CN115051363B - Distribution network area user change relation identification method and device and computer storage medium - Google Patents

Distribution network area user change relation identification method and device and computer storage medium Download PDF

Info

Publication number
CN115051363B
CN115051363B CN202210984079.0A CN202210984079A CN115051363B CN 115051363 B CN115051363 B CN 115051363B CN 202210984079 A CN202210984079 A CN 202210984079A CN 115051363 B CN115051363 B CN 115051363B
Authority
CN
China
Prior art keywords
component
user
matrix
reconstructed
analysis measure
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
Application number
CN202210984079.0A
Other languages
Chinese (zh)
Other versions
CN115051363A (en
Inventor
张殷
李新
王俊波
李国伟
唐琪
范心明
董镝
宋安琪
黎小龙
刘少辉
吴焯军
章涛
刘昊
王云飞
李雷
涂琬婧
梁年柏
刘崧
赖艳珊
李兰茵
王智娇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Foshan Power Supply Bureau of Guangdong Power Grid Corp
Original Assignee
Foshan Power Supply Bureau of Guangdong Power Grid Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Foshan Power Supply Bureau of Guangdong Power Grid Corp filed Critical Foshan Power Supply Bureau of Guangdong Power Grid Corp
Priority to CN202210984079.0A priority Critical patent/CN115051363B/en
Publication of CN115051363A publication Critical patent/CN115051363A/en
Application granted granted Critical
Publication of CN115051363B publication Critical patent/CN115051363B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/0084—Arrangements for measuring currents or voltages or for indicating presence or sign thereof measuring voltage only
    • 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

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to the technical field of power distribution network management of a power system, and discloses a distribution network area user variation relation identification method and device and a computer storage medium. The method comprises the steps of carrying out empirical mode decomposition on voltage measurement signals of transformers in distribution network areas and users, measuring the complexity of IMF components of the voltage measurement signals obtained by decomposition by sample entropy, reconstructing the IMF components of the voltage measurement signals with similar sample entropy into trend components, detail components and random components, and obtaining corresponding reconstructed component matrixes; taking the multiresolution analysis measure as similarity measurement between the transformer and the user and between the user and the user, carrying out K-means clustering on the reconstruction component matrix, and determining the transformer-area-user relationship according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices. The method and the device are suitable for large-scale normalization development, and accuracy of identification of the user variable relationship is improved.

