A method, system, device and medium for checking the topological relationship of a data-missing station area

Abstract

The application discloses a kind of topological relation checking method, system, computer device and medium for data missing station area, the method includes: obtaining the voltage data and electric quantity data of each distribution transformer and each user of low-voltage station area;Using matrix decomposition method to the voltage data, electric quantity data of each user is completed;According to the voltage data of each distribution transformer, using cosine similarity to group station area;According to the voltage data of each distribution transformer and each user, determine the group attribution of each user based on station area grouping;According to the electric quantity data of distribution transformer and user, the user in each group is solved user transformer relationship by quadratic programming, obtains the file of each station area.The application according to the voltage data and table code data of low-voltage station area distribution transformer and low-voltage user, mines its correlation, judges the station area attribution of low-voltage user by mixed integer quadratic programming algorithm, to realize station area topological relation checking.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power systems, in particular to a topology relationship checking method and system for data-missing transformer areas, a computer device and a computer readable storage medium. BACKGROUND

[0002] The transformer area and user transformer relationship information in the operation and distribution archives is of great significance for line loss management. Accurate transformer area and user transformer relationship can help management personnel analyze line loss sources and quickly propose optimization strategies. With the rapid increase in the coverage rate of intelligent collection devices, massive power measurement data provides the possibility for data-driven low-voltage transformer area and user transformer relationship checking. However, due to various conditions, the actual measurement data has different degrees of data missing. The current main topology relationship checking method is carrier communication, which directly regards the power line as a channel for communication work of the power communication line, and checks the existence of the carrier to determine the topology relationship of the low-voltage transformer area by using a handheld carrier communication instrument. The carrier communication checking method needs to check each household with a handheld carrier communication instrument. SUMMARY

[0003] In order to solve the above problems of the prior art, in view of the widespread data missing problem, the present application provides a topology relationship checking method and system for data-missing transformer areas, a device and a computer readable storage medium. The method mines the correlation of the voltage data and the meter code data of the low-voltage transformer area distribution transformer and the low-voltage user, and determines the transformer area belonging of the low-voltage user by using a mixed integer quadratic programming algorithm, so as to realize the topology relationship checking of the transformer area.

[0004] The first object of the present application is to provide a topology relationship checking method for data-missing transformer areas.

[0005] The second object of the present application is to provide a topology relationship checking system for data-missing transformer areas.

[0006] The third object of the present application is to provide a computer device.

[0007] The fourth object of the present application is to provide a computer readable storage medium.

[0008] The first object of the present application can be achieved by adopting the following technical solution:

[0009] A topology relationship checking method for data-missing transformer areas, the method comprising:

[0010] Obtaining voltage data and power data of each distribution transformer and each user in a low-voltage transformer area;

[0011] Using a matrix decomposition method to complete the data of the voltage data and the power data of each user;

[0012] According to the voltage data of each distribution transformer, the distribution area grouping is performed by using cosine similarity;

[0013] According to the voltage data of each distribution transformer and each user, the group attribution of each user is determined based on the distribution area grouping;

[0014] According to the power data of the distribution transformer and the user, the user-transformer relationship of each group is solved by quadratic programming, and the archives of each distribution area are obtained.

[0015] Further, the voltage data and the power data of each user are respectively denoted as matrixes and wherein K represents K distribution areas in a low-voltage distribution area, and N represents N users under the K distribution areas;

[0016] The matrix decomposition method is used to complete the data of the voltage data and the power data of each user, that is, the matrix decomposition method is used to complete the data of the matrix Y, wherein the matrix Y is respectively the matrix and the matrix

[0017] The matrix Y is completed by using the matrix decomposition method, comprising:

[0018] The matrix Y is decomposed into a plurality of matrixes by matrix decomposition;

[0019] The decomposed matrixes are multiplied to obtain an approximate matrix Y' of the matrix Y;

[0020] The missing values of the matrix Y are filled by using the values of the corresponding positions of the approximate matrix Y'.

