05-17 | 云南电网有限责任公司电力科学研究院 | Detection method of user's electricity consumption behavior of user based on clustering analysis |
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 |
CN107220906A (en) * | 2017-05-31 | 2017-09-29 | 国网上海市电力公司 | Multiple Time Scales multiplexing electric abnormality analysis method based on electricity consumption acquisition system |
CN107730395A (en) * | 2017-09-13 | 2018-02-23 | 国网天津市电力公司电力科学研究院 | A kind of multiplexing electric abnormality detection method based on power consumption deviation ratio for low-voltage customer |
Non-Patent Citations (1)
Title |
---|
王鹏伍 等: "模糊C均值聚类算法在用电异常稽查中的应用", 《华北电力技术》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110083986A (en) * | 2019-05-21 | 2019-08-02 | 国网湖南省电力有限公司 | Electrified energy-consuming device, which is opposed electricity-stealing, again simulates monitoring method, system, equipment and medium |
CN110321934A (en) * | 2019-06-12 | 2019-10-11 | 深圳供电局有限公司 | Method and system for detecting abnormal data of user electricity consumption |
CN110610121A (en) * | 2019-06-20 | 2019-12-24 | 国网重庆市电力公司 | Small-scale source load power abnormal data identification and restoration method based on curve clustering |
CN110610121B (en) * | 2019-06-20 | 2023-04-07 | 国网重庆市电力公司 | Small-scale source load power abnormal data identification and restoration method based on curve clustering |
CN110991555A (en) * | 2019-12-16 | 2020-04-10 | 国网上海市电力公司 | Method for monitoring abnormal electricity consumption of user in typical industry |
CN111178556A (en) * | 2019-12-25 | 2020-05-19 | 深圳供电局有限公司 | Electric quantity abnormality detection method and device, computer equipment and readable storage medium |
CN112925827A (en) * | 2021-03-04 | 2021-06-08 | 南京怡晟安全技术研究院有限公司 | User property abnormity analysis method based on power acquisition Internet of things data |
CN112925827B (en) * | 2021-03-04 | 2024-05-10 | 南京怡晟安全技术研究院有限公司 | User property anomaly analysis method based on electric power acquisition internet of things data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109636667A (en) | A kind of low-voltage customer multiplexing electric abnormality detection method based on user's week electrical feature | |
CN111199016B (en) | Daily load curve clustering method for improving K-means based on DTW | |
CN110336534B (en) | Fault diagnosis method based on photovoltaic array electrical parameter time series feature extraction | |
CN105678404B (en) | Based on online shopping electricity and dynamically associate the micro-grid load forecasting system and method for the factor | |
CN113554466B (en) | Short-term electricity consumption prediction model construction method, prediction method and device | |
CN105913079B (en) | Electronic nose isomeric data recognition methods based on the study of the target domain migration limit | |
CN110781332A (en) | Electric power resident user daily load curve clustering method based on composite clustering algorithm | |
CN106446967A (en) | Novel power system load curve clustering method | |
CN114330583B (en) | Abnormal electricity utilization identification method and abnormal electricity utilization identification system | |
CN106067034B (en) | Power distribution network load curve clustering method based on high-dimensional matrix characteristic root | |
Zhang et al. | Short-term power load forecasting using integrated methods based on long short-term memory | |
CN112307675B (en) | Neural network-based temperature-sensitive load separation identification method and system | |
CN110738232A (en) | grid voltage out-of-limit cause diagnosis method based on data mining technology | |
Rehman et al. | Comparative evaluation of machine learning models and input feature space for non-intrusive load monitoring | |
Qiu et al. | Failure rate prediction of electrical meters based on weighted hierarchical Bayesian | |
CN116148753A (en) | Intelligent electric energy meter operation error monitoring system | |
CN112418476A (en) | Ultra-short-term power load prediction method | |
CN114417971A (en) | Electric power data abnormal value detection algorithm based on K nearest neighbor density peak clustering | |
CN105653670B (en) | Intelligent electricity consumption data mining method based on manifold learning clustering algorithm | |
de Diego-Otón et al. | Recurrent LSTM architecture for appliance identification in non-intrusive load monitoring | |
CN112288157A (en) | Wind power plant power prediction method based on fuzzy clustering and deep reinforcement learning | |
CN114818827A (en) | Non-invasive load decomposition method based on seq2point network | |
CN118152857A (en) | Power consumption abnormality detection method and device and computer readable storage medium | |
CN114240687A (en) | Energy hosting efficiency analysis method suitable for comprehensive energy system | |
CN113987910A (en) | Method and device for identifying load of residents by coupling neural network and dynamic time planning |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190416 |