Description

Distribution network area user change relation identification method and device and computer storage medium
Technical Field
The invention relates to the technical field of power distribution network management of a power system, in particular to a method and a device for identifying a distribution network area user change relationship and a computer storage medium.
Background
Lean management of the power distribution network is an effective measure for realizing quality improvement and efficiency improvement of power enterprises, and accurate user variable relation information is a premise for developing lean management of the power distribution network. At present, topological information of the low-voltage distribution network depends on manual input, and due to the reasons that historical records are lost, the capacity expansion of distribution transformer and the information update is not timely when the line is moved and changed, the situation that a lot of errors exist in the relation of the household transformer in the transformer station account is caused, and the development of the lean management work of the transformer station area such as transformer station fault location, electricity stealing checking, line loss and three-phase imbalance management is seriously hindered.
The traditional identification method for the household variable relation mainly comprises a manual line patrol method, an instantaneous power failure method, a characteristic signal method and a transformer area identification instrument method. Wherein, the manual line patrol method depends on manual development and consumes manpower; the instantaneous power failure method influences the electricity consumption experience of customers and influences the reliability index of power supply; the characteristic signal method and the station area identification instrument method need to increase equipment investment and are easily influenced by interference noise. Therefore, the method is difficult to be normally carried out in a large scale, and the defects are obvious.
With the popularization and application of the electricity utilization acquisition system and the intelligent electric meter, massive user electricity utilization data are widely acquired, and the identification of the user variable relationship by using a data mining method becomes possible. Because the voltage fluctuation trends between users and the station transformers with similar electrical distances are similar, the distance and similarity measurement is used for calculating the correlation of voltage measurement curves between the users and the station transformers, so that the correlation between the users and the station transformers is measured, and the topological connection relation between the users and the station transformers is determined according to the distance and similarity results. However, the existing identification method for the user-variant relationship based on data mining mostly measures the feature space distance and the morphological similarity between the voltage measurement data from the overall perspective, but does not refine the local features of the voltage measurement data and does not realize the analysis of the feature space distance and the similarity under the multi-resolution, so the accuracy of the user-variant relationship identification still needs to be improved.
Disclosure of Invention
The invention provides a distribution network area user variable relation identification method, a distribution network area user variable relation identification device and a computer storage medium, which can realize large-scale normalized development and solve the technical problem that the user variable relation identification accuracy of the existing data mining-based user variable relation identification method is still to be improved.
The first aspect of the invention provides a distribution network area user variation relationship identification method, which comprises the following steps:
acquiring voltage measurement signals of a transformer and users in a distribution network area, and performing empirical mode decomposition on the basis of the voltage measurement signals to obtain a series of voltage measurement signal IMF components with different characteristics;
measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstructed component matrix;
taking the multi-resolution analysis measure as similarity measurement between transformers and users in the distribution network area and between users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-user variation relationship according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices.
According to a manner that can be realized by the first aspect of the present invention, the K-means clustering the reconstruction component matrix with a multi-resolution analysis measure as a similarity measure between the transformer of the transformer area and the user and between the users comprises:
selectingkEach transformer area is an initial clustering center;
according to the reconstruction component matrix, calculating the multiresolution analysis measure of the user and each initial clustering center, and classifying the user into a cluster corresponding to the clustering center with the closest distance or the largest similarity according to the calculation result;
updating the clustering center;
the classification and recalculation processes are iteratively performed until the error function converges or a maximum number of iterations is reached.
According to a possible implementation manner of the first aspect of the present invention, the calculating the multi-resolution analysis measure of the user and each initial cluster center includes:
let userxCorresponding reconstructed component matrix of
Figure 310733DEST_PATH_IMAGE001
，
Figure 216372DEST_PATH_IMAGE002
、
Figure 247258DEST_PATH_IMAGE003
And
Figure 958862DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 27312DEST_PATH_IMAGE005
、
Figure 103852DEST_PATH_IMAGE006
and
Figure 421701DEST_PATH_IMAGE007
are respectively as
Figure 874679DEST_PATH_IMAGE002
、
Figure 859953DEST_PATH_IMAGE003
And
Figure 107394DEST_PATH_IMAGE004
to (1) ajIndividual element, initial cluster centeryCorresponding reconstructed component matrix is
Figure 850222DEST_PATH_IMAGE008
，
Figure 841312DEST_PATH_IMAGE009
、
Figure 618775DEST_PATH_IMAGE010
And
Figure 40048DEST_PATH_IMAGE011
are respectively the initial cluster centeryCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 535751DEST_PATH_IMAGE012
、
Figure 2636DEST_PATH_IMAGE013
and
Figure 837868DEST_PATH_IMAGE014
are respectively as
Figure 427112DEST_PATH_IMAGE009
、
Figure 144533DEST_PATH_IMAGE010
And
Figure 474495DEST_PATH_IMAGE011
to (1)jAnd N is the number of elements, and the multi-resolution analysis measure of the user and each initial clustering center is calculated according to the following formula:
Figure 226551DEST_PATH_IMAGE015
in the formula,
Figure 986696DEST_PATH_IMAGE016
for the userxWith initial clustering centeryFor representing the userxWith initial clustering centeryMulti-resolution analysis of the measure.
According to an implementation manner of the first aspect of the present invention, the multi-resolution analysis measure is similarity between reconstruction component matrices, and the calculating the multi-resolution analysis measure between the user and each initial cluster center includes:
let userxCorresponding reconstructed component matrix of
Figure 191413DEST_PATH_IMAGE001
，
Figure 327996DEST_PATH_IMAGE002
、
Figure 872241DEST_PATH_IMAGE003
And
Figure 803288DEST_PATH_IMAGE004
are respectively usedHousexCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 424194DEST_PATH_IMAGE005
、
Figure 302151DEST_PATH_IMAGE006
and
Figure 763220DEST_PATH_IMAGE007
are respectively as
Figure 740534DEST_PATH_IMAGE002
、
Figure 854596DEST_PATH_IMAGE003
And
Figure 332982DEST_PATH_IMAGE004
to (1) ajElement, initial clustering centeryCorresponding reconstructed component matrix of
Figure 648557DEST_PATH_IMAGE008
，
Figure 249303DEST_PATH_IMAGE009
、
Figure 650328DEST_PATH_IMAGE010
And
Figure 932405DEST_PATH_IMAGE011
are respectively the initial cluster centeryCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 102486DEST_PATH_IMAGE012
、
Figure 811816DEST_PATH_IMAGE013
and
Figure 703067DEST_PATH_IMAGE014
are respectively as
Figure 523256DEST_PATH_IMAGE009
、
Figure 813423DEST_PATH_IMAGE010
And
Figure 693654DEST_PATH_IMAGE011
to (1)jThe number of the elements is one,
Figure 69272DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated as follows:
Figure 693151DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial cluster centery(ii) a multi-resolution analysis measure of (a);
Figure 775508DEST_PATH_IMAGE019
is a matrix
Figure 823711DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column are,
Figure 686625DEST_PATH_IMAGE021
is a matrix
Figure 910933DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
A second aspect of the present invention provides a distribution network area user change relationship identification device, including:
the signal decomposition module is used for acquiring voltage measurement signals of a transformer and a user in a distribution network area, and performing empirical mode decomposition on the basis of the voltage measurement signals to obtain a series of voltage measurement signal IMF components with different characteristics;
the signal reconstruction module is used for measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and acquiring a corresponding reconstruction component matrix;
the clustering module is used for taking the multi-resolution analysis measure as the similarity measure between transformers and users of the distribution network area and between users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-area user variation relation according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices.
According to an implementable manner of the second aspect of the present invention, the clustering module comprises:
a selection unit for selectingkEach transformer area is an initial clustering center;
the clustering unit is used for calculating the multiresolution analysis measure of the user and each initial clustering center according to the reconstructed component matrix and classifying the user into a cluster corresponding to the clustering center with the closest distance or the largest similarity according to the calculation result;
an updating unit for updating the clustering center;
and the iteration unit is used for iteratively executing the classification and recalculation processes until the error function is converged or the maximum iteration times is reached.
According to an implementable manner of the second aspect of the present invention, the multi-resolution analysis measure is a distance between the reconstructed component matrices, the clustering module comprises a first multi-resolution analysis measure calculating unit;
let userxCorresponding reconstructed component matrix of
Figure 175692DEST_PATH_IMAGE001
，
Figure 69830DEST_PATH_IMAGE002
、
Figure 357723DEST_PATH_IMAGE003
And
Figure 323405DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 207502DEST_PATH_IMAGE005
、
Figure 600437DEST_PATH_IMAGE006
and
Figure 172364DEST_PATH_IMAGE007
are respectively as
Figure 676157DEST_PATH_IMAGE002
、
Figure 915509DEST_PATH_IMAGE003
And
Figure 479345DEST_PATH_IMAGE004
to (1) ajElement, initial clustering centeryCorresponding reconstructed component matrix of
Figure 538568DEST_PATH_IMAGE008
，
Figure 843123DEST_PATH_IMAGE009
、
Figure 936981DEST_PATH_IMAGE010
And
Figure 671719DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 155921DEST_PATH_IMAGE012
、
Figure 63834DEST_PATH_IMAGE013
and
Figure 12199DEST_PATH_IMAGE014
are respectively as
Figure 855521DEST_PATH_IMAGE009
、
Figure 951653DEST_PATH_IMAGE010
And
Figure 603870DEST_PATH_IMAGE011
to (1)jAnd N is the number of elements, and the multi-resolution analysis measure of the user and each initial clustering center is calculated according to the following formula:
Figure 344424DEST_PATH_IMAGE015
in the formula,
Figure 358647DEST_PATH_IMAGE016
for the userxWith initial clustering centeryFor representing the userxWith initial cluster centeryMulti-resolution analysis of the measure.
According to an implementable aspect of the second aspect of the invention, the multiresolution analysis measure is a similarity between reconstructed component matrices, the clustering module comprises a second multiresolution analysis measure calculation unit;
let userxCorresponding reconstructed component matrix is
Figure 879758DEST_PATH_IMAGE001
，
Figure 270420DEST_PATH_IMAGE002
、
Figure 924867DEST_PATH_IMAGE003
And
Figure 172309DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 180716DEST_PATH_IMAGE005
、
Figure 171806DEST_PATH_IMAGE006
and
Figure 949269DEST_PATH_IMAGE007
are respectively as
Figure 429929DEST_PATH_IMAGE002
、
Figure 863316DEST_PATH_IMAGE003
And
Figure 392517DEST_PATH_IMAGE004
to (1)jElement, initial clustering centeryCorresponding reconstructed component matrix of
Figure 15698DEST_PATH_IMAGE008
，
Figure 808204DEST_PATH_IMAGE009
、
Figure 525625DEST_PATH_IMAGE010
And
Figure 858517DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 610572DEST_PATH_IMAGE012
、
Figure 370718DEST_PATH_IMAGE013
and
Figure 575434DEST_PATH_IMAGE014
are respectively as
Figure 646771DEST_PATH_IMAGE009
、
Figure 315650DEST_PATH_IMAGE010
And
Figure 184380DEST_PATH_IMAGE011
to (1)jThe number of the elements is one,
Figure 938709DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated as follows:
Figure 613404DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial cluster centery(ii) a multi-resolution analysis measure of (a);
Figure 74472DEST_PATH_IMAGE019
is a matrix
Figure 504317DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column(s) are,
Figure 683625DEST_PATH_IMAGE021
is a matrix
Figure 896432DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
The third aspect of the present invention provides a distribution network area change relationship identification device, including:
a memory to store instructions; the instruction is used for realizing the distribution network area user change relationship identification method in any one of the realizable modes;
a processor to execute the instructions in the memory.
A fourth aspect of the present invention is a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method for identifying a distribution network area subscriber identity relationship is implemented as described in any one of the above-mentioned implementable manners.
According to the technical scheme, the invention has the following advantages:
the method is based on voltage measurement signals of a transformer and users in a distribution network area to carry out empirical mode decomposition to obtain a series of voltage measurement signal IMF components with different characteristics; measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstructed component matrix; taking the multi-resolution analysis measure as similarity measurement between transformers in distribution network areas and users and between users, carrying out K-means clustering on the reconstructed component matrix, and determining the area-to-user variable relation according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices; the method and the device are not dependent on manual development, do not influence the electricity consumption experience of customers, do not need to increase equipment investment, are suitable for large-scale normalized development, refine local characteristics of voltage measurement data, realize multi-resolution characteristic space distance and similarity analysis of voltage measurement signals between users and between users and station transformers, effectively realize analysis of the station transformer relationship of the distribution transformer area, and improve the identification accuracy of the station transformer relationship.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive exercise.
Fig. 1 is a flowchart of a distribution network area change relationship identification method according to an optional embodiment of the present invention;
FIG. 2 is a detailed flow diagram of an exemplary distribution network area subscriber relationship identification algorithm;
