CN112649641B - Electricity stealing user judging method based on electricity stealing characteristics - Google Patents
Electricity stealing user judging method based on electricity stealing characteristics Download PDFInfo
- Publication number
- CN112649641B CN112649641B CN202011464556.8A CN202011464556A CN112649641B CN 112649641 B CN112649641 B CN 112649641B CN 202011464556 A CN202011464556 A CN 202011464556A CN 112649641 B CN112649641 B CN 112649641B
- Authority
- CN
- China
- Prior art keywords
- electricity
- user
- stealing
- time sequence
- electricity stealing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R11/00—Electromechanical arrangements for measuring time integral of electric power or current, e.g. of consumption
- G01R11/02—Constructional details
- G01R11/24—Arrangements for avoiding or indicating fraudulent use
-
- 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
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for judging electricity larceny users based on electricity larceny characteristics, which is characterized in that the electricity larceny judgment by utilizing historical data is realized by acquiring the electricity consumption time sequence of users to be diagnosed and the electricity consumption time sequence of typical electricity larceny users in the historical data, calculating the similarity according to electricity larceny indexes, primarily judging the users to be diagnosed, and calculating the similarity between the latest electricity consumption time sequence and the initial electricity consumption time sequence of the users which are primarily judged to be suspected of electricity larceny.
Description
Technical Field
The invention relates to a method for judging electricity stealing users based on electricity stealing characteristics, and belongs to the field of electricity stealing user judgment in the power industry.
Background
For a long time, the phenomena of electricity larceny, fraud and the like in society are often restrained, and the phenomena bring serious threat to the development of national economy, the standardization of enterprise operation management and power supply order, and even endanger the life and property safety of other users.
With the rapid development of economy, the power consumption is increased at a high speed, a few electric power users are driven by economic benefits, and illegal measures are used for stealing electricity, and the electricity is often used for a long time. Although power supply enterprises increase the research investment of the anti-electricity-stealing technology, the electricity-stealing technology is also greatly improved, more and more high-tech electricity-stealing modes are presented, and the characteristics of high concealment and high complexity are presented. In actual anti-electricity-theft work, the problem of difficult evidence collection still exists due to the hysteresis of field verification of an electric company, and how to pre-judge which mode is adopted by a user to steal electricity in advance is a key point of field accurate evidence collection.
At present, the electric power company relies on an electricity consumption information acquisition system to basically realize 'full acquisition and full coverage' of electricity consumption of users, can timely, completely and accurately control electricity consumption data and information of vast electric power users, and the electricity consumption information acquisition system acquires and deposits a large amount of historical data through operation for a plurality of years, but no electricity stealing user judging method utilizing the historical data exists at present.
Disclosure of Invention
The invention provides a method for judging electricity stealing users based on electricity stealing characteristics, which solves the problem.
In order to solve the technical problems, the invention adopts the following technical scheme:
a method for judging the user who steals electricity based on the characteristics of electricity stealing comprises,
acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed;
acquiring a power utilization time sequence of a typical power stealing user in historical data, and calculating a power stealing index of the typical power stealing user;
calculating the similarity A of the user electricity utilization time sequence to be diagnosed and the typical electricity utilization time sequence of the user electricity stealing according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user;
responding to the similarity A meeting a preset condition C, and calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed;
and responding to the similarity B meeting a preset condition D, and judging that the user to be diagnosed is a suspected user.
The user power time sequence to be diagnosed comprises an initial period of power time sequence and a last period of power time sequence.
The user types comprise large-scale private transformer users, medium-and small-scale private transformer users, three-phase general business users, single-phase general business users and resident users; the large-scale private transformer users and the medium-and small-scale private transformer users have consistent electricity stealing indexes, including electric quantity, current unbalance rate, voltage unbalance rate and active power; the three-phase general business users, the single-phase general business users and the resident users have consistent electricity stealing indexes, and the electricity is used as electric quantity.
