CN113128596A - Electricity stealing detection method and device and computer readable storage medium - Google Patents
Electricity stealing detection method and device and computer readable storage medium Download PDFInfo
- Publication number
- CN113128596A CN113128596A CN202110430297.5A CN202110430297A CN113128596A CN 113128596 A CN113128596 A CN 113128596A CN 202110430297 A CN202110430297 A CN 202110430297A CN 113128596 A CN113128596 A CN 113128596A
- Authority
- CN
- China
- Prior art keywords
- user
- electricity
- electricity stealing
- stealing
- day
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 271
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 239000013598 vector Substances 0.000 claims abstract description 41
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000004927 fusion Effects 0.000 claims description 10
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000006399 behavior Effects 0.000 abstract description 21
- 230000002354 daily effect Effects 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000005612 types of electricity Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 101100161752 Mus musculus Acot11 gene Proteins 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Power Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses an electricity stealing detection method, which comprises the following steps: determining the total error of electricity stealing of the transformer area within the specified number of days; for each user in the distribution area, determining the correlation between the total power consumption vector corresponding to the user and the total power stealing error as a first power stealing possibility corresponding to the user, wherein the total power consumption vector is obtained by combining the power consumption data of the user on each day in the specified number of days; for each user in the distribution area, determining a second electricity stealing possibility corresponding to the user according to the difference degree between the electricity utilization curves of the user on each day; determining electricity stealing users in the distribution area according to the first electricity stealing possibility and the second electricity stealing possibility of each user. The method disclosed by the application can detect electricity stealing behaviors of various different types, has excellent performance, and greatly improves the accuracy and adaptability of electricity stealing detection.
Description
Technical Field
The present application relates to the field of electricity stealing detection technologies, and in particular, to an electricity stealing detection method, an electricity stealing detection device, and a computer-readable storage medium.
Background
The electricity stealing is a behavior that the electricity is not metered or is less metered by adopting an illegal means to occupy the electric energy so as to achieve the purpose of paying less or not paying the electricity, the electricity stealing type mainly comprises the steps of changing the current (losing current), changing the voltage (losing voltage), changing the structure and the wiring mode of a meter, stealing the electricity by a strong alternating current magnetic field and the like, and the common characteristic of the electricity stealing is that the actual value of the electricity consumption is changed so as to achieve the purpose of metering less or not paying the electricity. With the increase of electricity stealing behaviors through a mode of changing electricity meter data, the electricity stealing behaviors become more diversified and difficult to detect, so how to accurately detect the electricity stealing behaviors of users is a technical problem to be solved urgently at present.
Disclosure of Invention
The application provides a method and a device for detecting electricity stealing and a computer readable storage medium, which can detect electricity stealing behaviors of various different types, have excellent performance and greatly improve the accuracy and the adaptability of electricity stealing detection.
In view of the above, a first aspect of the present application provides a method for detecting electricity theft, including:
determining the total error of electricity stealing of the transformer area within the specified number of days;
for each user in the distribution area, determining the correlation between the total power consumption vector corresponding to the user and the total power stealing error as a first power stealing possibility corresponding to the user, wherein the total power consumption vector is obtained by combining the power consumption data of the user on each day in the specified number of days;
for each user in the distribution area, determining a second electricity stealing possibility corresponding to the user according to the difference degree between the electricity utilization curves of the user on each day;
determining electricity stealing users in the distribution area according to the first electricity stealing possibility and the second electricity stealing possibility of each user.
Optionally, the correlation between the total power consumption vector corresponding to the user and the total power stealing error is calculated by a maximum mutual information coefficient algorithm.
Optionally, the determining, according to the difference between the power utilization curves of the user in each day, a second power stealing possibility corresponding to the user includes:
clustering the power utilization curves of the users on each day to obtain the difference between the power utilization curve of each day of the users and the power utilization curves of other days except the day;
and fusing the difference degrees corresponding to the electricity utilization curves of the user on each day to obtain a second electricity stealing possibility corresponding to the user.
Optionally, the clustering the power utilization curves of the users on each day includes:
and clustering the electricity utilization curves of the users in each day through a fast clustering algorithm based on density peak values.
Optionally, the clustering the power consumption curves of the users in each day to obtain the difference between the power consumption curve of the user in each day and the power consumption curves of other days except the day includes:
clustering the power utilization curves of the users on each day to obtain the power utilization curve density and the power utilization curve distance corresponding to each day of the users;
and fusing the power utilization curve density and the power utilization curve distance of each day to obtain the difference degree corresponding to the power utilization curve of each day.
Optionally, the determining power stealing users in the distribution area according to the first power stealing possibility and the second power stealing possibility of each user includes:
carrying out weighted fusion on the first electricity stealing possibility and the second electricity stealing possibility of each user to obtain fused electricity stealing possibility corresponding to each user;
and sequencing the fusion electricity stealing possibility corresponding to each user, and determining electricity stealing users in the distribution room according to a sequencing result.