fig. 3 is a connection block diagram of a structure of a distribution network area change relationship identification device according to an optional embodiment of the present invention.
Reference numerals are as follows:
1-a signal decomposition module; 2-a signal reconstruction module; and 3, a clustering module.
Detailed Description
The embodiment of the invention provides a distribution network area user-to-user relationship identification method, a distribution network area user-to-user relationship identification device and a computer storage medium, which are used for solving the technical problem that the user-to-user relationship identification accuracy of the existing data mining-based user-to-user relationship identification method still needs to be improved.
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
The invention provides a distribution network area user variation relation identification method.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for identifying a distribution network area subscriber relationship provided by an embodiment of the present invention.
The method for identifying the distribution network area user variation relationship comprises the steps S1-S3.
S1, voltage measurement signals of a transformer and a user in a distribution network area are obtained, empirical mode decomposition is carried out on the basis of the voltage measurement signals, and a series of voltage measurement signal IMF components with different characteristics are obtained.
In this embodiment, an Empirical Mode Decomposition (EMD) is adopted to adaptively decompose the voltage measurement signal into a series of subsequences with different characteristics, and to decouple the characteristic scale information, thereby completing the multi-resolution fine depiction of the original voltage measurement data of the user and the transformer.
And performing empirical mode decomposition on the voltage measurement signals, wherein data centralization processing can be performed on the voltage measurement signals of the distribution network transformer and users. The treatment process comprises the following steps: calculating an average value of the voltage measurement signals; and subtracting the average value from each sampling point of each voltage measurement signal to obtain the decentralized voltage measurement signal.
Wherein empirical mode decomposition is performed based on the voltage measurement signal, comprising:
step S1.1, extracting voltage measurement data from a measurement database, and searching for an original voltage measurement signal
Figure 539903DEST_PATH_IMAGE023
All maximum and minimum points;
s1.2, connecting all maximum value points by using a curve, and forming an upper envelope of the signal by fitting
Figure 815682DEST_PATH_IMAGE024
(ii) a Connecting all minimum value points by using a curve, and forming a lower envelope of the signal by fitting
Figure 482287DEST_PATH_IMAGE025
Calculating the average of the upper and lower envelopes:
Figure 498784DEST_PATH_IMAGE026
in the formula,
Figure 934445DEST_PATH_IMAGE027
the average value of the upper envelope line and the lower envelope line;
step S1.3, calculating the original voltage measurement signal
Figure 643775DEST_PATH_IMAGE023
And the average value
Figure 532096DEST_PATH_IMAGE027
The difference of (c):
Figure 352285DEST_PATH_IMAGE028
in the formula,
Figure 642452DEST_PATH_IMAGE029
measuring the signal for the original voltage
Figure 457437DEST_PATH_IMAGE023
And the average value
Figure 833054DEST_PATH_IMAGE027
A difference of (d);
step S1.4, if
Figure 191355DEST_PATH_IMAGE029
Not satisfying the requirements of intrinsic mode function
Figure 601607DEST_PATH_IMAGE029
Turning to step S1.1 for a new signal; if yes, then order
Figure 652740DEST_PATH_IMAGE029
Is an IMF component, wherein
Figure 515654DEST_PATH_IMAGE030
An IMF component
Figure 677645DEST_PATH_IMAGE031
Expressed as:
Figure 629156DEST_PATH_IMAGE032
step S1.5, the original voltage quantity is measuredSignal detection
Figure 913507DEST_PATH_IMAGE023
And
Figure 263717DEST_PATH_IMAGE031
the remaining components of the difference are taken as new signals and step S1.1 is carried out until all components are obtained;
step S1.6, original voltage measurement signal
Figure 167082DEST_PATH_IMAGE023
EMD decomposition into K IMF components
Figure 348665DEST_PATH_IMAGE031
And 1 residual component
Figure 476021DEST_PATH_IMAGE033
Then, then
Figure 313527DEST_PATH_IMAGE023
Can be expressed as:
Figure 817320DEST_PATH_IMAGE034
in the formula,
Figure 53742DEST_PATH_IMAGE035
the number of IMF components;
and S1.7, completing the multi-resolution decomposition of the original voltage measurement signals of the user and the station through the steps S1.1-S1.6 to obtain a series of subsequences of each measurement signal.
And S2, measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstructed component matrix.
After the voltage measurement signal is decomposed by adopting an EMD algorithm, more IMF components are generated, and partial IMF components have the phenomenon of similar decomposition characteristics. Based on this, the step further reconstructs the voltage measurement signal IMF component, measures the complexity of each measurement signal subsequence, namely the voltage measurement signal IMF component, by using Sample Entropy (SE), and completes the merging and recombination of similar subsequences to obtain a reconstructed component matrix, thereby completing the preparation of data for identifying the user variable relationship.
And reconstructing IMF components of the voltage measurement signals with similar sample entropies into a trend component, a detail component and a random component, wherein the steps comprise S2.1-S2.7.
Step S2.1, utilizing the fixed window width ofmSliding window of (2) the sub-sequence obtained by EMD decomposition
Figure 351999DEST_PATH_IMAGE036
Decompose into a groupmVector sequence of dimensions
Figure 676802DEST_PATH_IMAGE037
. Wherein,
Figure 984286DEST_PATH_IMAGE038
，Nis the number of data points in the sequence.
Step S2.2, defining vectors
Figure 812565DEST_PATH_IMAGE039
And
Figure 547303DEST_PATH_IMAGE040
the distance between them is:
Figure 93822DEST_PATH_IMAGE041
in the formula,
Figure 204997DEST_PATH_IMAGE042
and is made of
Figure 156291DEST_PATH_IMAGE043
，
Figure 61930DEST_PATH_IMAGE044
。
Step S2.3, setting similar tolerancetStatistical vector
Figure 33428DEST_PATH_IMAGE039
And
Figure 745033DEST_PATH_IMAGE040
at a distance of less thantNumber of (2) and total distanceN-mThe ratio of (A) to (B):
Figure 547903DEST_PATH_IMAGE045
in the formula, num represents the number statistics,
Figure 624444DEST_PATH_IMAGE046
and is
Figure 145555DEST_PATH_IMAGE047
。
Step S2.4, averaging the ratios:
Figure 395271DEST_PATH_IMAGE048
step S2.5, the sequence dimension is determined bymChange from vitamin to vitaminmDimension +1, repeating the steps S2.1-S2.4, and calculating to obtain
Figure 318227DEST_PATH_IMAGE049
。
Step S2.6 whenNFor finite values, the sample entropy is:
Figure 500423DEST_PATH_IMAGE050
wherein,mandtthe normal values are 2 and 0.2std, and std is the time sequence standard deviation.
Step S2.7, combining with steps S2.1-S2.6, calculates the original voltage measurement signal
Figure 571147DEST_PATH_IMAGE023
Is entropy of
Figure 562237DEST_PATH_IMAGE051
Calculating each IMF component
Figure 339700DEST_PATH_IMAGE031
Is entropy of
Figure 758043DEST_PATH_IMAGE052
And completing component reconstruction according to the entropy values of the subsequence samples to form random components, detail components and trend components so as to obtain data for identifying the user-variable relationship.
Wherein the reconstructed random componentRThe expression is as follows:
Figure 50484DEST_PATH_IMAGE053
in the formula,
Figure 845264DEST_PATH_IMAGE054
，γis the range threshold.γThe value is determined by the entropy distribution of the subsequence samples, and the value can be selected to be 0.05.
Reconstructed detail componentsDThe expression is as follows:
Figure 477234DEST_PATH_IMAGE055
in the formula,
Figure 128795DEST_PATH_IMAGE056
。
