CN109615004A - A kind of anti-electricity-theft method for early warning of multisource data fusion - Google Patents
A kind of anti-electricity-theft method for early warning of multisource data fusion Download PDFInfo
- Publication number
- CN109615004A CN109615004A CN201811493693.7A CN201811493693A CN109615004A CN 109615004 A CN109615004 A CN 109615004A CN 201811493693 A CN201811493693 A CN 201811493693A CN 109615004 A CN109615004 A CN 109615004A
- Authority
- CN
- China
- Prior art keywords
- electricity
- data
- information
- electricity consumption
- analysis
- 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
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The present invention provides a kind of anti-electricity-theft method for early warning of multisource data fusion, to realize the analysis and pre-alarming system that are directed to stealing and abnormal electricity consumption behavior that user occurs during electricity consumption.The system monitoring and prediction and warning diagnosis capability of electricity consumption behavior, the support technology applied with electricity consumption mass data are improved in this method using multi-source composite information；Multilayer Perception system is established, the integrated multi-source of power information system, deep layer extraction and omnidirection application are promoted；Monitoring and the cognitive method analyzed based on user power utilization are proposed simultaneously, and the related resources such as electricity consumption classification, electricity consumption strategy and electricity consumption behavioural norm are distributed in raising rationally；It proposes the discrimination method to user power utilization behavior property, provides technical support for power grid security, maintenance electricity marketing.
Description
Technical field
The present invention is to belong to electric network data analysis field, more accurately for be a kind of the anti-electricity-theft pre- of multisource data fusion
Alarm method.
Background technique
As most preposition, the most basic constitution element for facing user directly in smart grid composition, power consumer telecommunications
Breath acquisition system is the important and basic ring on electric power enterprise " big data " chain.Therefore in current smart electric grid system
Stealing and abnormal electricity consumption behavioural analysis are needed the acquisition of vast power consumer electricity consumption information data, and utilize the magnanimity acquired
Data combination marketing information system and other associated datas carry out reasonable drawing and analysis, form analysis conclusion.And how into
The acquisition of row data is extracted, is analyzed still without preferable implementation method.
Summary of the invention
Goal of the invention: to solve the above problems, the present invention provides a kind of anti-electricity-theft method for early warning of multisource data fusion,
To improve the system monitoring and prediction and warning diagnosis capability of electricity consumption behavior.
Technical solution: to achieve the above object, following technical solution can be used in the present invention.
A kind of anti-electricity-theft method for early warning of multisource data fusion, including Data acquisition and Proclssing, feature extraction and display, mould
Type analysis, aid decision；Wherein,
Data acquisition and Proclssing is arranged for data source, obtains valid data to obtain stealing typical case,
Collect marketing system customer default electricity consumption stealing related data information；
Feature extraction and display are by extracting user power utilization behavioural characteristic displaying power information；
Model analysis passes through abnormal information about power and carries out comprehensive analysis, obtains abnormal information about power model；Pass through magnanimity
Power information acquisition system extracts customer electricity data, and selected part stealing client's example carries out mould using logistic regression algorithm
The relevant technology beforehand research of type sample training；Build client's stealing probability big data analysis model；
Aid decision generates abnormal electricity consumption stealing report.
Further, the power information for extracting the displaying of user power utilization behavioural characteristic includes customer information, customers' credit
Information；Customer information includes user's classification, electricity consumption classification, trade classification, voltage class, contract capacity, importance rate, season
Property increase and decrease is held, whether is associated with power generation client, power generation client connects people's capacity, power generation client's grid-connected voltage grade, average moon electric flux,
The average moon electricity charge；Customers' credit information has included whether arrearage record, arrearage number, whether has had default electricity use record；Client uses
Electric behavioural information analyzes the electricity consumption behavior of client according to the acquisition data obtained from electric energy meter, including electric flux it is differential it is abnormal,
Electric energy meter cover opening, electric energy meter stop walking, three-phase imbalance, overcurrent, electric sampling open-phase, electric energy meter decompression, electric energy meter defluidization, electricity consumption
The information such as load.
Further, feature extraction and comparative studies are carried out using following methods, including,
Principal component method；
Based on self-organizing map neural network；
Former sequence is input in Recognition with Recurrent Neural Network one by one based on Recognition with Recurrent Neural Network and is used for regression forecasting, by per a period of time
It carves obtained hidden state all to converge, the data after then being converted using average Chi Hualai dimensionality reduction；
Further, after feature extraction, by the output of principal component analysis and self-organizing map neural network, display is special
Levy distribution situation of the data in two-dimensional space or three-dimensional space.