[0021] Further, the distribution area grouping is performed by using the cosine similarity according to the voltage data of each distribution transformer, comprising:

[0022] The cosine similarity formula is as follows:

[0023]

[0024] wherein A and B are two voltage data time series compared with each other, A z and B z represent the zth element of the voltage of each phase of two distribution areas, and η is the length of the time series;

[0025] The cosine similarity of the voltage data of each two distribution transformers is calculated according to the cosine similarity, and a matrix r i,j is obtained.

[0026] If r max ≥ 0.95 and r min≥0.85, the two distribution transformers belong to the same group, otherwise the two distribution transformers do not belong to the same group; wherein r max and r min respectively represent the maximum value and the minimum value in the matrix r i,j .

[0027] Further, the group belonging of each user is determined based on the grouping of the transformer stations according to the voltage data of each distribution transformer and each user, including:

[0028] The cosine similarity of the voltage data of each user and the distribution transformer is calculated, and the group to which the distribution transformer belongs corresponding to the maximum value is the group to which the user belongs; the user profiles {F1, F2,...F n} of each group and the corresponding user power data {X1, X2,...X n} are obtained according to the group information of the user, wherein n represents the number of groups.

[0029] Further, the user in each group is solved by quadratic programming to obtain the profile of each transformer station according to the power data of the distribution transformer and the user, including:

[0030] The power supply and user power data of the transformer stations in the group i are subjected to quadratic programming.

[0031] The mixed integer quadratic programming model in the quadratic programming is solved by using a heuristic algorithm, and the calculation result is output when the error of the optimization target is less than a set threshold value, that is, the profile of each transformer station.

[0032] Further, the quadratic programming includes:

[0033] The power supply of the transformer station is expressed as the sum of the power consumption of all users under the transformer station plus the line loss power, and the relationship between the power consumption and the power supply in a period of time is expressed as:

[0034] ∑TP=∑UP+λ

[0035] Wherein, UP represents the power consumption of the user, TP represents the power supply of the transformer station, and λ represents the total line loss power of the transformer station.

[0036] Suppose that there are m transformer stations and N i users in a group i, then for any single transformer station:

[0037]

[0038] Wherein, TPj represents the power supply of the transformer station j in the group i; 1 represents belonging, and 0 represents not belonging.

[0039] Let Then I represents a unit vector;

[0040] Based on the linear programming model, for any two substation constructs mixed integer quadratic programming model, including:

[0041] The optimization goal is:

[0042]

[0043] The constraint term is:

[0044]

[0045]

[0046] In the formula, is the distribution of power, represent all users of any two substation power consumption data, each column represents a user's power consumption; represent the profile of substation 1, the element is 0, which means that the user does not belong to this substation, 1 indicates that it belongs to; is a unit vector, represent the profile of substation 2;

[0047] Solving the mixed integer quadratic programming model with heuristic algorithm, when the error of the optimization goal is less than the set threshold, output the calculation result, that is, the profile of each substation.

[0048] Further, the solving of the mixed integer quadratic programming model with heuristic algorithm comprises:

[0049] Repeat the following process until it contains a single substation:

[0050] If the number of substation in the group m is greater than 2, the first m / 2 substation is merged into one substation, the remaining substation is merged into another substation, and the quadratic programming is carried out again to obtain the result And the result Is the output of the current number of divide and conquer;

[0051] Further, the acquisition of the voltage data and power data of each distribution transformer and each user of the low-voltage substation comprises:

[0052] Let the collection granularity of K substation data be x minutes, the collection time span be D days, and the length of complete data of each substation be T=D*24*60 / x;

[0053] The three-phase voltage matrix of each distribution transformer is denoted as TU l ∈R T×3 , and the power data sequence is denoted as TP l ∈RT×1 , l = 1, 2,..., K;

[0054] The single-phase voltage matrix and the power data matrix of the N users under the K subareas are denoted as and

[0055] The second object of the present application can be achieved by adopting the following technical solution:

[0056] A topology relationship checking system for data missing subareas, the system comprises:

[0057] A data acquisition module is configured to acquire voltage data and power data of each distribution transformer and each user in a low-voltage subarea.