The typical electricity stealing users are obtained by adopting a neighbor propagation clustering method, and the specific process is that,
according to the electricity stealing index, analyzing the electricity stealing characteristics of different electricity stealing modes of the user, and analyzing related electricity stealing methods from the electricity stealing bill text of the user;
and obtaining a typical electricity stealing user by utilizing a neighbor propagation clustering method according to the electricity stealing characteristic analysis result and related electricity stealing methods.
A typical electricity user electricity time sequence includes an electricity time sequence of normal electricity time and an electricity time sequence of electricity theft time.
And calculating the similarity A of the electricity utilization time sequence of the user to be diagnosed and the electricity utilization time sequence of the typical electricity stealing user by adopting a dynamic time bending method according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user.
And in response to the similarity A meeting the preset condition C, calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed by using a cosine function.
A power stealing user judging system based on a power stealing feature comprises,
the user computing module to be diagnosed: acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed;
typical electricity theft user computing modules: acquiring a power utilization time sequence of a typical power stealing user in historical data, and calculating a power stealing index of the typical power stealing user;
a first similarity module: calculating the similarity A of the user electricity utilization time sequence to be diagnosed and the typical electricity utilization time sequence of the user electricity stealing according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user;
a second similarity module: responding to the similarity A meeting a preset condition C, and calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed;
and a result judging module: and responding to the similarity B meeting a preset condition D, and judging that the user to be diagnosed is a suspected user.
A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform a power theft user determination method based on a power theft feature.
A computing device comprising one or more processors, one or more memories, and one or more programs, wherein one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing a power theft user determination method based on a power theft feature.
The invention has the beneficial effects that: 1. according to the invention, the electricity utilization time sequence of the user to be diagnosed and the electricity utilization time sequence of the typical electricity stealing user in the historical data are obtained, the similarity calculation is carried out according to the electricity stealing index, the user to be diagnosed is initially judged, and the similarity calculation of the latest electricity utilization time sequence and the initial electricity utilization time sequence is carried out on the user which is initially judged to be the suspected electricity stealing user, so that the electricity stealing judgment by utilizing the historical data is realized; 2. the invention utilizes the combination mode of the initial power utilization time sequence and the latest power utilization time sequence, can effectively eliminate the normal power utilization fluctuation caused by the production and life changes of users in the aspect of power utilization diagnosis, and is more accurate for positioning the power utilization users; 3. according to the invention, the electricity stealing indexes of different types of users are constructed, the original electricity consumption data are converted into the data representing the electricity stealing behavior of the users, the quality of the data is effectively improved, the data dimension is reduced, and a new direction and a new solution are provided for the identification of the electricity stealing users; 4. compared with the traditional method for judging the electricity stealing users, the method can judge whether the users steal electricity or not, and meanwhile, the electricity stealing means adopted by the users can be clear, so that the method has great significance for on-site electricity stealing check.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for more clearly illustrating the technical aspects of the present invention, and are not intended to limit the scope of the present invention.
As shown in fig. 1, a method for determining a user to steal electricity based on a characteristic of stealing electricity includes the following steps:
step 1, acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed; and acquiring a power utilization time sequence of the typical power stealing user in the historical data, and calculating the power stealing index of the typical power stealing user.
The user types include large-scale private transformer users, medium-and small-scale private transformer users, three-phase general business users, single-phase general business users and residential users.
The large-sized private transformer users and the medium-sized private transformer users are high-voltage users, the large-sized private transformer users (A class) refer to users with the electricity consumption of 100kVA or more, and the medium-sized private transformer users (B class) refer to users with the electricity consumption of 100kVA or less; the private transformer of the private transformer user refers to a transformer special for a certain customer, and the property rights of the private transformer are owned by the customer, and the private transformer user refers to a power user provided with the special transformer relative to the public transformer.
The three-phase general business users, the single-phase general business users and the resident users are low-voltage users, the three-phase general business users (C type) refer to non-resident three-phase power users with commercial, small power, office and other electric properties, and the single-phase general business users (D type) refer to non-resident single-phase power users with low-voltage 220V commercial, small power, office and other electric properties.
The user power utilization time sequence to be diagnosed comprises an initial period of power utilization time sequence and a last period of power utilization time sequence; typically a user metering point commissioning begins an initial continuous 6 month time series of electricity usage (daily), and a last continuous 3 month time series of electricity usage (daily).