Further, the analysis of client's stealing probability big data analysis model, which builds to acquire based on power information, is
System, using clustering methodology as multivariate statistics tool, including K-means algorithm, aggregate clustering algorithm and EM algorithm system
Clustering procedure of uniting and K mean cluster method；Customer electricity data, selected part stealing visitor are extracted by magnanimity power information acquisition system
Family example carries out the relevant technology beforehand research of model sample training using logistic regression algorithm.
The utility model has the advantages that the present invention provides a kind of anti-electricity-theft method for early warning of multisource data fusion, based on user in electricity consumption
The probability early warning analysis of the anti-electricity-theft and multiplexing electric abnormality behavior occurred in journey can be based on power information acquisition system and marketing industry
A large amount of customer electricity information of application system of being engaged in accumulation, comprehensively consider various factors, establish customer electricity behavior probability analysis mould
Type carries out probability supposition and early warning to stealing suspicion family in real time by big data technology analysis means, is used in violation of rules and regulations electricity consumption enterprise
Electricity is precisely judged and trend analysis, and realization precisely identifies doubtful stealing family, establishes the Closed loop operation of early warning, investigation and processing
Mechanism, increases and punishes dynamics to the investigation of multiplexing electric abnormality behavior, and maintenance normally for electricity consumption order, ensures enterprise management efficiency.
Detailed description of the invention
Attached drawing 1 is system work structure chart.
Attached drawing 2 is functional framework figure.
Attached drawing 3 is data sampling figure.
Attached drawing 4 is that principal component is shown and SOM display schematic diagram.
Attached drawing 5 is clustering figure.
Specific embodiment
In the following, being described in further details in conjunction with attached drawing to the present invention.
Referring to FIG. 1, Fig. 1 is functional structure chart；
System establishes the data information collection around user power utilization by the collection to electricity consumption basic data.Pass through building
Simple stealing analysis model simultaneously chooses corresponding multiplexing electric abnormality classification indicators and delimit multiplexing electric abnormality range as foundation, together
When the user for the condition that meets is received in people's anomaly analysis pond, quantitative analysis is carried out according to multiplexing electric abnormality behavior property rule, respectively
It generates low pressure, specially become family stealing early warning detail；Then, according to flexibly customized oppose electricity-stealing expert tactics and user behavior track
It analyzes multiplexing electric abnormality situation and carries out comprehensive diagnos, such as have electricity stealing feature, great stealing suspicion family will be set as, and by system
Detailed abnormity diagnosis report is generated, live verification can be carried out according to report diagnosis with inspection personnel；Finally, will verification
As a result case library is fed back in system and be added to, case training analysis function sustainable improvement multiplexing electric abnormality attribution rule is passed through
And decision plan.The process flow of system is divided into on-line monitoring, online screening, on-line analysis, verification feedback, online feedback and holds
It is continuous to improve 6 parts.On-line monitoring mainly carries out quantification of targets to the multiplexing electric abnormality of selected multiplexing electric abnormality user；Online screening
Mainly by stealing decision rule to thering is the user of electricity stealing feature to screen in abnormal electricity consumption user；On-line analysis master
Its stealing suspicion is determined if carrying out electricity consumption action trail to the user for having electricity stealing feature and analyzing；It puts in order and investigates online and is anti-
Feedback mainly issues abnormal work order to stealing suspicion family online, feeds back result after carrying out live verification with inspection personnel；Persistently change
Into being to improve electricity consumption behavior property rule by the analysis of stealing case and oppose electricity-stealing to determine specially
Family's strategy.
Referring to FIG. 2, Fig. 2 is functional framework figure.
In system architecture, using the database service interface for supporting a variety of data exchange agreements, from marketing management system, meter
It measures automated system and battalion and obtains basic data information with different data sources such as information integrated platforms.In data processing, pass through
Mass data processing technology stores data, is merged, in combination with the research achievement to model of opposing electricity-stealing, algorithm and strategy
Concurrent processing and analysis are carried out to data, provide high-speed data access service to last time application；Meanwhile in data analysis layer
It plans integrated data platform of opposing electricity-stealing, provides data application support specifically to oppose electricity-stealing to apply.It constructs and opposes electricity-stealing in application layer
On-line monitoring and intelligent diagnostics analysis system, provide the stealing of low pressure family on-line monitoring analysis, specially become family stealing on-line monitoring analysis,
The analysis of user power utilization behavior integration assists oppose electricity-stealing information inquiry and statistical analysis and report capability.User can pass through
System real-time monitoring electricity stealing, and the configuration feature that can be provided by system is realized to stealing accumulation of knowledge and utilization.