[0058] A data completion module is configured to complete data of voltage data and power data of each user by using a matrix decomposition method.

[0059] A subarea grouping module is configured to group subareas by using cosine similarity according to voltage data of each distribution transformer.

[0060] A user group determination module is configured to determine group attribution of each user based on subarea grouping according to voltage data of each distribution transformer and each user.

[0061] A user transformer relationship solving module is configured to solve user transformer relationship of users in each group by using quadratic programming according to power data of distribution transformers and users, and obtain archives of each subarea.

[0062] The third object of the present application can be achieved by adopting the following technical solution:

[0063] A computer device comprises a processor and a memory for storing a program executable by the processor, and the processor implements the topology relationship checking method when executing the program stored in the memory.

[0064] The fourth object of the present application can be achieved by adopting the following technical solution:

[0065] A computer readable storage medium stores a program, and the program is executed by a processor to implement the topology relationship checking method.

[0066] The present application has the following beneficial effects compared with the prior art:

[0067] The application provides a method and system for checking the topological relationship of a data-missing power distribution area, a computer device and a computer readable storage medium. BRIEF DESCRIPTION OF DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0069] Figure 1 The flow chart of the method for checking the topological relationship of a data-missing power distribution area in Embodiment 1 of the present application.

[0070] Figure 2a and 2b are respectively the data before and after voltage completion of some users under a power distribution area in Embodiment 1 of the present application.

[0071] Figure 3 The schematic diagram of the power distribution voltage data after grouping the power distribution areas in Embodiment 1 of the present application.

[0072] Figure 4 The structural block diagram of the system for checking the topological relationship of a data-missing power distribution area in Embodiment 2 of the present application.

[0073] Figure 5 The structural block diagram of the computer device in Embodiment 3 of the present application. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort fall within the protection scope of the present application. It should be understood that the described specific embodiments are only used to explain the present application, and are not used to limit the present application.

[0075] Embodiment 1:

[0076] As Figure 1As shown, the embodiment provides a topology relationship checking method for data missing substations, according to voltage data and meter code data of low-voltage substation distribution transformers and users, correlation thereof is mined, and a mixed integer quadratic programming algorithm is used to determine the substation attribution of the users, thereby realizing substation topology relationship checking, and specifically comprising the following steps:

[0077] S101, voltage data and power data of each distribution transformer and each user in a low-voltage substation are acquired.

[0078] The data collection granularity of K substations is x minutes, that is, one data point is collected every x minutes, the collection time span is D days, and the complete data length of each substation is T=D*24*60 / x. The three-phase voltage matrix of each distribution transformer is denoted as TU i ∈R T×3 (i=1, 2,..., K), and the power data sequence is denoted as The single-phase voltage matrix and the power data matrix of N users under K substations are denoted as and

[0079] S102, the voltage data and the power data of each user are completed.

[0080] The single-phase voltage matrix and the power data matrix of the user are completed. Matrix completion is to complete the missing part of a matrix containing missing values. Matrix completion can decompose a matrix Y containing missing values into two (or more) matrices, and then the multiplication of these decomposed matrices can obtain the approximation Y' of the original matrix. Then, the missing part of the original matrix Y is filled with the value of the approximation matrix Y'.

[0081] Suppose that the observed n-dimensional time series data is organized in a matrix Y, Y∈R n×T , Y is approximated by the combination of a dimensionality characteristic matrix W∈R n×T and a time characteristic matrix X∈R n×T , thereby repairing the data, wherein Y≈WX. To solve W and X, the least squares and gradient descent methods are used to solve the following minimization problem, thereby approximating the matrix W and the matrix X.