Suppose the user U installs the meter and connects the electricity date to be D 0 At month of M 0 The method comprises the steps of carrying out a first treatment on the surface of the The current statistical analysis date is D c At month of M c Then, as shown in table 1, the user U power time sequence is constructed in 4 ways:
when D is 0 Before month 20 (containing), the value M is taken 0 ～(M 0 +5) 6 consecutive natural months, extending smoothly across the year;
when D is 0 After month 21 (containing), the value (M) 0 +1)～(M 0 +6) for 6 consecutive natural months, extending smoothly across the year.
When D is c Before month 10 (containing), the value (M) c -3)～(M c -1) 3 consecutive natural months;
when D is c After month 11 (containing), the value (M) c -2)～(M c ) Continuously for 3 natural months;
TABLE 1 user U Power time series construction
The electricity stealing indexes of different user categories are different, wherein the electricity stealing indexes of the large-scale private transformer users and the medium-scale private transformer users are consistent, and the electricity stealing indexes comprise electric quantity, current unbalance rate, voltage unbalance rate and active power; the three-phase general business users, the single-phase general business users and the resident users have consistent electricity stealing indexes, and the electricity is used as electric quantity.
Current unbalance rate= (maximum phase current-minimum phase current)/maximum phase current, which reflects current undercurrent phenomenon, namely difference among three-phase currents, wherein the three-phase currents are kept stable when normal power is used, and the higher the current unbalance rate is, the higher the current abnormality probability is represented; the voltage unbalance rate= (maximum phase voltage-minimum phase voltage)/maximum phase voltage reflects the voltage undervoltage phenomenon, namely the difference between three-phase voltages, the three-phase voltages are kept stable during normal power utilization, and the higher the voltage unbalance rate is, the higher the probability of representing voltage abnormality is, and the specific is shown in table 2.
TABLE 2 electric larceny index
Typical electricity stealing users are obtained from historical data, and particularly are obtained from an electricity stealing checking work order by adopting a neighbor propagation clustering method, and the method comprises the following steps of:
1) According to the electricity stealing index, analyzing the electricity stealing characteristics of different electricity stealing modes of the user, and analyzing related electricity stealing methods from the electricity stealing bill text of the user;
2) And obtaining a typical electricity stealing user by utilizing a neighbor propagation clustering method according to the electricity stealing characteristic analysis result and related electricity stealing methods.
Meanwhile, the typical electricity stealing user selection needs to meet the following requirements: 1) The user electricity time sequence data is complete; 2) The user steals electricity and checks the time definitely; 3) The time period of electricity stealing of the user is clear, and the data characteristics are consistent; 4) The electricity stealing and checking time is within 5 years, and is in the current electricity utilization state; 5) The electricity stealing means adopted by the user when electricity is stolen is clear.
The typical electricity stealing user electricity time sequence comprises an electricity time sequence of normal electricity time and an electricity time sequence of electricity stealing time; the electricity stealing period electricity utilization data of 3 months is selected based on the electricity stealing discovery and check time, and the normal electricity utilization data of 3 months is selected after the electricity stealing check or the normal electricity utilization data of 3 months is selected before the electricity stealing starts.
Let the user steal electricity U-shaped meter and connect the electricity date as D 0 At month of M 0 The check date is D E At month of M E The starting time of electricity larceny is D B At month of M B The time series data of the user U are constructed in 4 ways:
mode one: when M B -M 0 >=3 and D E -D B >At=3 months, take M B -3～M B +2 continuous 6 natural month data, extending smoothly across years;
mode two: when M B -M 0 >=3 and D E -D B <At 3 months, M was taken B -3～D E Continuous natural month data, and year-across and year-forward;
mode three: when M B -M 0 <3 and D E -D B >At=3 months, take M E -3～M E +2 continuous 6 natural month data, extending smoothly across years;
mode four: when M B -M 0 <3 and D E -D B <Taking D3 months later B ～M E +2 consecutive several natural months of data, extending smoothly across the year.
The calculation of the electricity stealing index is similar to that of the user to be diagnosed.