Referring to FIG. 3, Fig. 3 is data sampling figure.
Under big data analysis background, especially using neural network as the machine learning analysis method of representative, need to occupy
Mass data.Wherein neural network is particularly suitable for the complex data analysis of multidimensional, to cope with the more of different complicated user types
Label analysis, needs to acquire a large amount of multi-dimensional datas.
The data that can be got include marketing subscriber profile data, five class data of power information acquisition system, have sent out
Raw stealing case data and the power load data of user etc..
Subscriber profile data of marketing includes customer information, customers' credit information etc., can be obtained by marketing system.Client
Information include user's classification, electricity consumption classification, trade classification, voltage class, contract capacity, importance rate, seasonal increase and decrease hold,
Whether association power generation client, power generation client connect people's capacity, power generation client's grid-connected voltage grade, average moon electric flux, average moon electricity
Take；Customers' credit information included whether arrearage record, arrearage number, whether have default electricity use record etc..
Customer electricity behavioural information obtains.According to the acquisition data obtained from electric energy meter, the electricity consumption behavior of client is analyzed,
Stop walking including the differential exception of electric flux, electric energy meter cover opening, electric energy meter, three-phase imbalance, overcurrent, electric sampling open-phase, electric energy meter
The information such as decompression, electric energy meter defluidization, power load.
With popularizing for intelligent electric meter, what the electricity consumption of user can be convenient is obtained, it might even be possible to detect in a few minutes
Once.It, being capable of a more complete description user characteristics by the pretreatment to user power consumption load data.Pass through the sampling interval
Difference can be converted into user power utilization load data a minute electricity consumption, hour electricity consumption, daily power consumption, moon electricity consumption, Ji Yong
Electricity and year electricity consumption form pyramid feature, so that user power utilization mode is preferably described in different time dimension,
It lays the foundation for following model analysis.
Referring to FIG. 4, Fig. 4 is that principal component is shown and SOM display schematic diagram.
Good feature, can be anti-interference, and can improve the accuracy of model, while reducing model complexity, improves model
Generalization Capability, so extract feature, be an important component of this project research.The following method of proposed adoption carries out special
Sign is extracted and comparative studies.
Principal component method (PCA)
PCA is that each feature of comprehensive former data recalculates a set of new entirely different integration characteristics to illustrate original
There are data.The objective function of PCA is to keep error sum of squares minimum, it is desirable to can find out the low-rank table to former data using this target
It reaches, so that not losing many information after being mapped to higher-dimension, still can be very good expression data.
Based on Self-organizing Maps (SOM) neural network
The dimensionality reduction mapping from the input space (n dimension) to output plane (2 dimension) may be implemented in SOM network, and maps to have and open up
Flutter feature retention properties.And since the spatial position of SOM network output node embodies the inner link of input sample, that is, have
The input of like attribute is mapped on neighbouring SOM output node, so the visualization that the output of SOM can be used for clustering again.
Former sequence is input in LSTM one by one based on Recognition with Recurrent Neural Network (LSTM) and is used for regression forecasting, by each moment
Obtained hidden state all converges, and the data after then being converted using average Chi Hualai dimensionality reduction, the data can be made
It is characterized with the clustering methods such as k-means later.After feature extraction, if higher-dimension can be shown in two dimension or three-dimensional space
The data distribution of feature, the judgement that can be characterized quality provide intuitive foundation.Pass through principal component analysis (PCA) and SOM mind
Output through network can show distribution situation of the characteristic in two-dimensional space or three-dimensional space.
Referring to FIG. 5, Fig. 5 is clustering figure.
By the experiment to a variety of clustering methods and compare, the smallest clustering method of error identifying with display, and passes through conjunction
Suitable cluster display methods provides the intuitive display effect of user power utilization feature, provides auxiliary for further aid decision and refer to
Mark.The research of clustering method focuses primarily upon the higher K-means clustering method of traditional hierarchy clustering method and efficiency.