[0082] The above matrix decomposition method only targets the non-zero elements in the original matrix in the calculation, and does not further standardize the missing elements in the original matrix. For power measurement data, the measurement data of different users often shows certain similarity, for example, the voltage fluctuation of users with close electrical distance is often similar, and the power consumption habits of users in the same residential community are often similar. For the user i with missing data, suppose that the position set of the missing data is Ω' t, the shortest distance user can be found by calculating the distance with other users through Euclidean distance, and it is judged whether the data of the user in the set Ω' t is complete, if not, the next shortest distance user is selected, and so on until the user meeting the condition is found, and the element of the user in the set Ω' t is y' i，t , the minimization objective function is:

[0083]

[0084] Where, Ω is the set of positions of non-zero elements in the original matrix, Ω' is the set of positions of zero elements in the original matrix; R ω (W) and R x (X) are the 2-norm of the residual matrix, R ω (W) and R x (X) are the regularization terms of W and X respectively, to prevent overfitting, ω i is the column vector of W, x t is the column vector of X.

[0085]

[0086]

[0087] Since data missing occurs mostly in user voltage matrix and power matrix, this embodiment takes user single-phase voltage matrix completion as an example for illustration.

[0088] The matrix UU is decomposed into two matrices W and X, for the user i with missing data, the set of positions of missing elements in the original matrix is Ω', and the set of positions of other elements in the original matrix is Ω, the shortest distance user can be found by calculating the distance with other users through Euclidean distance, and it is judged whether the data of the user in the set Ω' i,t is complete, if not, the next shortest distance user is selected, and so on until the user meeting the condition is found, and the element of the user in the set Ω' t is y' i，t .

[0089] W and X are iterated by gradient descent method to minimize the following objective function:

[0090]

[0091] Finally, the missing data in the original matrix UU is filled according to W and X:

[0092] Uu (i，t)∈Ω = WX (i，t)∈Ω .

[0093] WX (i,t)∈ΩThe index matrix W represents the range of i columns and t columns in X belonging to Ω.

[0094] The raw data of the voltage of some users under a certain substation provided by the embodiment is shown in Figure 2a After the data is completed by the matrix decomposition method, it is shown in Figure 2b .

[0095] S103, according to the voltage data of each distribution transformer, the cosine similarity is used for substation grouping, and then the group attribution of all users is determined, and the mixed integer quadratic programming is used to solve the users in each group, and the profile of each substation is obtained.

[0096] (1) According to the voltage data of each distribution transformer, the cosine similarity is used for substation grouping.

[0097] The cosine similarity is to measure the similarity between two vectors according to the cosine value of the included angle. The cosine value of 0 degree angle is 1, indicating that the two vectors completely coincide, while the cosine value of any other angle is not greater than 1; and its minimum value is -1. Two attribute vectors A, B are two-by-two compared voltage data time series, A z , B z represent the zth element of the voltage of two substations, respectively, and η is the length of the time series, and the cosine similarity is given by the dot product and the vector length, as follows:

[0098]

[0099] According to the above formula, the cosine similarity of the three-phase voltage matrix TU i of two-by-two distribution transformers is calculated, and the matrix r i,j ∈R 3×3 is obtained. If r max ≥ 0.95 and r min ≥ 0.85, it is considered that the two transformers belong to the same group. According to the cosine similarity of the voltage, the substation grouping G = {g1, g1,... g n} is obtained, where n represents the number of groups, and j represents the substation number in the grouping i.

[0100] In this embodiment, the voltage data of 10 substations is selected, the total number of users is 2105, the data length for analysis is 3 days, the data acquisition granularity is 15 minutes, and the total number of points is 288. The substations are grouped by cosine similarity, and the grouping results are shown in Figure 3 and Table 1.

[0101] Table 1 Substation grouping results

[0102] First group Second group Third group Fourth group Fifth group Station area 1 Station area 3 Station area 4 Station area 6 Station area 9 Station area 2 Station area 5 - Station area 7 - Station area 10 - - Station area 8 -

[0103] (2) According to the voltage data of each distribution transformer and each user, the group affiliation of all users is determined based on the grouping of the transformer area.

[0104] The cosine similarity of the voltage matrix of each user and the distribution transformer is calculated and the maximum value is taken, and the group to which the distribution transformer belongs is the group to which the user belongs, thereby obtaining the user profile {F1, F2,...F n} of each group. n

[0105] According to the voltage data of the 10 transformer areas, the total number of users is 2105, and the grouping results of the users are shown in Table 2.