And 2, calculating the similarity A of the power utilization time sequence of the user to be diagnosed and the power utilization time sequence of the typical power stealing user by adopting a dynamic time warping method (Dynamic Time Wrapping, DTW) according to the power stealing index of the user to be diagnosed and the power stealing index of the typical power stealing user.
Firstly, calculating a daily electric quantity time sequence DTW distance between a typical electricity stealing user and a user to be diagnosed, and if the total active power DTW distance between the typical electricity stealing user and the user to be diagnosed is a special transformer user, calculating a three-phase voltage unbalance time sequence DTW distance between the typical electricity stealing user and the user to be diagnosed, and calculating a three-phase current unbalance time sequence DTW distance between the typical electricity stealing user and the user to be diagnosed; the total distance of the users to be diagnosed, i.e. the similarity a, is then calculated.
For a special variant user, the total distance S is calculated as follows:
S special transformer =0.5×daily electrical quantity time series DTW distance+0.3×active power time series DTW distance+0.1×three-phase voltage unbalance time series DTW distance+0.1×three-phase current unbalance time series DTW distance.
For non-private users, the total distance S is calculated as:
S non-proprietary transformer Time series of daily electrical quantity DTW distance.
When the users of the same class have a plurality of typical electricity stealing users, calculating the total distance S between the user to be diagnosed and each typical electricity stealing user i one by one i Selecting S with minimum distance i * As a value of the distance of the user to be diagnosed.
Step 3, if the similarity A is smaller than or equal to a threshold K, primarily judging that the user to be diagnosed is a suspected electricity larceny user; otherwise, ending.
For all the checked electricity stealing users, the ith group G is set after the users are grouped according to the user category i The number of the fraudulent use of electricity is C i . If C i >1000, then G i0 = { slave G i 1/10 proportion electricity stealing users are selected randomly; if 500 degrees f<C i <=1000, then G i0 = { slave G i 1/5 proportion electricity stealing users are selected randomly; if 100 is<C i <=500, then G i0 = { slave G i 1/2 proportion electricity stealing users are selected randomly; if C i <=100, then G i0 ＝G i . For G i0 All users stealing electricity calculate the distance S between every two users (k, j) according to the DTW method kj ，K＝∑S kj /C i1 ，k、j∈G i0 ，C i1 Is G i0 Number of users.
Large-scale private transformer users and medium-and small-scale private transformer users: and when the similarity between the user daily electric quantity time sequence to be diagnosed and the typical electricity stealing user daily electric quantity time sequence is smaller than or equal to a threshold value, and the similarity between the user power curve time sequence to be diagnosed and the typical electricity stealing user power curve time sequence is smaller than or equal to the threshold value, preliminarily judging that the user to be diagnosed is an electricity stealing suspected user.
Three-phase general business users, single-phase general business users and residential users: when the similarity between the user daily electric quantity time sequence to be diagnosed and the typical electricity stealing user daily electric quantity time sequence is smaller than or equal to a threshold value, the user to be diagnosed is primarily judged to be the electricity stealing suspected user.
Step 4, calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed by using a cosine function, and judging the user to be diagnosed as a suspected electricity stealing user if the similarity B is smaller than or equal to a threshold k; wherein k takes 80% value according to the correlation experience value.
The similarity B of the last 1 month electricity time series (judged according to the current statistical date) and the initial 6 months (consistent with step 1) electricity time series (electricity daily time series) of the user to be diagnosed is calculated using the cosine function.
The last 1 month is denoted as P c Current statistics dayDenoted as D c At month of M c Record P c Last 1 day is D E Day, P c ＝[1,D E ]。
P c The value is as follows:
when D is c <When=10, P c ＝M c -1, the month before the month of the statistical day, D E ＝(M c -1) last 1 days of the month;
when D is c >At number 10, P c ＝M c I.e. counting the month D of day E ＝D c -1。
Calculating initial power consumption of a user for 6 months and the latest 1 month P c Daily electricity amount time sequence similarity; let M k E month of { initial 6 months }, EDay (M k ) Represents M k The last 1 day of the month.