reconstructed trend componentTThe expression is as follows:
Figure 40689DEST_PATH_IMAGE057
in the formula,
Figure 435898DEST_PATH_IMAGE058
。
s3, taking the multi-resolution analysis measure as similarity measurement between transformers and users of the distribution network area and between the users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-area relationship according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices.
In the embodiment, the K-means clustering method is improved, all transformer areas to be analyzed are used as initial clustering centers, the multi-resolution analysis measure is used as similarity measurement between users and between station changes, and the users are divided into user clusters belonging to each station change.
In one implementation, the K-means clustering the reconstructed component matrix with a multi-resolution analysis measure as a similarity measure between a transformer and a user and between users includes:
selectingkEach transformer area is an initial clustering center;
according to the reconstructed component matrix, calculating the multi-resolution analysis measure of the user and each initial clustering center, and classifying the user into a cluster corresponding to the clustering center with the closest distance or the largest similarity according to the calculation result;
updating the clustering center;
the classification and recalculation processes are iteratively performed until the error function converges or a maximum number of iterations is reached.
For example, when the distance between the reconstruction component matrices is selected as the multi-resolution analysis measure, the user is classified into the cluster corresponding to the cluster center closest to the distance or having the greatest similarity according to the calculation result, specifically:
calculate the firsteIndividual userU e And a firstfIndividual cluster centerTr f (ii) a multi-resolution analysis measure dis: (U e ,Tr f ) Wherein, in the process,
Figure 922374DEST_PATH_IMAGE059
let us order
Figure DEST_PATH_IMAGE060
，zTo and from the usereThe cluster center number with the minimum multiresolution analysis measure of (2) is obtained, then the usereAscribed to the clustering centerzThe user cluster.
Let userxCorresponding reconstructed component matrix of
Figure 885782DEST_PATH_IMAGE001
，
Figure 90499DEST_PATH_IMAGE002
、
Figure 289399DEST_PATH_IMAGE003
And
Figure 630381DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 827007DEST_PATH_IMAGE005
、
Figure 453773DEST_PATH_IMAGE006
and
Figure 128468DEST_PATH_IMAGE007
are respectively as
Figure 651854DEST_PATH_IMAGE002
、
Figure 19381DEST_PATH_IMAGE003
And
Figure 933110DEST_PATH_IMAGE004
to (1)jElement, initial clustering centeryCorresponding reconstructed componentsThe matrix is
Figure 411496DEST_PATH_IMAGE008
，
Figure 992650DEST_PATH_IMAGE009
、
Figure 327817DEST_PATH_IMAGE010
And
Figure 994421DEST_PATH_IMAGE011
are respectively the initial cluster centeryCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 10919DEST_PATH_IMAGE012
、
Figure 449509DEST_PATH_IMAGE013
and
Figure 893260DEST_PATH_IMAGE014
are respectively as
Figure 47161DEST_PATH_IMAGE009
、
Figure 805032DEST_PATH_IMAGE010
And
Figure 95199DEST_PATH_IMAGE011
to (1) ajAnd N is the number of elements, and the multi-resolution analysis measure of the user and each initial clustering center is calculated according to the following formula:
Figure 975431DEST_PATH_IMAGE015
in the formula,
Figure 351048DEST_PATH_IMAGE016
for the userxWith initial clustering centeryDistance between the reconstructed component matrices of (1)For representing usersxWith initial cluster centeryMulti-resolution analysis of the measure.
In another implementation manner, the multiresolution analysis measure is similarity between the reconstructed component matrices, and the multiresolution analysis measure of the user and each initial cluster center is calculated according to the following formula:
Figure 706419DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial cluster centery(ii) a multi-resolution analysis measure of (a);
Figure 116672DEST_PATH_IMAGE019
is a matrix
Figure 902225DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column are,
Figure 765139DEST_PATH_IMAGE021
is a matrix
Figure 927130DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
In another implementation, the multiresolution analysis measures are the distance and similarity between the reconstructed component matrices, and the multiresolution analysis measure of the user and each initial cluster center is calculated according to the following formula:
Figure 191889DEST_PATH_IMAGE061
in the formula,
Figure 413923DEST_PATH_IMAGE062
representing a multi-resolution analysis measure of the user from each initial cluster center,
Figure 764133DEST_PATH_IMAGE063
to represent
Figure 526552DEST_PATH_IMAGE016
The normalized value of (a) is calculated,
Figure 660467DEST_PATH_IMAGE064
to representr(x,y) The normalized value of (a) is calculated,
Figure 53402DEST_PATH_IMAGE065
is a weight of the degree of similarity,
Figure 890908DEST_PATH_IMAGE066
is the distance weight.
As a way to be realized, when updating the clustering center, the average value of the user voltage measurements in each cluster is made to be a new clustering center;
Figure 457018DEST_PATH_IMAGE067
in the formula,
Figure 430791DEST_PATH_IMAGE068
in clusters ofC i The cluster center of (a);xis the voltage measurement of the user.
As a way to do this, the error function is set to:
Figure 994627DEST_PATH_IMAGE069
in the formula,
Figure 319429DEST_PATH_IMAGE070
is the cluster squared error.
According to the method of the embodiment of the invention, if the distance or the similarity between the reconstructed component matrixes is taken as the multi-resolution analysis measure, a corresponding distribution network area user variable relationship identification algorithm can be set in specific implementation.
Illustratively, as shown in fig. 2, the detailed flow of the distribution network area user-to-user relationship identification algorithm includes steps 1-11.
Step 1, taking out voltage measurement data of a low-voltage user electric meter and voltage measurement data of a transformer low-voltage side, and subtracting a sample average value from each voltage sequence sample value to finish data centralization processing;
step 2, extracting IMF components and residual components of the voltage sequence data by using EMD, and recombining the subsequence into a trend component, a detail component and a random component by using SE to obtain data for identifying the user variable relationship;
step 3, initializing parameters of the improved K-means clustering algorithm and the number of clustering categorieskTaking a variable of a platform to be analyzed, taking the three-phase voltage average value on the low voltage side of the platform as an initial clustering center, and selecting a multi-resolution analysis measure type;
step 4, makee=1, count number of users to be analyzed
Figure 361335DEST_PATH_IMAGE071
；
Step 5, if
Figure DEST_PATH_IMAGE072
Executing step 6, otherwise, executing step 9;
step 6, if the selected measure type is the multi-resolution distance measure, executing step 7, otherwise, executing step 8;
step 7, calculating the usereAnd withkThe voltage measurement data of each cluster center has multi-resolution distance, and users can use the distanceeClassifying the cluster with the minimum distance to the cluster centere=e+1, go to step 5;
step 8, calculating the usereAnd withkThe multi-resolution similarity of the voltage measurement data of the cluster centers enables users to share the voltage measurement dataeIs assigned to the cluster with the highest similarity cluster center, ande=e+1, go to step 5;
step 9, updating the clustering centers, and enabling the mean value of each cluster sample to be a new clustering center;
step 10, if the clustering error function is converged, executing step 11, otherwise, extracting IMF components and residual components of new clustering center voltage sequence data by using EMD, recombining to obtain trend components, detail components and random components, and executing step 4;
and 11, finishing the algorithm flow, and finally obtaining the user attribution clustering cluster number which is the user attribution station change number, thereby finishing the user change relationship identification.
The invention also provides a distribution network area user change relationship identification device which can be used for executing the distribution network area user change relationship identification method in any embodiment.
Referring to fig. 3, fig. 3 is a block diagram illustrating a structural connection of a distribution network area-to-user relationship identification device according to an embodiment of the present invention.
The device for identifying the household transformation relationship of the distribution network area provided by the embodiment of the invention comprises:
the system comprises a signal decomposition module 1, a voltage measurement module and a power supply module, wherein the signal decomposition module is used for acquiring voltage measurement signals of a transformer and a user in a distribution network area, and performing empirical mode decomposition on the basis of the voltage measurement signals to obtain a series of voltage measurement signal IMF components with different characteristics;
the signal reconstruction module 2 is used for measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstruction component matrix;
the clustering module 3 is used for taking the multi-resolution analysis measure as the similarity measure between transformers and users of the distribution network area and between users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-user variation relation according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices.
In an implementable manner, the clustering module 3 comprises:
a selection unit for selectingkEach transformer area is an initial clustering center;
the clustering unit is used for calculating the multiresolution analysis measure of the user and each initial clustering center according to the reconstructed component matrix and classifying the user into a cluster corresponding to the clustering center with the closest distance or the largest similarity according to the calculation result;
an updating unit for updating the clustering center;
and the iteration unit is used for iteratively executing the classification and recalculation processes until the error function is converged or the maximum iteration times is reached.
In an implementable manner, the multi-resolution analysis measure is a distance between reconstructed component matrices, the clustering module 3 comprises a first multi-resolution analysis measure calculation unit;
let userxCorresponding reconstructed component matrix of
Figure 389946DEST_PATH_IMAGE001
，
Figure 124684DEST_PATH_IMAGE002
、
Figure 733520DEST_PATH_IMAGE003
And
Figure 844695DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 793060DEST_PATH_IMAGE005
、
Figure 698699DEST_PATH_IMAGE006
and
Figure 794831DEST_PATH_IMAGE007
are respectively as
Figure 178539DEST_PATH_IMAGE002
、
Figure 246989DEST_PATH_IMAGE003
And
Figure 323529DEST_PATH_IMAGE004
to (1)jIndividual element, initial cluster centeryCorresponding reconstructed component matrix is
Figure 906957DEST_PATH_IMAGE008
，
Figure 97286DEST_PATH_IMAGE009
、
Figure 20242DEST_PATH_IMAGE010
And
Figure 267684DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 276091DEST_PATH_IMAGE012
、
Figure 267181DEST_PATH_IMAGE013
and
Figure 920011DEST_PATH_IMAGE014
are respectively as
Figure 23839DEST_PATH_IMAGE009
、
Figure 253964DEST_PATH_IMAGE010
And
Figure 48744DEST_PATH_IMAGE011
to (1) ajAnd N is the number of elements, and the multi-resolution analysis measure of the user and each initial clustering center is calculated according to the following formula:
Figure 609608DEST_PATH_IMAGE015
in the formula，
Figure 198852DEST_PATH_IMAGE016
For the userxWith initial cluster centeryFor representing the userxWith initial clustering centeryMulti-resolution analysis of (2).
In an implementable manner, the multi-resolution analysis measure is a similarity between the reconstructed component matrices, the clustering module 3 comprises a second multi-resolution analysis measure calculating unit;
let userxCorresponding reconstructed component matrix is
Figure 181852DEST_PATH_IMAGE001
，
Figure 514744DEST_PATH_IMAGE002
、
Figure 1220DEST_PATH_IMAGE003
And
Figure 89262DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 293978DEST_PATH_IMAGE005
、
Figure 164982DEST_PATH_IMAGE006
and
Figure 768614DEST_PATH_IMAGE007
are respectively as
Figure 27557DEST_PATH_IMAGE002
、
Figure 453991DEST_PATH_IMAGE003
And
Figure 394265DEST_PATH_IMAGE004
to (1) ajElement, initial clustering centeryCorresponding reconstructed component matrix of
Figure 855333DEST_PATH_IMAGE008
，
Figure 19598DEST_PATH_IMAGE009
、
Figure 198907DEST_PATH_IMAGE010
And
Figure 677293DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 55185DEST_PATH_IMAGE012
、
Figure 593613DEST_PATH_IMAGE013
and
Figure 994639DEST_PATH_IMAGE014
are respectively as
Figure 279645DEST_PATH_IMAGE009
、
Figure 715306DEST_PATH_IMAGE010
And
Figure 221373DEST_PATH_IMAGE011
to (1) ajThe number of the elements is one,
Figure 375274DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated according to the following formula:
Figure 195463DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial clustering centery(ii) a multi-resolution analysis measure of (a);
Figure 485630DEST_PATH_IMAGE019
is a matrix
Figure 100282DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column(s) are,
Figure 475900DEST_PATH_IMAGE021
is a matrix
Figure 99779DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
The invention also provides a distribution network area user change relationship identification device, which comprises:
a memory to store instructions; the instruction is used for realizing the distribution network area user change relationship identification method in any one of the embodiments;
a processor to execute the instructions in the memory.
The invention further provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when being executed by a processor, the computer program realizes the distribution network area user change relationship identification method according to any one of the above embodiments.
According to the embodiment of the invention, manual development is not relied on, the electricity consumption experience of customers is not influenced, the investment of newly-added equipment is not required, the method is suitable for large-scale normalized development, the local characteristics of the voltage measurement data are refined by the identification method, the multi-resolution characteristic space distance and similarity analysis of the voltage measurement signals between users and between station transformers are realized, the analysis of the station transformer area station transformer relationship can be effectively realized, and the identification accuracy of the station transformer area station transformer relationship is improved.
It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described apparatuses, modules and units may refer to the corresponding processes in the foregoing method embodiments, and the specific beneficial effects of the above-described apparatuses, modules and units may refer to the corresponding beneficial effects in the foregoing method embodiments, which are not described herein again.
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 ways. 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.
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 this embodiment.
In addition, functional modules in the embodiments of the present invention may be integrated into one processing module, or each module 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 invention may be embodied in the form of 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 invention. 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 various media capable of storing program codes.
The above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will 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 of the embodiments of the present invention.

Claims (4)

1. A distribution network area user change relationship identification method is characterized by comprising the following steps:
acquiring voltage measurement signals of a transformer and users in a distribution network area, and performing empirical mode decomposition on the basis of the voltage measurement signals to obtain a series of voltage measurement signal IMF components with different characteristics;
measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstructed component matrix;
taking the multi-resolution analysis measure as similarity measurement between transformers and users in the distribution network area and between users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-user variation relationship according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices;
the K-means clustering is carried out on the reconstruction component matrix by taking the multi-resolution analysis measure as the similarity measurement between the transformer of the transformer area and the user and between the users, and comprises the following steps:
selectingkEach transformer area is an initial clustering center;
according to the reconstructed component matrix, calculating the multi-resolution analysis measure of the user and each initial clustering center, and classifying the user into a cluster corresponding to the clustering center with the closest distance or the largest similarity according to the calculation result;
updating the clustering center;
iteratively executing the classification and recalculation processes until the error function converges or the maximum iteration times is reached;
when the multiresolution analysis measure is the distance between the reconstruction component matrixes, the calculating the multiresolution analysis measure of the user and each initial clustering center comprises the following steps:
let userxCorresponding reconstructed component matrix of
Figure 909901DEST_PATH_IMAGE001
，
Figure 246336DEST_PATH_IMAGE002
、
Figure 738497DEST_PATH_IMAGE003
And
Figure 527593DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 130612DEST_PATH_IMAGE005
、
Figure 587133DEST_PATH_IMAGE006
and
Figure 922299DEST_PATH_IMAGE007
are respectively as
Figure 713538DEST_PATH_IMAGE002
、
Figure 605401DEST_PATH_IMAGE003
And
Figure 165696DEST_PATH_IMAGE004
to (1)jIndividual element, initial cluster centeryCorresponding reconstructed component matrix is
Figure 484813DEST_PATH_IMAGE008
，
Figure 763347DEST_PATH_IMAGE009
、
Figure 645853DEST_PATH_IMAGE010
And
Figure 811386DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 816251DEST_PATH_IMAGE012
、
Figure 84813DEST_PATH_IMAGE013
and
Figure 567747DEST_PATH_IMAGE014
are respectively as
Figure 40317DEST_PATH_IMAGE009
、
Figure 701236DEST_PATH_IMAGE010
And
Figure 688784DEST_PATH_IMAGE011
to (1) ajAnd N is the number of elements, and the multi-resolution analysis measure of the user and each initial clustering center is calculated according to the following formula:
Figure 913092DEST_PATH_IMAGE015
in the formula,
Figure 53217DEST_PATH_IMAGE016
for the userxWith initial clustering centeryFor representing the user, the distance between the reconstructed component matrices ofxWith initial cluster centery(ii) a multi-resolution analysis measure of (a);
when the multiresolution analysis measure is the similarity between the reconstruction component matrixes, the calculating the multiresolution analysis measure of the user and each initial clustering center comprises the following steps:
let userxCorresponding reconstructed component matrix is
Figure 399885DEST_PATH_IMAGE001
，
Figure 625461DEST_PATH_IMAGE002
、
Figure 450197DEST_PATH_IMAGE003
And
Figure 444829DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 962398DEST_PATH_IMAGE005
、
Figure 862221DEST_PATH_IMAGE006
and
Figure 241381DEST_PATH_IMAGE007
are respectively as
Figure 339787DEST_PATH_IMAGE002
、
Figure 778990DEST_PATH_IMAGE003
And
Figure 228426DEST_PATH_IMAGE004
to (1)jElement, initial clustering centeryCorresponding reconstructed component matrix of
Figure 145697DEST_PATH_IMAGE008
，
Figure 301872DEST_PATH_IMAGE009
、
Figure 161244DEST_PATH_IMAGE010
And
Figure 577270DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 813079DEST_PATH_IMAGE012
、
Figure 636810DEST_PATH_IMAGE013
and
Figure 604766DEST_PATH_IMAGE014
are respectively as
Figure 763214DEST_PATH_IMAGE009
、
Figure 22289DEST_PATH_IMAGE010
And
Figure 215373DEST_PATH_IMAGE011
to (1) ajThe number of the elements is one,
Figure 354230DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated as follows:
Figure 750707DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial clustering centery(ii) a multi-resolution analysis measure of (a);
Figure 62740DEST_PATH_IMAGE019
is a matrix
Figure 861063DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column(s) are,
Figure 170821DEST_PATH_IMAGE021
is a matrix
Figure 303862DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
2. The utility model provides a join in marriage net platform district family and become relation identification device which characterized in that includes:
the signal decomposition module is used for acquiring voltage measurement signals of a transformer and a user in a distribution network area, and performing empirical mode decomposition on the basis of the voltage measurement signals to obtain a series of voltage measurement signal IMF components with different characteristics;
the signal reconstruction module is used for measuring the complexity of each voltage measurement signal IMF component by using sample entropy, reconstructing the voltage measurement signal IMF components with similar sample entropy into a trend component, a detail component and a random component, and obtaining a corresponding reconstruction component matrix;
the clustering module is used for taking the multi-resolution analysis measure as the similarity measure between transformers and users of the distribution network area and between users, carrying out K-means clustering on the reconstruction component matrix, and determining the area-to-area user variation relation according to the obtained clustering result; wherein the multi-resolution analysis measure is a distance and/or a similarity between the reconstructed component matrices;
the clustering module comprises:
a selection unit for selectingkEach transformer area is an initial clustering center;
the clustering unit is used for calculating the multi-resolution analysis measure of the user and each initial clustering center according to the reconstructed component matrix, and classifying the user into a cluster corresponding to the clustering center closest to or most similar to the user according to the calculation result;
an updating unit for updating the clustering center;
the iteration unit is used for iteratively executing the classification and recalculation processes until the error function converges or the maximum iteration times is reached;
when the multi-resolution analysis measure is the distance between the reconstruction component matrixes, the clustering module comprises a first multi-resolution analysis measure calculating unit;
let userxCorresponding reconstructed component matrix of
Figure 170318DEST_PATH_IMAGE001
，
Figure 72415DEST_PATH_IMAGE002
、
Figure 366125DEST_PATH_IMAGE003
And
Figure 658566DEST_PATH_IMAGE004
are respectively usersxCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 577980DEST_PATH_IMAGE005
、
Figure 85316DEST_PATH_IMAGE006
and
Figure 799194DEST_PATH_IMAGE007
are respectively as
Figure 651700DEST_PATH_IMAGE002
、
Figure 46910DEST_PATH_IMAGE003
And
Figure 658019DEST_PATH_IMAGE004
to (1) ajElement, initial clustering centeryCorresponding reconstructed component matrix is
Figure 559111DEST_PATH_IMAGE008
，
Figure 888461DEST_PATH_IMAGE009
、
Figure 634831DEST_PATH_IMAGE010
And
Figure 303710DEST_PATH_IMAGE011
are respectively initial cluster centersyCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 624970DEST_PATH_IMAGE012
、
Figure 926769DEST_PATH_IMAGE013
and
Figure 991677DEST_PATH_IMAGE014
are respectively as
Figure 515062DEST_PATH_IMAGE009
、
Figure 492377DEST_PATH_IMAGE010
And
Figure 796319DEST_PATH_IMAGE011
to (1) ajThe number of the elements is one,
Figure 150071DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated as follows:
Figure 527963DEST_PATH_IMAGE015
in the formula,
Figure 191025DEST_PATH_IMAGE016
for the userxWith initial clustering centeryFor representing the userxWith initial clustering centery(ii) a multi-resolution analysis measure of (a);
when the multi-resolution analysis measure is the similarity between the reconstruction component matrixes, the clustering module comprises a second multi-resolution analysis measure calculating unit;
let userxCorresponding reconstructed component matrix is
Figure 467417DEST_PATH_IMAGE001
，
Figure 874128DEST_PATH_IMAGE002
、
Figure 372105DEST_PATH_IMAGE003
And
Figure 691222DEST_PATH_IMAGE004
are respectively usedHousexCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 969757DEST_PATH_IMAGE005
、
Figure 659452DEST_PATH_IMAGE006
and
Figure 11936DEST_PATH_IMAGE007
are respectively as
Figure 751222DEST_PATH_IMAGE002
、
Figure 2206DEST_PATH_IMAGE003
And
Figure 750719DEST_PATH_IMAGE004
to (1)jIndividual element, initial cluster centeryCorresponding reconstructed component matrix is
Figure 957709DEST_PATH_IMAGE008
，
Figure 884208DEST_PATH_IMAGE009
、
Figure 871756DEST_PATH_IMAGE010
And
Figure 909113DEST_PATH_IMAGE011
are respectively the initial cluster centeryCorresponding to the trend component, detail component and random component of the reconstructed component matrix,
Figure 236189DEST_PATH_IMAGE012
、
Figure 582857DEST_PATH_IMAGE013
and
Figure 542854DEST_PATH_IMAGE014
are respectively as
Figure 570853DEST_PATH_IMAGE009
、
Figure 814752DEST_PATH_IMAGE010
And
Figure 83053DEST_PATH_IMAGE011
to (1) ajThe number of the elements is one,
Figure 45193DEST_PATH_IMAGE017
for the number of elements, the multiresolution analysis measure of the user and each initial cluster center is calculated according to the following formula:
Figure 611304DEST_PATH_IMAGE018
in the formula,r(x,y) Is a samplexAndyfor representing the userxWith initial clustering centery(ii) a multi-resolution analysis measure of (a);
Figure 460442DEST_PATH_IMAGE019
is a matrix
Figure 148913DEST_PATH_IMAGE020
To middleiGo to the firstjThe elements of the column(s) are,
Figure 83502DEST_PATH_IMAGE021
is a matrix
Figure 453303DEST_PATH_IMAGE022
To middleiGo to the firstjThe elements of the column.
3. The utility model provides a join in marriage net platform district family and become relation identification device which characterized in that includes:
a memory to store instructions; the instruction is used for realizing the distribution network area user change relationship identification method as claimed in claim 1;
a processor to execute the instructions in the memory.
4. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when executed by a processor, the computer program implements the distribution network change relationship identification method according to claim 1.
CN202210984079.0A 2022-08-17 2022-08-17 Distribution network area user change relation identification method and device and computer storage medium Active CN115051363B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210984079.0A CN115051363B (en) 2022-08-17 2022-08-17 Distribution network area user change relation identification method and device and computer storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210984079.0A CN115051363B (en) 2022-08-17 2022-08-17 Distribution network area user change relation identification method and device and computer storage medium