Claims (5)
1. a kind of anti-electricity-theft method for early warning of multisource data fusion, it is characterised in that:
Including Data acquisition and Proclssing, feature extraction and display, model analysis, aid decision；
Wherein,
Data acquisition and Proclssing is arranged for data source, is obtained valid data to obtain stealing typical case, that is, is received
Collect marketing system customer default electricity consumption stealing related data information；
Feature extraction and display are by extracting user power utilization behavioural characteristic displaying power information；
Model analysis passes through abnormal information about power and carries out comprehensive analysis, obtains abnormal information about power model；Pass through magnanimity electricity consumption
Information acquisition system extracts customer electricity data, and selected part stealing client's example carries out model sample using logistic regression algorithm
The relevant technology beforehand research of this training；Build client's stealing probability big data analysis model；
Aid decision generates abnormal electricity consumption stealing report.
2. anti-electricity-theft method for early warning according to claim 1, it is characterised in that: the extraction user power utilization behavioural characteristic exhibition
The power information shown includes customer information, customers' credit information；Customer information include user's classification, electricity consumption classification, trade classification,
Voltage class, contract capacity, importance rate, seasonal increase and decrease is held, whether is associated with power generation client, power generation client connects people's capacity,
Generate electricity client's grid-connected voltage grade, average moon electric flux, the average moon electricity charge；Customers' credit information included whether arrearage record,
Whether arrearage number has default electricity use record；Customer electricity behavioural information is according to the acquisition data obtained from electric energy meter, analysis
The electricity consumption behavior of client, including the differential exception of electric flux, electric energy meter cover opening, electric energy meter stop walking, three-phase imbalance, overcurrent,
The information such as electric sampling open-phase, electric energy meter decompression, electric energy meter defluidization, power load.
3. anti-electricity-theft method for early warning according to claim 2, it is characterised in that: carry out feature extraction using following methods
And comparative studies, including,
Principal component method；
Based on self-organizing map neural network；
Former sequence is input in Recognition with Recurrent Neural Network one by one based on Recognition with Recurrent Neural Network and is used for regression forecasting, each moment is obtained
To hidden state all converge, the data after then being converted using average Chi Hualai dimensionality reduction.
4. anti-electricity-theft method for early warning according to claim 3, it is characterised in that: after feature extraction, pass through principal component point
The output of analysis and self-organizing map neural network shows distribution situation of the characteristic in two-dimensional space or three-dimensional space.
5. anti-electricity-theft method for early warning according to claim 4, it is characterised in that: client's stealing probability big data analysis
The analysis of model is built based on power information acquisition system, using clustering methodology as multivariate statistics tool, including K-
Means algorithm, aggregate clustering algorithm and EM algorithmic system clustering procedure and K mean cluster method；It is acquired by magnanimity power information
System extracts customer electricity data, and selected part stealing client's example carries out model sample training phase using logistic regression algorithm
The technology beforehand research of pass.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811493693.7A CN109615004A (en) | 2018-12-07 | 2018-12-07 | A kind of anti-electricity-theft method for early warning of multisource data fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811493693.7A CN109615004A (en) | 2018-12-07 | 2018-12-07 | A kind of anti-electricity-theft method for early warning of multisource data fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109615004A true CN109615004A (en) | 2019-04-12 |
Family
ID=66008465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811493693.7A Pending CN109615004A (en) | 2018-12-07 | 2018-12-07 | A kind of anti-electricity-theft method for early warning of multisource data fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615004A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045209A (en) * | 2019-05-10 | 2019-07-23 | 广东电网有限责任公司 | Detection method, device, equipment and the readable storage medium storing program for executing of electricity consumption data exception |