Firstly, calculating the similarity: m is M k Month [1, min (D) E ,EDay(M k ))]Vector a and P composed of daily electric quantity c [1, min (D) E ,EDay(M k ))]Calculating cosine included angle values of vectors a and b according to a vector b formed by daily electric quantity;
if the user is a special user, the following distances are continuously calculated:
calculating P of the user in the initial 6 months and the last 1 month c Total active power time series similarity;
calculating the similarity of the three-phase voltage unbalance time sequence of the user in the initial 6 months and the last 45 days;
and calculating the three-phase current imbalance time sequence similarity of the user in the initial 6 months and the last 45 days.
And calculating the total time sequence similarity of the user.
For a special variant user, the total similarity S is calculated as follows:
S special transformer =0.5×Σ daily electrical quantity time series similarity/total month number) +0.3×Σ active power time series similarity/total month number) +0.1×Σ three-phase voltage unbalance time series similarity/total month number) +0.1×Σ three-phase current unbalance time series similarity/total month number.
For non-private users, the total similarity S is calculated as:
S non-proprietary transformer The daily electrical time series similarity/total number of months.
According to the method, the electricity utilization time sequence of the user to be diagnosed and the electricity utilization time sequence of the typical electricity stealing user in the historical data are obtained, the similarity calculation is carried out according to the electricity stealing index, the user to be diagnosed is initially judged, and the similarity calculation of the latest electricity utilization time sequence and the initial electricity utilization time sequence is carried out on the user which is initially judged to be the suspected electricity stealing user, so that the electricity stealing judgment by utilizing the historical data is realized.
The method utilizes the combination mode of the initial electricity utilization time sequence and the latest electricity utilization time sequence, can effectively eliminate the normal electricity utilization fluctuation caused by the production and life changes of users in the aspect of electricity stealing diagnosis, and is more accurate in positioning electricity stealing users.
The method constructs the electricity stealing indexes of different types of users, converts the original electricity consumption data into the data representing the electricity stealing behavior of the users, effectively improves the quality of the data, reduces the dimension of the data, and provides a new direction and solution for the identification of the electricity stealing users.
Compared with the traditional electricity stealing user judging method, the method adopts the dynamic time bending method to calculate the similarity, can determine the electricity stealing means adopted by the user while judging whether the user steals electricity, and has great significance for on-site electricity stealing check.
A power stealing user judging system based on a power stealing feature comprises,
the user computing module to be diagnosed: acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed;
typical electricity theft user computing modules: acquiring a power utilization time sequence of a typical power stealing user in historical data, and calculating a power stealing index of the typical power stealing user;
a first similarity module: calculating the similarity A of the user electricity utilization time sequence to be diagnosed and the typical electricity utilization time sequence of the user electricity stealing according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user;
a second similarity module: responding to the similarity A meeting a preset condition C, and calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed;
and a result judging module: and responding to the similarity B meeting a preset condition D, and judging that the user to be diagnosed is a suspected user.
A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform a power theft user determination method based on a power theft feature.
A computing device comprising one or more processors, one or more memories, and one or more programs, wherein one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing a power theft user determination method based on a power theft feature.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The foregoing is illustrative of the present invention and is not to be construed as limiting thereof, but rather as providing for the use of additional embodiments and advantages of all such modifications, equivalents, improvements and similar to the present invention are intended to be included within the scope of the present invention as defined by the appended claims.
Claims (7)
1. A method for judging electricity stealing users based on electricity stealing characteristics is characterized in that: comprising the steps of (a) a step of,
acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed;
acquiring a power utilization time sequence of a typical power stealing user in historical data, and calculating a power stealing index of the typical power stealing user; the typical electricity stealing users are obtained by adopting a neighbor propagation clustering method, and the specific process is as follows: according to the electricity stealing index, the electricity stealing characteristics of different electricity stealing modes of the user are analyzed, related electricity stealing methods are analyzed from the electricity stealing bill text of the user, and a typical electricity stealing user is obtained by utilizing a neighbor propagation clustering method according to the analysis result of the electricity stealing characteristics and the related electricity stealing methods;
according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user, calculating the similarity A of the electricity utilization time sequence of the user to be diagnosed and the electricity utilization time sequence of the typical electricity stealing user by adopting a dynamic time bending method;
responding to the similarity A meeting a preset condition C, and calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed by using a cosine function;
and responding to the similarity B meeting a preset condition D, and judging that the user to be diagnosed is a suspected user.