Publications (2)

Publication Number Publication Date
CN115051363A CN115051363A (en) 2022-09-13
CN115051363B true CN115051363B (en) 2023-01-13

Family

ID=83168417

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210984079.0A Active CN115051363B (en) 2022-08-17 2022-08-17 Distribution network area user change relation identification method and device and computer storage medium

Country Status (1)

Country Link
CN (1) CN115051363B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115687952B (en) * 2023-01-03 2023-11-07 南昌工程学院 Distribution network line transformation relation identification method and device based on Riemann manifold clustering

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839197A (en) * 2014-03-19 2014-06-04 国家电网公司 Method for judging abnormal electricity consumption behaviors of users based on EEMD method
CN109726865A (en) * 2018-12-27 2019-05-07 国网江苏省电力有限公司电力科学研究院 Customer charge probability density prediction technique, device and storage medium based on EMD-QRF
CN111327118A (en) * 2020-03-31 2020-06-23 西安理工大学 Non-invasive power load identification method
WO2020162937A1 (en) * 2019-02-07 2020-08-13 General Electric Company Automated model validation system for electrical grid
CN111612074A (en) * 2020-05-22 2020-09-01 王彬 Identification method and device of non-invasive load monitoring electric equipment and related equipment
CN113780440A (en) * 2021-09-15 2021-12-10 江苏方天电力技术有限公司 Low-voltage station area phase identification method for improving data disturbance resistance
CN114611869A (en) * 2022-01-24 2022-06-10 国家电网有限公司 Low-voltage transformer area subscriber identification method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114024308A (en) * 2021-11-09 2022-02-08 广东电网有限责任公司广州供电局 Power distribution network platform relationship identification method and system

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103839197A (en) * 2014-03-19 2014-06-04 国家电网公司 Method for judging abnormal electricity consumption behaviors of users based on EEMD method
CN109726865A (en) * 2018-12-27 2019-05-07 国网江苏省电力有限公司电力科学研究院 Customer charge probability density prediction technique, device and storage medium based on EMD-QRF
WO2020162937A1 (en) * 2019-02-07 2020-08-13 General Electric Company Automated model validation system for electrical grid
CN111327118A (en) * 2020-03-31 2020-06-23 西安理工大学 Non-invasive power load identification method
CN111612074A (en) * 2020-05-22 2020-09-01 王彬 Identification method and device of non-invasive load monitoring electric equipment and related equipment
CN113780440A (en) * 2021-09-15 2021-12-10 江苏方天电力技术有限公司 Low-voltage station area phase identification method for improving data disturbance resistance
CN114611869A (en) * 2022-01-24 2022-06-10 国家电网有限公司 Low-voltage transformer area subscriber identification method

Also Published As

Publication number Publication date
CN115051363A (en) 2022-09-13

Similar Documents

Publication Publication Date Title
CN112149873B (en) Low-voltage station line loss reasonable interval prediction method based on deep learning
CN113126019A (en) Intelligent ammeter error remote estimation method, system, terminal and storage medium
CN111368259B (en) Distribution network voltage data missing filling method and related device
CN114519514B (en) Low-voltage transformer area reasonable line loss value measuring and calculating method, system and computer equipment
CN112819649B (en) Method and device for determining user change relation of station area
CN115051363B (en) Distribution network area user change relation identification method and device and computer storage medium
CN112308341B (en) Power data processing method and device
CN115618249A (en) Low-voltage power distribution station area phase identification method based on LargeVis dimension reduction and DBSCAN clustering
CN112418476A (en) Ultra-short-term power load prediction method
CN112232381B (en) Model parameter post-processing method and device for dominant load parameter noise identification
CN112380998A (en) Low-voltage transformer area missing data completion method based on matrix completion
CN113595071A (en) Transformer area user identification and voltage influence evaluation method
CN112163781A (en) Park electricity utilization group life cycle evaluation method based on multi-dimensional index clustering
CN111259965A (en) Method and system for carrying out mean value clustering on electrical characteristic data based on dimension reduction
CN110703038B (en) Harmonic impedance estimation method suitable for fan access power distribution network
Ao et al. Entropy estimation via normalizing flow
CN117040114A (en) Method, medium and terminal for identifying topology of platform area
CN114839586B (en) Low-voltage station metering device misalignment calculation method based on EM algorithm
CN116029614A (en) Power quality assessment method, device and computer equipment for power distribution network area
CN112989965B (en) SVD-LCD-BCM-based fault feature extraction method
CN113589034A (en) Electricity stealing detection method, device, equipment and medium for power distribution system
CN112699920A (en) Method and system for determining main influence factors of line loss rate of passive station area
CN111062538A (en) CEEMD method for determining reasonable maximum value of cellular load in space load prediction
CN113723835B (en) Water consumption evaluation method and terminal equipment for thermal power plant
Navon et al. Capturing between-tasks covariance and similarities using multivariate linear mixed models

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