[0106] Table 2 Grouping results of users

[0107] First group Second group Third group Fourth group Fifth group Real number of users 721 422 203 574 185 Algorithm grouping result 673 412 0 560 185 Number of correctly grouped users 673 412 0 560 185 Accuracy 100％ 100％ 100％ 100％ 100％

[0108] The total number of users in the 10 transformer areas is 2105, and after excluding users with more than 60% missing data, the total number of users is 1830. The third group, i.e. the users in the fourth transformer area, have no voltage data, so no user is assigned to this group. Taking the first group as an example, the first group includes transformer areas 1, 2 and 10, and contains 721 users. The algorithm assigns 673 of the 2105 users to this group, and of these 673 users, the number of users that actually belong to this group (belonging to transformer areas 1, 2 or 10) is 673, i.e. all the grouping is correct, with an accuracy of 100%.

[0109] (3) The user in each group is solved by quadratic programming to obtain the profile of each transformer area.

[0110] According to the results of the transformer area grouping, the power of the transformer area is checked. The power supply of the transformer area can be represented as the sum of the power consumption of all users under the transformer area plus the line loss power. The relationship between the power consumption and the power supply in a period of time can be represented by the following formula:

[0111] ∑TP=∑UP+λ

[0112] Where UP represents the power consumption of the user, TP represents the power supply of the transformer area, and λ represents the total line loss power of the transformer area.

[0113] Suppose there are m transformer areas and N i users in a group, then for any single transformer area j:

[0114]

[0115] Where represents whether user u in group i belongs to transformer area j, 1 means belongs to, and 0 means does not belong to; ​N represents the power supply of the cell j in the group i, N i N represents the total number of users under the group i.

[0116] Let Then I represents a unit vector,

[0117] Quadratic programming includes: linear programming (LP) model is to study the mathematical theory and method of linear objective function under linear constraint condition extreme value problem, and for mixed integer quadratic programming, the objective function is quadratic and not linear, variable limit is integer, because the variable here indicates whether the corresponding user belongs to a certain cell and can only belong to or not belong to, so the integer variable here is 0 or 1, and the constraint condition limits the power consumption of the cell to be less than the power supply of the distribution transformer. In order to improve the efficiency of the algorithm, set the advanced constraint condition of quadratic programming, and constrain the user's cell attribution to only belong to a few cells, rather than belong to any cell. Calculate the cosine similarity of user voltage and three-phase voltage of distribution transformer to get matrix r i,j ∈R 3×3 If r max ≥ 0.95 and r min ≥ 0.85, the user is likely to belong to the cell, otherwise the user is not likely to belong to the cell, which includes:

[0118] Based on the linear programming model, for any two cells, construct a mixed integer quadratic programming model:

[0119] Optimization goal:

[0120]

[0121] Constraint term:

[0122]

[0123]

[0124] In the formula, is the power supply of the distribution transformer, represents the power consumption matrix of all users under the two cells, and each column represents the power consumption of a user; represents the profile of cell 1, and the element 0 represents that the user does not belong to the cell, and 1 represents that the user belongs to the cell; is a unit vector, represents the profile of cell 2.

[0125] The meaning of the constraint term is that the power supply is greater than the power consumption.

[0126] If the number of the station areas in the group m is greater than 2, the first m / 2 station areas are merged into one station area, and the rest of the station areas are merged into another station area, and then the second-order programming is performed to obtain the result and the result i.e. the result and the result represents the output of the first-time division and merger;

[0127] If the output of the first-time division and merger contains more than 2 station areas, the second-order programming is repeatedly performed until the output contains a single station area;

[0128] The mixed integer second-order programming model is solved by using a heuristic algorithm, and the calculation result is output when the error of the optimization target is less than a threshold value, as shown in Table 3. The heuristic algorithm includes genetic algorithm, ant colony algorithm, etc.

[0129] Table 3 outputs the number of archives of each station area

[0130]

[0131] From the calculation result, it can be seen that the correct archives of the station area 1 contain 211 users, and the algorithm correctly finds 197 of them, and the electricity consumption of the 197 users accounts for 96% of the electricity consumption of the station area. The correct archives of the station area 2 contain 197 users, and the algorithm correctly finds 167 of them, and the accuracy is relatively low, but the electricity consumption of the 167 users accounts for 99% of the total electricity consumption of the station area, which shows that most of the misclassified users are some users with very small electricity consumption in the station area. For users with very small or even zero electricity consumption, it is difficult to determine their station area from the algorithm principle, and these users also have little impact on the line loss of the station area.

[0132] Those skilled in the art can understand that all or part of the steps in the method of implementing the above embodiments can be instructed by a program to relevant hardware, and the corresponding program can be stored in a computer readable storage medium.

[0133] It should be noted that although the method operations of the above embodiments are described in a specific order in the accompanying drawings, this does not require or imply that the operations must be performed in this specific order, or that all of the shown operations must be performed to achieve the desired result. On the contrary, the depicted steps can change the order of execution. Additionally or alternatively, certain steps can be omitted, combined into one step, and / or divided into multiple steps.

[0134] Embodiment 2:

[0135] As Figure 4As shown, the embodiment provides a topology relationship checking system for data missing substations, which comprises a data acquisition module 401, a data completion module 402, a substation grouping module 403, a user group determination module 404 and a user-substation relationship solving module 405, wherein:

[0136] The data acquisition module 401 is configured to acquire voltage data and power data of each distribution transformer and each user in a low-voltage substation.

[0137] The data completion module 402 is configured to complete data of voltage data and power data of each user by using a matrix decomposition method.

[0138] The substation grouping module 403 is configured to group substations by using cosine similarity according to voltage data of each distribution transformer.

[0139] The user group determination module 404 is configured to determine group attribution of each user based on substation grouping according to voltage data of each distribution transformer and each user.

[0140] The user-substation relationship solving module 405 is configured to solve user-substation relationship of each user in each group by using quadratic programming according to power data of each distribution transformer and each user, and obtain archives of each substation.

[0141] The specific implementation of each module in the embodiment can refer to the above-mentioned embodiment 1, which will not be repeated here; it should be noted that the system provided in the embodiment is only exemplified by the division of the above-mentioned functional modules, and in actual application, the above-mentioned functions can be completed by different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the functions described above.

[0142] Embodiment 3

[0143] The embodiment provides a computer device, which can be a computer, such as Figure 5 As shown, the processor 502, the memory, the input device 503, the display 504 and the network interface 505 connected by the system bus 501, the processor is used to provide computing and control ability, the memory includes a non-volatile storage medium 506 and an internal memory 507, the non-volatile storage medium 506 stores an operating system, a computer program and a database, the internal memory 507 provides an environment for the running of the operating system and the computer program in the non-volatile storage medium, when the processor 502 executes the computer program stored in the memory, the topology relationship checking method of the above-mentioned embodiment 1 is realized, as follows:

[0144] Acquire voltage data and power data of each distribution transformer and each user in a low-voltage substation.

[0145] The voltage data and the power data of each user are complemented by using a matrix decomposition method;

[0146] According to the voltage data of each distribution transformer, the distribution areas are grouped by using cosine similarity;

[0147] According to the voltage data of each distribution transformer and each user, the group attribution of each user is determined based on the distribution area grouping;

[0148] According to the power data of the distribution transformer and the user, the user-transformer relationship of each user in each group is solved by using a quadratic programming method, and the archives of each distribution area are obtained.

[0149] Embodiment 4:

[0150] The embodiment provides a computer readable storage medium storing a computer program, and when the computer program is executed by a processor, a topology relationship checking method in the above embodiment 1 is realized, as follows:

[0151] The voltage data and the power data of each distribution transformer and each user in a low-voltage distribution area are acquired;

[0152] The voltage data and the power data of each user are complemented by using a matrix decomposition method;

[0153] According to the voltage data of each distribution transformer, the distribution areas are grouped by using cosine similarity;

[0154] According to the voltage data of each distribution transformer and each user, the group attribution of each user is determined based on the distribution area grouping;

[0155] According to the power data of the distribution transformer and the user, the user-transformer relationship of each user in each group is solved by using a quadratic programming method, and the archives of each distribution area are obtained.

[0156] It should be noted that the computer readable storage medium of the embodiment can be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection with one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0157] The above merely describes preferred embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can make equivalent replacements or changes to the technical scheme and inventive concept of the present application within the scope disclosed by the present application, and such replacements or changes shall also fall within the protection scope of the present application.

Claims

1. A method for checking topological relations of data missing cell areas, characterized in that, The method comprises: acquiring voltage data and power data of each distribution transformer and each user in a low-voltage transformer area; completing data of the voltage data and the power data of each user by using a matrix decomposition method; grouping the transformer areas according to the voltage data of each distribution transformer by using cosine similarity; determining the group attribution of each user based on the grouping of the transformer areas according to the voltage data of each distribution transformer and each user; solving the user-transformer relationship of the users in each group by quadratic programming to obtain the archives of each transformer area according to the power data of the distribution transformers and the users; wherein the acquiring of the voltage data and the power data of each distribution transformer and each user in a low-voltage transformer area comprises: assuming that the data collection granularity of K transformer area data is x minutes, the collection time span is D days, and the complete data length of each transformer area is T = D * 24 * 60 / x; The three-phase voltage matrix of each distribution transformer is denoted by TU l ∈ R T×3 The power data sequence is denoted by TP l ∈ R T×1 l = 1, 2,... K; Let the single-phase voltage matrix and the power data matrix of N users under K districts be denoted as UU∈R T×N and UP∈R T×N ; the completing of the data of the voltage data and the power data of each user by using the matrix decomposition method comprises: decomposing the matrix Y into a plurality of matrices by matrix decomposition; the matrix Y is respectively a matrix UU∈R T×N and a matrix UP∈R T×N ; multiplying the decomposed matrices to obtain an approximate matrix Y of the matrix ′ ; filling in missing values of the matrix Y with values from a corresponding position of the matrix Y ′ filling in missing values of the matrix Y with values from a corresponding position of the matrix Y the grouping of the transformer areas according to the voltage data of each distribution transformer by using the cosine similarity comprises: the cosine similarity formula is as follows: Wherein, A, B are two voltage data time series in contrast, A z , B z respectively represent the zth element of the voltage of a certain phase of two transformer areas, and η is the length of the time series. According to the cosine similarity, the cosine similarity of the voltage data of each pair of distribution transformers is calculated to obtain a matrix r i,j ; If r max ≥ 0.95 and r min ≥ 0.85, then the two distribution transformers belong to the same group, otherwise the two distribution transformers do not belong to the same group; wherein r max and r min represent the maximum and minimum values in the matrix r i,j , respectively.

2. The method of claim 1, wherein the determining of the group attribution of each user based on the grouping of the transformer areas according to the voltage data of each distribution transformer and each user comprises: The cosine similarity of the voltage data of each user and the distribution transformer is calculated, and the group in which the distribution transformer corresponding to the maximum value is located is the group to which the user belongs; the user profiles of each group are obtained according to the group information of the user, that is, {F1, F2, … F n}, and the corresponding user power data is {X1, X2, … X n}, wherein n represents the number of groups.

3. The method of claim 2, wherein: the solving of the user-transformer relationship of the users in each group by quadratic programming to obtain the archives of each transformer area according to the power data of the distribution transformers and the users comprises: performing quadratic programming according to the power supply data and the user power data of the transformer areas in the group i; solving the mixed integer quadratic programming model in the quadratic programming by using a heuristic algorithm, and outputting the calculation result when the error of the optimization target is less than a set threshold value, that is, the archives of each transformer area.

4. The method of claim 3, wherein the performing of the quadratic programming comprises: the power supply of the transformer area is expressed as the sum of the power consumption of all users under the transformer area and the line loss power, and the relationship between the power consumption and the power supply in a period of time is expressed as: ∑TP=∑UP+λ wherein UP represents the power consumption of the user, TP represents the power supply of the transformer area, and λ represents the total line loss power of the transformer area; Suppose that a certain group i contains m stations and N stations. i For any single station area j, with a user: where TP i j represents the power supply of the substation j in the group i; represents whether the user u in the group i belongs to the substation j, 1 represents belonging, and 0 represents not belonging. Let Then I denotes a unit vector; based on the linear programming model, a mixed integer quadratic programming model is constructed for any two transformer areas, comprising: the optimization target is: the constraint term is: In the formula, is the distribution power, represents the power consumption data of all users under any two power distribution areas, and each column represents the power consumption of a user; represents the profile of power distribution area 1, and the element 0 represents that the user does not belong to the power distribution area, and 1 indicates that it belongs to; is a unit vector, represents the profile of power distribution area 2; the mixed integer quadratic programming model is solved by using a heuristic algorithm, and the calculation result is output when the error of the optimization target is less than a set threshold value, that is, the archives of each transformer area.

5. The method of claim 4, wherein, the solving of the mixed integer quadratic programming model by using the heuristic algorithm comprises: If the number of station areas m in the group is greater than 2, the first m / 2 station areas are merged into one station area, and the rest of the station areas are merged into another station area, and then quadratic programming is performed to obtain a result and The result is the output of the current time division and conquer. if the output of the divide-and-conquer contains more than 2 transformer areas, return if the number m of transformer areas in the group is greater than 2, and continue to execute the subsequent operations until the output contains a single transformer area.

6. A system for checking topological relations for data missing cell areas, characterized in that, The system comprises: a data acquisition module configured to acquire voltage data and power data of each distribution transformer and each user in a low-voltage transformer area; a data completion module configured to complete data of the voltage data and the power data of each user by using a matrix decomposition method; a transformer area grouping module configured to group the transformer areas according to the voltage data of each distribution transformer by using cosine similarity; a user group determination module configured to determine the group attribution of each user based on the grouping of the transformer areas according to the voltage data of each distribution transformer and each user; a user group determination module configured to determine the group attribution of each user based on the grouping of the transformer areas according to the voltage data of each distribution transformer and each user; The household transformer relationship solving module is configured to solve the household transformer relationship of each group of users by quadratic programming according to the voltage data and the power data of the distribution transformer and the users, and obtain the archives of each transformer area. The method comprises the following steps: Suppose that the collection granularity of K transformer area data is x minutes, the collection time span is D days, and the complete data length of each transformer area is T=D*24*60 / x; The three-phase voltage matrix of each distribution transformer is denoted by TU l ∈R T×3 The power data sequence is denoted by TP l ∈R T×1 l = 1, 2,... K; Let the single-phase voltage matrix and the power data matrix of N users under K districts be denoted as UU∈R T×N and UP∈R T×N ; The method for data completion of the voltage data and the power data of each user by using the matrix decomposition method comprises the following steps: decomposing the matrix Y into a plurality of matrices by matrix decomposition; the matrix Y is respectively a matrix UU∈R T×N and a matrix UP∈R T×N ; multiplying the decomposed matrices to obtain an approximate matrix Y of the matrix ′ ; filling in missing values of the matrix Y with values from a corresponding position of the matrix Y ′ filling in missing values of the matrix Y with values from a corresponding position of the matrix Y The method for grouping transformer areas by using the cosine similarity according to the voltage data of each distribution transformer comprises the following steps: The cosine similarity formula is as follows: Wherein, A, B are two voltage data time series in contrast, A z , B z respectively represent the zth element of the voltage of a certain phase of two transformer areas, and η is the length of the time series. According to the cosine similarity, the cosine similarity of the voltage data of each pair of distribution transformers is calculated to obtain a matrix r i,j ; If r max ≥ 0.95 and r min ≥ 0.85, then the two distribution transformers belong to the same group, otherwise the two distribution transformers do not belong to the same group; wherein r max and r min represent the maximum and minimum values in the matrix r i,j , respectively.

7. A computer-readable storage medium storing a program, characterized in that, The program is executed by the processor to implement the topological relationship checking method of any one of claims 1-5.

Images