2. The electricity theft user judging method based on the electricity theft feature according to claim 1, wherein: the user power time sequence to be diagnosed comprises an initial period of power time sequence and a last period of power time sequence.
3. The electricity theft user judging method based on the electricity theft feature according to claim 1, wherein: the user types comprise large-scale private transformer users, medium-and small-scale private transformer users, three-phase general business users, single-phase general business users and resident users; the large-scale private transformer users and the medium-and small-scale private transformer users have consistent electricity stealing indexes, including electric quantity, current unbalance rate, voltage unbalance rate and active power; the three-phase general business users, the single-phase general business users and the resident users have consistent electricity stealing indexes, and the electricity is used as electric quantity.
4. The electricity theft user judging method based on the electricity theft feature according to claim 1, wherein: a typical electricity user electricity time sequence includes an electricity time sequence of normal electricity time and an electricity time sequence of electricity theft time.
5. A steal electricity user judging system based on steal electricity characteristic is characterized in that: comprising the steps of (a) a step of,
the user computing module to be diagnosed: acquiring a power utilization time sequence of a user to be diagnosed, and calculating a power stealing index of the user to be diagnosed;
typical electricity theft user computing modules: acquiring a power utilization time sequence of a typical power stealing user in historical data, and calculating a power stealing index of the typical power stealing user; the typical electricity stealing users are obtained by adopting a neighbor propagation clustering method, and the specific process is as follows: according to the electricity stealing index, the electricity stealing characteristics of different electricity stealing modes of the user are analyzed, related electricity stealing methods are analyzed from the electricity stealing bill text of the user, and a typical electricity stealing user is obtained by utilizing a neighbor propagation clustering method according to the analysis result of the electricity stealing characteristics and the related electricity stealing methods;
a first similarity module: according to the electricity stealing index of the user to be diagnosed and the electricity stealing index of the typical electricity stealing user, calculating the similarity A of the electricity utilization time sequence of the user to be diagnosed and the electricity utilization time sequence of the typical electricity stealing user by adopting a dynamic time bending method;
a second similarity module: responding to the similarity A meeting a preset condition C, and calculating the similarity B of the latest power utilization time sequence and the initial power utilization time sequence of the user to be diagnosed by using a cosine function;
and a result judging module: and responding to the similarity B meeting a preset condition D, and judging that the user to be diagnosed is a suspected user.
6. A computer readable storage medium storing one or more programs, characterized by: the one or more programs include instructions, which when executed by a computing device, cause the computing device to perform any of the methods of claims 1-4.
7. A computing device, characterized by: comprising the steps of (a) a step of,
one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, the one or more programs comprising instructions for performing any of the methods of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011464556.8A CN112649641B (en) | 2020-12-14 | 2020-12-14 | Electricity stealing user judging method based on electricity stealing characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011464556.8A CN112649641B (en) | 2020-12-14 | 2020-12-14 | Electricity stealing user judging method based on electricity stealing characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112649641A CN112649641A (en) | 2021-04-13 |
| CN112649641B true CN112649641B (en) | 2023-05-02 |
Family
ID=75353874
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011464556.8A Active CN112649641B (en) | 2020-12-14 | 2020-12-14 | Electricity stealing user judging method based on electricity stealing characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112649641B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116862116B (en) * | 2023-09-05 | 2024-07-30 | 国网天津市电力公司营销服务中心 | Intelligent early warning method and system for preventing electricity larceny, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105203924A (en) * | 2015-10-10 | 2015-12-30 | 上海博英信息科技有限公司 | Electricity usage trend abnormity suspicion analysis method and anti-electric-larceny monitoring system |