| CN110082579A (en) * | 2019-05-21 | 2019-08-02 | 国网湖南省电力有限公司 | A kind of area's Intelligent power-stealing prevention monitoring method, system, equipment and medium |
| CN110223196A (en) * | 2019-06-04 | 2019-09-10 | 国网浙江省电力有限公司电力科学研究院 | Analysis method of opposing electricity-stealing based on typical industry feature database and sample database of opposing electricity-stealing |
| CN110516011A (en) * | 2019-08-28 | 2019-11-29 | 北京思维造物信息科技股份有限公司 | A kind of multi-source solid data fusion method, device and equipment |
| CN111008193A (en) * | 2019-12-03 | 2020-04-14 | 国网天津市电力公司电力科学研究院 | Data cleaning and quality evaluation method and system |
| CN111489073A (en) * | 2020-03-31 | 2020-08-04 | 深圳市康拓普信息技术有限公司 | Classification algorithm-based user electricity consumption price situation early warning method |
| CN111506636A (en) * | 2020-05-12 | 2020-08-07 | 上海积成能源科技有限公司 | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm |
| CN111525697A (en) * | 2020-05-09 | 2020-08-11 | 西安交通大学 | Medium and low voltage power distribution network electricity larceny prevention method and system based on current monitoring and line topology analysis |
| CN111612054A (en) * | 2020-05-14 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | User electricity stealing behavior identification method based on non-negative matrix factorization and density clustering |
| CN112418623A (en) * | 2020-11-12 | 2021-02-26 | 国网河南省电力公司郑州供电公司 | Anti-electricity-stealing identification method based on bidirectional long-time and short-time memory network and sliding window input |
| CN113724098A (en) * | 2021-07-30 | 2021-11-30 | 国网山东省电力公司济南供电公司 | Clustering and neural network-based electricity stealing user detection method and system |
| CN114154999A (en) * | 2021-10-27 | 2022-03-08 | 国网河北省电力有限公司营销服务中心 | Electricity stealing prevention method, device, terminal and storage medium |
| WO2023109527A1 (en) * | 2021-12-17 | 2023-06-22 | 广东电网有限责任公司东莞供电局 | Electricity theft behavior detection method and apparatus, computer device and storage medium |
| 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 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015158198A1 (en) * | 2014-04-17 | 2015-10-22 | 北京泰乐德信息技术有限公司 | Fault recognition method and system based on neural network self-learning |
| CN107145966A (en) * | 2017-04-12 | 2017-09-08 | 山大地纬软件股份有限公司 | Logic-based returns the analysis and early warning method of opposing electricity-stealing of probability analysis Optimized model |
| CN107507076A (en) * | 2017-09-26 | 2017-12-22 | 贵州电网有限责任公司 | The method of the composite rating of power customer based on data mining |
| CN107633050A (en) * | 2017-09-18 | 2018-01-26 | 安徽蓝杰鑫信息科技有限公司 | A kind of method that stealing probability is judged based on big data analysis electricity consumption behavior |
| CN107742127A (en) * | 2017-10-19 | 2018-02-27 | 国网辽宁省电力有限公司 | A kind of improved anti-electricity-theft intelligent early-warning system and method |
-
2018
- 2018-12-07 CN CN201811493693.7A patent/CN109615004A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015158198A1 (en) * | 2014-04-17 | 2015-10-22 | 北京泰乐德信息技术有限公司 | Fault recognition method and system based on neural network self-learning |
| CN107145966A (en) * | 2017-04-12 | 2017-09-08 | 山大地纬软件股份有限公司 | Logic-based returns the analysis and early warning method of opposing electricity-stealing of probability analysis Optimized model |
| CN107633050A (en) * | 2017-09-18 | 2018-01-26 | 安徽蓝杰鑫信息科技有限公司 | A kind of method that stealing probability is judged based on big data analysis electricity consumption behavior |
| CN107507076A (en) * | 2017-09-26 | 2017-12-22 | 贵州电网有限责任公司 | The method of the composite rating of power customer based on data mining |
| CN107742127A (en) * | 2017-10-19 | 2018-02-27 | 国网辽宁省电力有限公司 | A kind of improved anti-electricity-theft intelligent early-warning system and method |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110045209A (en) * | 2019-05-10 | 2019-07-23 | 广东电网有限责任公司 | Detection method, device, equipment and the readable storage medium storing program for executing of electricity consumption data exception |
| CN110082579A (en) * | 2019-05-21 | 2019-08-02 | 国网湖南省电力有限公司 | A kind of area's Intelligent power-stealing prevention monitoring method, system, equipment and medium |
| CN110223196A (en) * | 2019-06-04 | 2019-09-10 | 国网浙江省电力有限公司电力科学研究院 | Analysis method of opposing electricity-stealing based on typical industry feature database and sample database of opposing electricity-stealing |
| CN110516011B (en) * | 2019-08-28 | 2022-06-21 | 北京思维造物信息科技股份有限公司 | Multi-source entity data fusion method, device and equipment |
| CN110516011A (en) * | 2019-08-28 | 2019-11-29 | 北京思维造物信息科技股份有限公司 | A kind of multi-source solid data fusion method, device and equipment |
| CN111008193A (en) * | 2019-12-03 | 2020-04-14 | 国网天津市电力公司电力科学研究院 | Data cleaning and quality evaluation method and system |
| CN111008193B (en) * | 2019-12-03 | 2023-10-31 | 国网天津市电力公司电力科学研究院 | Data cleaning and quality evaluation method and system |
| CN111489073A (en) * | 2020-03-31 | 2020-08-04 | 深圳市康拓普信息技术有限公司 | Classification algorithm-based user electricity consumption price situation early warning method |
| CN111525697A (en) * | 2020-05-09 | 2020-08-11 | 西安交通大学 | Medium and low voltage power distribution network electricity larceny prevention method and system based on current monitoring and line topology analysis |
| CN111525697B (en) * | 2020-05-09 | 2022-10-25 | 西安交通大学 | Medium and low voltage power distribution network electricity larceny prevention method and system based on current monitoring and line topology analysis |
| CN111506636A (en) * | 2020-05-12 | 2020-08-07 | 上海积成能源科技有限公司 | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm |
| CN111612054A (en) * | 2020-05-14 | 2020-09-01 | 国网河北省电力有限公司电力科学研究院 | User electricity stealing behavior identification method based on non-negative matrix factorization and density clustering |
| CN112418623A (en) * | 2020-11-12 | 2021-02-26 | 国网河南省电力公司郑州供电公司 | Anti-electricity-stealing identification method based on bidirectional long-time and short-time memory network and sliding window input |
| CN113724098A (en) * | 2021-07-30 | 2021-11-30 | 国网山东省电力公司济南供电公司 | Clustering and neural network-based electricity stealing user detection method and system |
| CN113724098B (en) * | 2021-07-30 | 2023-10-13 | 国网山东省电力公司济南供电公司 | Method and system for detecting electricity stealing users based on clustering and neural network |
| CN114154999A (en) * | 2021-10-27 | 2022-03-08 | 国网河北省电力有限公司营销服务中心 | Electricity stealing prevention method, device, terminal and storage medium |
| WO2023109527A1 (en) * | 2021-12-17 | 2023-06-22 | 广东电网有限责任公司东莞供电局 | Electricity theft behavior detection method and apparatus, computer device and storage medium |
| 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 |
|---|---|---|
| CN109615004A (en) | A kind of anti-electricity-theft method for early warning of multisource data fusion | |
| CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
| Ahmad et al. | Review of various modeling techniques for the detection of electricity theft in smart grid environment | |
| CN106093707B (en) | The data processing method of intelligent electricity anti-theft analysis system | |
| CN104407589B (en) | Workshop manufacturing process-oriented active sensing and anomaly analysis method of real-time production performance | |
| CN110223196A (en) | Analysis method of opposing electricity-stealing based on typical industry feature database and sample database of opposing electricity-stealing | |
| Guo et al. | Complex power system status monitoring and evaluation using big data platform and machine learning algorithms: a review and a case study | |
| CN112859822B (en) | Equipment health analysis and fault diagnosis method and system based on artificial intelligence | |
| CN104156403B (en) | A kind of big data normal mode extracting method and system based on cluster | |
| CN110458230A (en) | A kind of distribution transforming based on the fusion of more criterions is with adopting data exception discriminating method | |
| CN110108914A (en) | One kind is opposed electricity-stealing intelligent decision making method, system, equipment and medium | |
| CN108959424A (en) | A kind of operating method of the city electricity consumption map for power system load monitoring | |
| CN108985467A (en) | Secondary device lean management-control method based on artificial intelligence | |
| Zou et al. | A novel network security algorithm based on improved support vector machine from smart city perspective | |
| Ünal et al. | Big data-driven detection of false data injection attacks in smart meters | |
| CN117132025A (en) | Power consumption monitoring and early warning system based on multisource data fusion | |
| CN115062725B (en) | Hotel income anomaly analysis method and system | |
| CN115099579A (en) | Marketing cue value evaluation method based on migrated federal learning | |
| Rosenlund et al. | Clustering and dimensionality-reduction techniques applied on power quality measurement data | |
| Poudel et al. | Artificial intelligence for energy fraud detection: a review | |
| CN106384308A (en) | Power consumption load data processing method and apparatus for power system | |
| Sidhu et al. | Clustering in data mining | |
| Xie et al. | Knowledge Acquisition for Transformer Condition Assessment Using Synthetic Minority Over-sampling Technique and Decision Tree Algorithm | |
| Lu et al. | Time series power anomaly detection based on Light Gradient Boosting Machine | |
| Benítez Sánchez et al. | Classification of customers based on temporal load profile patterns |
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 |