Optionally, the weight corresponding to the first power stealing possibility and the weight corresponding to the second power stealing possibility are adaptively changed.
Optionally, the determining the total error of power stealing in the distribution area within the specified number of days includes:
acquiring a measured value of the total power consumption of a distribution area in a specified number of days and a measured value of the power consumption corresponding to each user in the distribution area;
and calculating the total electricity stealing error corresponding to the distribution room according to the sum of the measured value corresponding to the total electricity consumption and the measured value corresponding to the user.
This application second aspect provides an electricity larceny detection device, includes:
the total error determining module is used for determining the total error of electricity stealing in the distribution area within the specified days;
a first electricity stealing possibility determining module, configured to determine, for each user in the distribution room, a correlation between a total electricity consumption vector corresponding to the user and the total electricity stealing error as a first electricity stealing possibility corresponding to the user, where the total electricity consumption vector is obtained by combining electricity consumption data of the user on each day within the specified number of days;
the second electricity stealing possibility determining module is used for determining a second electricity stealing possibility corresponding to each user in the distribution room according to the difference degree between the electricity utilization curves of the users on each day;
and the electricity stealing user determining module is used for determining electricity stealing users in the transformer area according to the first electricity stealing possibility and the second electricity stealing possibility of each user.
A third aspect of the present application provides a computer-readable storage medium for storing program code for executing the electricity larceny detection method of the first aspect described above.
According to the technical scheme, the method has the following advantages:
the embodiment of the application provides an electricity stealing detection method, which can determine a first electricity stealing possibility corresponding to a user through the correlation degree of a total electricity utilization vector corresponding to the user and a total electricity stealing error, and determine a second electricity stealing possibility corresponding to the user according to the difference degree between electricity utilization curves of the user in each day, so that electricity stealing users in a distribution room can be determined according to the first electricity stealing possibility and the second electricity stealing possibility of each user. The electricity stealing detection method provided by the embodiment of the application can detect electricity stealing behaviors of various different types, has excellent performance, and greatly improves the accuracy and adaptability of electricity stealing detection.
Drawings
FIG. 1 is a first flowchart of a power theft detection method provided by an embodiment of the present application;
FIG. 2 is a second flowchart of a power theft detection method provided by an embodiment of the present application;
FIG. 3 is a graph of six types of electricity stealing provided by an embodiment of the present application;
FIG. 4 is a graph of the number of electricity stealing users versus AUC provided by an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electricity stealing detection device provided by an embodiment of the present application.
Detailed Description
In order to make the technical solutions of the present application better understood, the technical solutions of the present application will be clearly and completely described below with reference to the drawings in the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The electricity stealing is a behavior that the electricity is not measured or is less measured by adopting an illegal means to occupy the electric energy, so as to achieve the purpose of paying less or even not paying the electricity fee. Accurate detection of electricity stealing behavior of a user is an important task for maintaining the safety and the benefit of a power grid.
In order to detect the electricity stealing behavior of a user and screen the user with the electricity stealing behavior, an embodiment of the application provides an electricity stealing detection method, which may be referred to in fig. 1, where fig. 1 is a first flowchart of the electricity stealing detection method provided in the embodiment of the application, and the method includes:
s101, determining the total error of electricity stealing of the transformer area within the specified number of days.
The transformer area refers to a power supply range or area of a transformer, and the transformer area comprises a plurality of power users. The total electricity stealing error is a measurement error of the total electricity consumption caused by electricity stealing behaviors of users, specifically, the sum of the measured total electricity consumption and the electricity consumption corresponding to each user in the distribution area is unequal. In an embodiment, the measured value of the total power consumption of the distribution area in the specified days can be obtained, that is, the power consumption data measured by the distribution area total table in the specified days can be obtained, and the measured value of the power consumption corresponding to each user in the distribution area in the specified days can be obtained, that is, the power consumption data measured by each user sub-table in the specified days can be obtained, so that the sum of the measured value corresponding to the total power consumption and the measured value of the power consumption corresponding to each user is subtracted, and the total error of electricity stealing corresponding to the distribution area can be calculated.
It is understood that the number of the specified days can be set according to the requirement, for example, the number of the specified days can be 30 days, 300 days, and the like, and the embodiment of the present application is not limited thereto.
In one example, the electricity consumption data of n users in m days of a platform area can be collected, and the total electricity stealing error e of the platform area caused by electricity stealing behavior can be calculated by the following formulat：
Wherein A is the set of n electricity users in the platform area, EtIs a measured value of a table of the station area,
an electricity meter measurement representing the electricity usage of the ith subscriber in the area. And because the total error e of non-electricity-stealing users on electricity stealingtHas no influence, so etAnd can be represented as:
wherein, gamma is the set of users stealing electricity in the platform area, xi,tThe true value of the electricity consumption of the ith user.
After the power consumption data corresponding to each user in the station area is obtained, in an embodiment, the power consumption data of the user may be normalized, and specifically, the power consumption data of each user may be divided by the maximum value of the power consumption in the time series. In the above example, the normalized electricity consumption data of each user may be represented by n × m vectors, that is, each user of the n users corresponds to m vectors, each vector corresponds to electricity consumption data of the user for 1 day, and m vectors corresponds to electricity consumption data of the user for m days.
S102, for each user in the distribution area, determining the correlation between the total power consumption vector corresponding to the user and the total power stealing error as a first power stealing possibility corresponding to the user.
Here, each user may determine a corresponding total power consumption vector, where the total power consumption vector of one user is obtained by arranging and combining power consumption data of the user in each day of a specified number of days according to a chronological order, for example, for an ith user, the power consumption data of m days corresponding to the ith user may correspond to m vectors, and the m vectors are arranged and combined according to the chronological order to obtain the total power consumption vector corresponding to the ith user, and the total power consumption vector corresponding to the ith user may be recorded as ci(i＝1,2,...,n)。
In S102, a correlation between the total power consumption vector and the total power stealing error of each user may be calculated separately for each user, and the correlation corresponding to each user may be used as the first power stealing possibility corresponding to the user. It can be understood that the higher the correlation degree corresponding to a user is, the higher the probability that the user has electricity stealing behavior, i.e. the higher the corresponding first electricity stealing possibility.
There are many calculation methods for the correlation between the total power consumption vector and the total error of power stealing. In one embodiment, the correlation between the total power consumption vector and the total error of power stealing can be calculated by a maximum mutual information coefficient algorithm. Specifically, the total power consumption vector c corresponding to each useriTotal error from stealing electricity etThe correlation degree of (c) can be calculated by the following steps:
(1) given a finite set D of ordered pairs comprising (x, y), where x and y each represent a total electricity vector c for a single useriError from total electricity stealing et。
(2) X is divided into m parts and y is divided into n parts. Then the two-dimensional plane xOy is divided into a grid of m × n, this grid is denoted as G, and the distribution of the elements in the set D in each cell in the grid is denoted as D ∞G. For the
As defined below:
where max represents the maximum value among all cells in grid G, I (DG) Represents D-GThe MI value of (1).
(3) The definition of the feature matrix is:
when the set D is determined, the (x, y) ordered pairs have | D | pairs, and the number of grids is less than B (n), then the estimation value is determined as follows:
determining whether B (| D |) ═ D | is Y-Y is selected empirically0.6. The maximum mutual information coefficient falls within a range of 0,1]The larger the value, the stronger the correlation, and the stronger the correlation, the greater the possibility of electricity theft. The maximum mutual information coefficient algorithm can represent the nonlinear fuzzy relation of two vectors, and has the characteristic that the higher the dimension of the input vector is, the more accurate the obtained estimated value is.
In an embodiment, the ranking may be performed according to the degree of correlation corresponding to each user, and the rank value with higher ranking is larger, to obtain a first ranking of probability of electricity stealing, which is denoted as rank 1.
S103, for each user in the distribution area, determining a second electricity stealing possibility corresponding to the user according to the difference degree between the electricity utilization curves of the user in each day.
As described above, the daily power consumption data of each user in a specified number of days may be obtained, where the daily power consumption data may include the power consumption of the user at different times of the day, and therefore, the daily power consumption curve of the user may be determined according to the daily power consumption data of the user.
For each user, the difference degree between the power utilization curves of the user on each day can be determined, so that the second power stealing possibility corresponding to the user can be determined according to the difference degree. Because the electricity-stealing user has a larger difference between the electricity utilization curve of the same day and the normal electricity utilization curve without the electricity-stealing behavior when implementing the electricity-stealing behavior, if the difference between the electricity utilization curves of each day of a user is larger, the possibility that the user is the electricity-stealing user is also larger.
In determining the difference between the power curves of the user on each day, in one embodiment, the power curves of the user on each day may be clustered, so that the difference between the power curve of the user on each day and the power curves of other days except the day may be obtained. Here, when determining the degree of difference corresponding to the power usage curve of the user for each day, specifically, the power usage curves of the user for each day may be clustered, and the power usage curve density corresponding to each day after the clustering of all the power usage curves of the user and the distance between the power usage curve for each day and the power usage curve for the other days other than the day may be obtained, and the distance may be referred to as a power usage curve distance. After the power consumption curve density and the power consumption curve distance corresponding to each day of the user are obtained, the power consumption curve density and the power consumption curve distance can be fused for each day, and therefore the difference degree corresponding to the power consumption curve for each day can be obtained. And then, the difference degrees corresponding to the electricity utilization curves of the user in each day can be fused, so that the second electricity stealing possibility corresponding to the user can be obtained.
In clustering the power usage curves of the user for each day, in one embodiment, the power usage curves of the user for each day may be clustered through a fast clustering algorithm based on density peaks. For the sake of understanding, the specific steps of the fast clustering algorithm based on density peaks will be described below with reference to the previous example of n users for m days.
The power consumption curve of the user i in each day of m days can be obtained, and the power consumption curve of the user j day can be marked as Ci,j. In the fast clustering algorithm of the density peak, there are two indexes for measuring the power utilization curve of the user: density of electricity consumption curve rhojDistance delta from power utilization curvej。ρjRepresenting the density, delta, corresponding to the power utilization curve of the j day after the clustering of all the power utilization curves of a certain userjRepresents the distance between the power curve of the j day of a certain user and the power curve of other days of the user. RhojAnd deltajIs calculated by the distance d between the data point and thej,l(1 ═ 1, 2., m, l ≠ j) is relevant, ρjCan be calculated by the following formula:
wherein d iscRepresenting the truncation distance, χ (·) represents the kernel function, which satisfies the following equation:
δjcan be calculated by the following formula:
that is, the distance between the power consumption curve with the density higher than that of the power consumption curve on the day j and the power consumption curve on the day j is the shortest, and particularly, when the density of the power consumption curve on a certain day is the highest, the distance delta is setjComprises the following steps:
the electricity utilization curve of the electricity stealing users is different from the electricity utilization curve of the normal users in characteristics, so after clustering, the electricity stealing users are clustered into two different categories. In practice, the number of electricity stealing users is often much less than that of normal users, so the density rho of electricity stealing users after electricity utilization curve clusteringjLower than normal users, and a distance δjTherefore, in an embodiment, the power consumption curve density and the power consumption curve distance on the jth day of the user i may be fused to obtain the difference degree corresponding to the power consumption curve on the jth day of the user i, where the difference degree may also be referred to as a power consumption curve evaluation index and may be recorded as ξi,jIt can be determined by the following formula:
the difference degree xi corresponding to the power utilization curve of the j day of the user ii,jThe larger the probability that there is a power theft action on user i day j. Further, for each user, the difference degrees corresponding to m days of the user may be fused, and a specific fusion manner may be to calculate an arithmetic average of the difference degrees corresponding to m days, and the arithmetic average obtained by fusion may be used as the second electricity stealing possibility corresponding to the user.
In an embodiment, the second electricity stealing probabilities corresponding to the users may be ranked according to sizes, and the rank2 is obtained.
Compared with clustering algorithms such as K-means clustering and the like, the fast clustering algorithm based on the density peak value does not need to consider the shape and select any parameter. Moreover, the fast clustering algorithm based on the density peak value is simple, and only needs to calculate rhojThen delta is obtainedjAnd xii,jWithout going through any iteration.
S104, determining electricity stealing users in the distribution room according to the first electricity stealing possibility and the second electricity stealing possibility of each user.
In an embodiment, the first power stealing possibility and the second power stealing possibility of each user may be weighted and fused to obtain a fused power stealing possibility corresponding to each user, and then a specified number of users with the highest fused power stealing possibility may be determined as power stealing users in the platform area. In an embodiment, the merged power stealing possibility corresponding to each user may be ranked, and a user with a higher ranking (for example, greater than a preset serial number) may be determined as a power stealing user in the platform area.
It is noted that in an embodiment, the weight corresponding to the first power stealing likelihood and the weight corresponding to the second power stealing likelihood may be adaptively changed when fusing the first power stealing likelihood and the second power stealing likelihood. Here, the above description may still be made in connection with the example of m-day n users.
After the first power stealing possibility ranking rank1 and the second power stealing possibility ranking rank2 are obtained, the two power stealing possibility rankings can be fused by using the weight which changes in a self-adaptive manner to obtain the comprehensive final power stealing possibility ranking:
wherein alpha is1,α2Always satisfy alpha1+α 21 and varies with rank1 and rank2, respectively. If the arithmetic mean (alpha) of two sorts is simply chosen1＝α20.5), when there is a high probability that the electricity stealing subscriber is in rank1, but the probability is low in rank2, the result of the comprehensive ranking is not ideal. The adaptively changing weight can ensure that the higher the ranking in the comprehensive ranking rank, the higher the possibility of electricity stealing of the users, i.e. the higher the weight occupied in the rank. In the algorithm of the technical scheme, alpha1,α2The determination method of (2) is as follows:
wherein, betapN is the number of users as a fraction of the likelihood that the users steal power in the rank. To ensure alpha1+α2×1，α1,α2The following are selected:
the comprehensive ranking rank is the final ranking, wherein the user with higher ranking can be determined as the electricity stealing user.
The embodiment of the application provides an electricity stealing detection method, which can determine a first electricity stealing possibility corresponding to a user through the correlation degree of a total electricity utilization vector corresponding to the user and a total electricity stealing error, and determine a second electricity stealing possibility corresponding to the user according to the difference degree between electricity utilization curves of the user in each day, so that electricity stealing users in a distribution room can be determined according to the first electricity stealing possibility and the second electricity stealing possibility of each user. The electricity stealing detection method provided by the embodiment of the application can detect electricity stealing behaviors of various different types, has excellent performance, and greatly improves the accuracy and adaptability of electricity stealing detection.
Referring to fig. 2, fig. 2 is a second flowchart of a method for detecting electricity theft provided by an embodiment of the present application, where the method may include:
s201, obtaining a measured value of the total electricity consumption of the distribution area in a specified number of days and a measured value of the electricity consumption corresponding to each user in the distribution area.
S202, calculating the total electricity stealing error corresponding to the distribution room according to the sum of the measured value corresponding to the total electricity consumption and the measured value corresponding to the user.
And S203, for each user in the distribution area, combining the power utilization data of the user in each day in the specified number of days to obtain a total power utilization vector corresponding to the user.
S204, calculating the correlation between the total power consumption vector corresponding to the user and the total power stealing error according to a maximum mutual information coefficient algorithm, and determining the correlation as the first power stealing possibility corresponding to the user.
S205, clustering the power utilization curves of the users in each day through a density peak value-based rapid clustering algorithm for each user in the distribution area to obtain the power utilization curve density and the power utilization curve distance corresponding to each day of the users.
S206, the power utilization curve density and the power utilization curve distance of the user every day are fused to obtain the difference degree between the power utilization curve of the user every day and the power utilization curves of other days except the day.
And S207, fusing the difference degrees corresponding to the electricity utilization curves of the user on each day to obtain a second electricity stealing possibility corresponding to the user.
S208, carrying out weighted fusion on the first electricity stealing possibility and the second electricity stealing possibility of each user to obtain the fused electricity stealing possibility corresponding to each user.
S209, sorting the fusion electricity stealing possibility corresponding to each user, and determining electricity stealing users in the distribution room according to the sorting result.
The embodiment of the application provides an electricity stealing detection method, which can determine a first electricity stealing possibility corresponding to a user through the correlation degree of a total electricity utilization vector corresponding to the user and a total electricity stealing error, and determine a second electricity stealing possibility corresponding to the user according to the difference degree between electricity utilization curves of the user in each day, so that electricity stealing users in a distribution room can be determined according to the first electricity stealing possibility and the second electricity stealing possibility of each user. The electricity stealing detection method provided by the embodiment of the application can detect electricity stealing behaviors of various different types, has excellent performance, and greatly improves the accuracy and adaptability of electricity stealing detection.
An embodiment for performing electricity theft detection using the above method is provided below.
The electricity stealing patterns can be categorized into the six types shown in table 1:
TABLE 1 six types of electricity stealing
In Table 1, xtThe actual value of the used amount of electricity that changes with time for the user,
corresponding measured values for the user. The six electricity stealing types are six basic electricity stealing types, and other electricity stealing modes are combinations of the six basic electricity stealing types, so that experimental results carried out on the six electricity stealing types have general significance. Reference may be made to fig. 3, fig. 3 being a graph of six types of electricity stealing provided by embodiments of the present application.
The following feasibility verification is performed on the six electricity stealing modes by combining a specific example, a calculation formula and the electricity stealing detection algorithm based on the correlation and cluster fusion technology, which is mentioned above, and specifically includes:
in order to verify the effectiveness of the technical scheme, the method and the device are applied to a platform area containing more than four thousand residents, 485 middle and small-sized enterprises and other users, the inspected power utilization data exceeding 500 days are collected, and the power utilization data can be regarded as the actual power consumption of the users. Here, the electricity consumption data of the 485 small-sized enterprise in the district in 2009 in 8 months (except 31 days) can be selected, the obtained electricity consumption data can be regarded as 485 × 30 groups of vectors, and the electricity consumption of the district is measured once in half an hour, so that each group of vectors contains 48 data. 485 users are divided into 12 groups, about 40 users are provided in each group, 1 to 5 users are randomly selected from each group as electricity stealing users, the number of the electricity stealing users is sequentially increased, and the actual values of the electricity stealing users are randomly changed by the six different electricity stealing modes to be used as sub-meter measurement values. Experiments were performed to verify the performance of the algorithm following the steps of the electricity theft detection method provided previously.
In order to evaluate the performance of the technical solution proposed in the embodiments of the present application, the area under the working characteristic curve (AUC) of the subject can be used as an evaluation index. The test subject work characteristic curve is a curve obtained by drawing a graph with the true rate as a vertical axis and the false positive rate as a horizontal axis. All users are divided into a normal user set N and a power stealing user set gamma, and the | N | and the | gamma | respectively represent the number of users in the set. All users are ranked in order of possible power stealing from small to large. The AUC formula is as follows:
of the 12 groups of users, 5 were randomly selected as electricity stealing users. The electricity stealing users account for around 12.5% of the group of users. The electricity stealing detection method provided by the embodiment of the application and other electricity stealing detection methods are respectively subjected to 100 electricity stealing user detection experiments, the average value is obtained, and the result is shown in table 2:
TABLE 2 mean value of the Electricity stealing detection index (AUC)
As can be seen from the above table, after the maximum mutual information coefficient algorithm is combined with the fast clustering algorithm based on the density peak, the detection capability of the sixth electricity stealing type is greatly improved, and the detection capability of other electricity stealing types also has higher detection accuracy. The experiment proves the effectiveness of the algorithm provided by the embodiment of the application.
Experiment the electricity stealing type 6 is tested by adopting the electricity stealing detection method provided by the embodiment of the application and other electricity stealing detection methods, and can be seen in fig. 4, and fig. 4 is a change curve graph of the number of electricity stealing users and AUC provided by the embodiment of the application.
It can be seen from the curve that, as the number of electricity stealing users increases, the performance of other electricity stealing detection methods is obviously deteriorated, and the fast clustering algorithm based on the density peak value is less affected by the increase of the users, so that the stable detection accuracy can be maintained, and the performance is better in the station areas with a large number of electricity stealing users.
The embodiment of the application provides a fusion electricity stealing detection algorithm based on a maximum mutual information coefficient and density peak value fast clustering algorithm, has effectiveness on six basic electricity stealing types, is suitable for common electricity utilization scenes, and has strong applicability.
The maximum mutual information coefficient algorithm can obtain the relevance between the total electricity stealing error and the total electricity consumption of each user during the measurement period under the condition of the minimum obtained data information, overcomes the defects that the Pearson correlation coefficient algorithm can only judge the linear correlation degree of two vectors and cannot detect more complex two-time and three-time function relations, time-varying relations and the like, can detect more kinds of relevance, is not limited to linear relations and is applied to detecting various kinds of relevance.
The fast clustering algorithm of the density peak can detect abnormal users according to the density characteristics, the load curves of the abnormal users have randomness compared with the load curves of normal users, the algorithm can better overcome the randomness, the deficiency of the algorithm based on the correlation in the detection of partial electricity stealing types is compensated, and the algorithm has good effect in big data clustering and abnormal detection.
The technical scheme provided by the embodiment of the application adopts the fast clustering algorithm of the density peak value on the basis of detecting the correlation between the management line loss and the electricity stealing behavior of the user by the maximum mutual information coefficient algorithm, positions the user with electricity stealing possibility, can detect various electricity stealing types, combines the advantages of two algorithms, supplements the defect of respective single algorithm, and greatly improves the accuracy and the adaptability of detecting the electricity stealing behavior.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electricity stealing detection apparatus provided in an embodiment of the present application, where the apparatus includes:
a total electricity stealing error determining module 510, configured to determine a total electricity stealing error of the distribution room within a specified number of days;
a first electricity stealing possibility determining module 520, configured to determine, for each user in the distribution area, a correlation between a total electricity consumption vector corresponding to the user and the total electricity stealing error as a first electricity stealing possibility corresponding to the user, where the total electricity consumption vector is obtained by combining electricity consumption data of the user on each day within the specified number of days;
a second electricity stealing possibility determining module 530, configured to determine, for each user in the distribution room, a second electricity stealing possibility corresponding to the user according to a difference between electricity utilization curves of the user on each day;
a power stealing subscriber determining module 540, configured to determine power stealing subscribers in the distribution room according to the first power stealing probability and the second power stealing probability of each subscriber.
The embodiment of the application provides an electricity stealing detection device, can confirm the first electricity stealing possibility that the user corresponds through the relevance degree of the total power consumption vector that the user corresponds and the total error of stealing electricity, according to the difference degree between the power consumption curves of each day of user, confirm the second electricity stealing possibility that the user corresponds to can confirm the electricity stealing user in the platform district according to the first electricity stealing possibility and the second electricity stealing possibility of each user. The electricity stealing detection method provided by the embodiment of the application can detect electricity stealing behaviors of various different types, has excellent performance, and greatly improves the accuracy and adaptability of electricity stealing detection.
The present application further provides a computer-readable storage medium for storing program code for executing any one of the electricity stealing detection methods provided by the embodiments of the present application.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and modules may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of operation in sequences other than those illustrated or otherwise described herein. Moreover, the terms "comprises," "comprising," and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus.
It should be understood that in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and in actual implementation, there may be other divisions, for example, multiple modules or components may be combined or integrated into another system, or some features may be omitted, or not implemented. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or modules, and may be in an electrical, mechanical or other form.
The modules described as separate parts may or may not be physically separate, and parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
In addition, functional modules in the embodiments of the present application may be integrated into one processing module, or each of the modules may exist alone physically, or two or more modules are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode.
The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
Claims (10)
1. A method of detecting theft of electricity, comprising:
determining the total error of electricity stealing of the transformer area within the specified number of days;
for each user in the distribution area, determining the correlation between the total power consumption vector corresponding to the user and the total power stealing error as a first power stealing possibility corresponding to the user, wherein the total power consumption vector is obtained by combining the power consumption data of the user on each day in the specified number of days;
for each user in the distribution area, determining a second electricity stealing possibility corresponding to the user according to the difference degree between the electricity utilization curves of the user on each day;
determining electricity stealing users in the distribution area according to the first electricity stealing possibility and the second electricity stealing possibility of each user.
2. The method according to claim 1, wherein the correlation between the total power consumption vector corresponding to the user and the total error of power stealing is calculated by a maximum mutual information coefficient algorithm.
3. The method of claim 1, wherein determining a second power stealing probability corresponding to the user according to a difference between power usage curves of the user on each day comprises:
clustering the power utilization curves of the users on each day to obtain the difference between the power utilization curve of each day of the users and the power utilization curves of other days except the day;
and fusing the difference degrees corresponding to the electricity utilization curves of the user on each day to obtain a second electricity stealing possibility corresponding to the user.
4. The method of claim 3, wherein clustering the power usage profile of the user for each day comprises:
and clustering the electricity utilization curves of the users in each day through a fast clustering algorithm based on density peak values.
5. The method of claim 3, wherein the clustering the power usage curves of the users for each day to obtain the difference between the power usage curve of each day of the users and the power usage curves of other days except the day comprises:
clustering the power utilization curves of the users on each day to obtain the power utilization curve density and the power utilization curve distance corresponding to each day of the users;
and fusing the power utilization curve density and the power utilization curve distance of each day to obtain the difference degree corresponding to the power utilization curve of each day.
6. The method of claim 1, wherein determining electricity stealing subscribers in the cell based on the first and second electricity stealing probabilities for each subscriber comprises:
carrying out weighted fusion on the first electricity stealing possibility and the second electricity stealing possibility of each user to obtain fused electricity stealing possibility corresponding to each user;
and sequencing the fusion electricity stealing possibility corresponding to each user, and determining electricity stealing users in the distribution room according to a sequencing result.
7. The method of claim 6, wherein the weight corresponding to the first electricity stealing likelihood is adaptively varied from the weight corresponding to the second electricity stealing likelihood.
8. The method of claim 1, wherein determining the total error of power stealing in a cell on a given number of days comprises:
acquiring a measured value of the total power consumption of a distribution area in a specified number of days and a measured value of the power consumption corresponding to each user in the distribution area;
and calculating the total electricity stealing error corresponding to the distribution room according to the sum of the measured value corresponding to the total electricity consumption and the measured value corresponding to the user.
9. An electricity theft detection device, comprising:
the total error determining module is used for determining the total error of electricity stealing in the distribution area within the specified days;
a first electricity stealing possibility determining module, configured to determine, for each user in the distribution room, a correlation between a total electricity consumption vector corresponding to the user and the total electricity stealing error as a first electricity stealing possibility corresponding to the user, where the total electricity consumption vector is obtained by combining electricity consumption data of the user on each day within the specified number of days;
the second electricity stealing possibility determining module is used for determining a second electricity stealing possibility corresponding to each user in the distribution room according to the difference degree between the electricity utilization curves of the users on each day;
and the electricity stealing user determining module is used for determining electricity stealing users in the transformer area according to the first electricity stealing possibility and the second electricity stealing possibility of each user.
10. A computer-readable storage medium for storing program code for performing the power theft detection method of any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110430297.5A CN113128596A (en) | 2021-04-21 | 2021-04-21 | Electricity stealing detection method and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110430297.5A CN113128596A (en) | 2021-04-21 | 2021-04-21 | Electricity stealing detection method and device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113128596A true CN113128596A (en) | 2021-07-16 |
Family
ID=76778532
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110430297.5A Pending CN113128596A (en) | 2021-04-21 | 2021-04-21 | Electricity stealing detection method and device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113128596A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113884734A (en) * | 2021-10-27 | 2022-01-04 | 广东电网有限责任公司 | Non-invasive electricity utilization abnormity diagnosis method and device |
| 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 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108593990A (en) * | 2018-06-04 | 2018-09-28 | 国网天津市电力公司 | A kind of stealing detection method and application based on electric power users electricity consumption behavior pattern |
| CN108664990A (en) * | 2018-03-29 | 2018-10-16 | 清华大学 | The stealing detection method and device of comprehensive entropy method and Density Clustering method |
| CN110824270A (en) * | 2019-10-09 | 2020-02-21 | 中国电力科学研究院有限公司 | Electricity stealing user identification method and device combining transformer area line loss and abnormal events |
| CN111507611A (en) * | 2020-04-15 | 2020-08-07 | 北京中电普华信息技术有限公司 | Method and system for determining electricity stealing suspected user |
| CN111667144A (en) * | 2020-04-30 | 2020-09-15 | 北京中电普华信息技术有限公司 | User identification method and device |
-
2021
- 2021-04-21 CN CN202110430297.5A patent/CN113128596A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108664990A (en) * | 2018-03-29 | 2018-10-16 | 清华大学 | The stealing detection method and device of comprehensive entropy method and Density Clustering method |
| CN108593990A (en) * | 2018-06-04 | 2018-09-28 | 国网天津市电力公司 | A kind of stealing detection method and application based on electric power users electricity consumption behavior pattern |
| CN110824270A (en) * | 2019-10-09 | 2020-02-21 | 中国电力科学研究院有限公司 | Electricity stealing user identification method and device combining transformer area line loss and abnormal events |
| CN111507611A (en) * | 2020-04-15 | 2020-08-07 | 北京中电普华信息技术有限公司 | Method and system for determining electricity stealing suspected user |
| CN111667144A (en) * | 2020-04-30 | 2020-09-15 | 北京中电普华信息技术有限公司 | User identification method and device |
Non-Patent Citations (1)
| Title |
|---|
| KEDI ZHENG ET AL.: "A Novel Combined Data-Driven Approach for Electricity Theft Detection", 《IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS》, vol. 15, no. 3, pages 1809 - 1819, XP011713836, DOI: 10.1109/TII.2018.2873814 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113884734A (en) * | 2021-10-27 | 2022-01-04 | 广东电网有限责任公司 | Non-invasive electricity utilization abnormity diagnosis method and device |
| CN113884734B (en) * | 2021-10-27 | 2024-04-19 | 广东电网有限责任公司 | Non-invasive electricity consumption abnormality diagnosis method and device |
| 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 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2529186B1 (en) | Robust automated determination of the hierarchical structure of utility monitoring systems | |
| Su | Probabilistic load-flow computation using point estimate method | |
| Piao et al. | Subspace projection method based clustering analysis in load profiling | |
| CN110988422B (en) | Electricity stealing identification method and device and electronic equipment | |
| CN113128596A (en) | Electricity stealing detection method and device and computer readable storage medium | |
| Chicco et al. | Emergent electricity customer classification | |
| Abul-Haggag et al. | Application of fuzzy logic for risk assessment using risk matrix | |
| CN111310120B (en) | Abnormal electricity utilization user identification method, device, terminal and medium based on big data | |
| CN111178396A (en) | Method and device for identifying abnormal electricity consumption user | |
| CN112001409A (en) | Power distribution network line loss abnormity diagnosis method and system based on K-means clustering algorithm | |
| KR101733708B1 (en) | Method and system for rating measured values taken from a system | |
| CN112288303A (en) | Method and device for determining line loss rate | |
| CN109389517B (en) | Analysis method and device for quantifying line loss influence factors | |
| CN114066261A (en) | Tampering detection method and device for electric meter, computer equipment and storage medium | |
| CN116502894A (en) | Photovoltaic transformer area power failure risk assessment method and device, electronic equipment and storage medium | |
| CN115829334A (en) | Risk assessment method and system for power grid service | |
| CN113592371B (en) | Comprehensive risk analysis system, method and equipment based on multi-dimensional risk matrix | |
| CN114266485A (en) | Construction method and construction system of power information communication data quality detection model | |
| CN114240102A (en) | Line loss abnormal data identification method and device, electronic equipment and storage medium | |
| Sun et al. | A calculation method for a power user’s CIC under specific conditions in smart distribution grid | |
| Trevizan et al. | Distribution test system for nontechnical loss detection | |
| Peng et al. | Estimating robustness through Kirchhoff index in mesh graphs | |
| Khani et al. | An unsupervised learning based MCDM approach for optimal placement of fault indicators in distribution networks | |
| CN117709273B (en) | Battery risk prediction method, device, computer equipment and storage medium | |
| CN116388275A (en) | Distribution network safety analysis method and related device for distributed photovoltaic grid connection |
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 |