| CN107328974A (en) * | 2017-08-03 | 2017-11-07 | 北京中电普华信息技术有限公司 | A kind of stealing recognition methods and device |
| CN109146705A (en) * | 2018-07-02 | 2019-01-04 | 昆明理工大学 | A kind of method of electricity consumption characteristic index dimensionality reduction and the progress stealing detection of extreme learning machine algorithm |
| CN109583679A (en) * | 2018-09-30 | 2019-04-05 | 国网浙江长兴县供电有限公司 | A kind of stealing Suspected Degree analysis method of more algorithm fusions |
| CN109784388A (en) * | 2018-12-29 | 2019-05-21 | 北京中电普华信息技术有限公司 | Stealing user identification method and device |
| AU2019101183A4 (en) * | 2019-10-02 | 2020-01-16 | Han, Yining MISS | Feature Extraction and Fusion for Industrial Data |
-
2020
- 2020-12-14 CN CN202011464556.8A patent/CN112649641B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105203924A (en) * | 2015-10-10 | 2015-12-30 | 上海博英信息科技有限公司 | Electricity usage trend abnormity suspicion analysis method and anti-electric-larceny monitoring system |
| CN107328974A (en) * | 2017-08-03 | 2017-11-07 | 北京中电普华信息技术有限公司 | A kind of stealing recognition methods and device |
| CN109146705A (en) * | 2018-07-02 | 2019-01-04 | 昆明理工大学 | A kind of method of electricity consumption characteristic index dimensionality reduction and the progress stealing detection of extreme learning machine algorithm |
| CN109583679A (en) * | 2018-09-30 | 2019-04-05 | 国网浙江长兴县供电有限公司 | A kind of stealing Suspected Degree analysis method of more algorithm fusions |
| CN109784388A (en) * | 2018-12-29 | 2019-05-21 | 北京中电普华信息技术有限公司 | Stealing user identification method and device |
| AU2019101183A4 (en) * | 2019-10-02 | 2020-01-16 | Han, Yining MISS | Feature Extraction and Fusion for Industrial Data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112649641A (en) | 2021-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110231528B (en) | Transformer household variation common knowledge identification method and device based on load characteristic model library | |
| WO2020252785A1 (en) | Abnormal electricity use recognition method and device, and computer readable storage medium | |
| CN110288039B (en) | Electricity stealing detection method based on user electricity load characteristics | |
| CN106022951A (en) | Electricity consumption abnormity analysis method and apparatus | |
| CN112836738B (en) | BP neural network-based electricity stealing behavior detection method | |
| CN110503136A (en) | Platform area line loss exception analysis method, computer readable storage medium and terminal device | |
| CN110749784B (en) | Line electricity stealing detection method based on electric power data wavelet analysis | |
| CN112649641B (en) | Electricity stealing user judging method based on electricity stealing characteristics | |
| CN110244099A (en) | Stealing detection method based on user's voltage | |
| CN111652761B (en) | Multi-source feature fusion electricity stealing behavior detection method based on evidence theory | |
| CN110390440B (en) | Clustering and deep neural network-based intelligent ammeter user aggregate load prediction method | |
| CN106685086B (en) | Remote power utilization management system | |
| CN115166625A (en) | Intelligent ammeter error estimation method and device | |
| Yan et al. | Comparative study of electricity-theft detection based on gradient boosting machine | |
| Xu et al. | A combined unsupervised learning approach for electricity theft detection and loss estimation | |
| Fu et al. | A developmental state: How to allocate electricity efficiently in a developing country | |
| Farrugia | Imputation and anomaly detection for consumer load profiles | |
| Guo et al. | Research on recognition and classification of user stealing detection based on weighted naive Bayes | |
| CN115859331B (en) | Smart city information security guarantee system | |
| CN115826909B (en) | Big data analysis-based electricity larceny detection system | |
| CN116910596B (en) | User electricity stealing analysis method, device and storage medium based on improved DBSCAN clustering | |
| Mbunwe | Analysis of energy crisis and how it affects production sector and economic growth of Nigeria | |
| CN115168780A (en) | Voltage sag homologous identification method and device | |
| Oluyomi | Detecting False Data Injection in a Large-Scale Water Distribution Network | |
| CN109615555B (en) | User abnormal electricity behavior identification method and system based on integrated learning algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |