A data modeling method for low voltage distribution network equipment based on non-negative matrix factorization

Abstract

The present invention discloses a method for modeling low-voltage distribution network equipment data by joint non-negative matrix decomposition, including storing and standardizing the type, operating environment, asset identification, and electrical characteristic data information of the primary equipment of distribution transformers, low-voltage junction boxes, and user meter boxes in the form of a matrix; using a joint non-negative matrix decomposition method to project all data sets into the same space, taking vectors with larger coefficients in the same basis vector in the space as heterogeneous vectors; forming a modal space for the heterogeneous vectors to characterize the key information of the equipment; and completing the rapid identification of distribution network equipment information and modeling of low-voltage distribution network equipment data information based on heterogeneous vector information. The present invention adopts a non-negative matrix decomposition method to achieve dimensionality reduction of low-voltage distribution network equipment data information, filter out information irrelevant to the equipment, and help save information processing resources and improve computing efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of data information modeling, and in particular to a low-voltage distribution network equipment data modeling method combined with non-negative matrix decomposition. Background Art

[0002] With the increasing demand for national economic development and the advancement of distribution network technology, the scale of my country's low-voltage distribution network is showing a trend of continuous expansion, and the safety and stability of low-voltage distribution network power supply equipment is the basis for the safe and reliable operation of the power grid. Once a low-voltage distribution network power supply equipment fails, it will affect people's normal production and life at the least; it will cause a safety accident, resulting in major economic losses and adverse social impacts. Therefore, real-time monitoring of the operating status of low-voltage distribution network power supply equipment, early warning of potential failures of power supply equipment, and timely maintenance and replacement of damaged equipment are of great significance to improving the reliable operation of the power grid. However, there are many low-voltage distribution network equipment, including distribution transformers, low-voltage distribution boxes, user meter boxes and other primary equipment, and each equipment contains a variety of information such as type, operating environment, asset identification, electrical characteristics, etc. How to model the above massive information is the key to accurately locate and identify equipment, and it is also the core of implementing online monitoring of equipment status. Summary of the invention

[0003] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a low-voltage distribution network equipment data modeling method combined with non-negative matrix decomposition, which can solve the problem that the prior art cannot perform data modeling on the above-mentioned massive information to achieve accurate positioning and identification of equipment.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: including, storing and standardizing the type, operating environment, asset identification, and electrical characteristic data information of the primary equipment of the distribution transformer, low-voltage tapping box, and user meter box in the form of a matrix; using a joint non-negative matrix decomposition method to project all data sets into the same space, and taking the vectors with larger coefficients in the same basis vector in the space as heterogeneous vectors; characterizing the key information of the equipment in the modal space of the heterogeneous vectors; and completing the rapid identification of distribution network equipment information and modeling of low-voltage distribution network equipment data information based on the heterogeneous vector information.

[0007] As a preferred scheme of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition described in the present invention, it includes: encoding the type of the distribution transformer, the low-voltage junction box, the user meter box primary equipment, and the asset identification information; merging with the operating environment of the equipment and the electrical characteristic parameter information, converting them into a matrix form for storage; standardizing and normalizing all data information to complete data preprocessing.

[0008] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition described in the present invention, it includes: using the non-negative matrix decomposition method to complete the dimensionality reduction of the low-voltage distribution network equipment data information and filter out information irrelevant to the equipment.

[0009] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition described in the present invention, the distribution network equipment data information X is a non-negative matrix, and the matrix decomposition is performed according to the following formula:

[0010]

[0011] stW≥0,H≥0

[0012] Among them, ||·|| F represents the Frobenius norm, W∈R m×r The column vectors are basis vectors, H∈R r×n The row vectors are coefficients of each variable in X on the basis vectors in W, r is the number of basis vectors, and in most cases r<min(m,n).

[0013] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition described in the present invention, it includes: for the column vector xj, approximating it with a non-negative linear combination under the basis vector, as follows,

[0014]

[0015] Among them, the r basis vectors wi (i=1, 2, ..., r) can be regarded as the skeleton of the data set X, and the r-dimensional coefficient vector hj describes the weight of the variable xj on the skeleton.

[0016] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition described in the present invention, it includes: completing the dimensionality reduction of the distribution network equipment data information based on the distribution network equipment data information X and the column vector xj, and filtering out information irrelevant to the equipment.

[0017] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition of the present invention, it includes: the distribution network equipment data information is projected into the same low-dimensional space after the joint non-negative matrix decomposition, so that they share the same set of basis vectors, as follows,

[0018]

[0019] stW≥0,H i ≥0, i=1, 2, 3

[0020] Among them, X i Represents the data set of each device in the distribution network, W∈R m×r The r column vectors in H form the basis of the same space shared by n data sets. i Represents the coefficient of the data variable under the basis.

[0021] As a preferred solution of the low-voltage distribution network equipment data modeling method of the joint non-negative matrix decomposition of the present invention, it includes: the matrix H i The coefficient vectors in are used to select variables from each data set to construct the modal space, and the matrix H is calculated using the following formula i The z-score of the row vector element,

[0022]

[0023] Among them, μ i is the matrix H i The mean of the i-th row, σ i is the standard deviation of the i-th row, for H i , z ij >T indicates that the jth variable in the data set X is a member variable of the modal space, and T>0 is a pre-given threshold.

[0024] The beneficial effects of the present invention are as follows: the present invention encodes the type, asset identification and other information of primary equipment such as distribution network transformers, low-voltage distribution boxes, and user meter boxes, establishes database information of the equipment, and combines it with the operating environment information and electrical characteristic information of the equipment to complete the fusion of relevant information of the equipment; a non-negative matrix decomposition method is used to achieve dimensionality reduction of data information of low-voltage distribution network equipment, and filter out information irrelevant to the equipment, which is helpful to save information processing resources and improve computing efficiency; a joint non-negative matrix decomposition method is used to project all data information of distribution network equipment into the same space, and heterogeneous vectors in the projection space can be extracted based on the comparison of coefficient vectors. The heterogeneous vectors represent the most critical vectors of the distribution network equipment. In practice, detailed information of the equipment can be extracted based on the identification of heterogeneous vectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0026] Figure 1 A schematic flow chart of a low-voltage distribution network equipment data modeling method using a joint non-negative matrix decomposition method according to a first embodiment of the present invention;

[0027] Figure 2 It is a schematic diagram of comparison curves of the low-voltage distribution network equipment data modeling method based on joint non-negative matrix decomposition described in the first embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0029] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0030] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0031] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0032] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0033] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0034] Example 1

[0035] Reference Figure 1 , which is the first embodiment of the present invention, provides a low-voltage distribution network equipment data modeling method based on joint non-negative matrix decomposition, comprising:

[0036] S1: The type, operating environment, asset identification, and electrical characteristic data information of the primary equipment of distribution transformers, low-voltage junction boxes, and user meter boxes are stored and standardized in the form of a matrix.

[0037] S2: All data sets are projected into the same space using the joint non-negative matrix factorization method, and vectors with larger coefficients in the same basis vector in the space are considered heterogeneous vectors.

[0038] S3: Heterogeneous vectors form the key information of the modal space characterization device.

[0039] S4: Complete the rapid identification of distribution network equipment information and modeling of low-voltage distribution network equipment data information based on heterogeneous vector information.

[0040] Specifically, it should be noted that the type and asset identification information of the primary equipment of the distribution transformer, low-voltage junction box, and user meter box are encoded; merged with the operating environment and electrical characteristic parameter information of the equipment, converted into a matrix form for storage; all data information is standardized and normalized to complete data preprocessing.

[0041] Furthermore, the non-negative matrix decomposition method is used to complete the dimensionality reduction of the low-voltage distribution network equipment data information, and filter out the information irrelevant to the equipment. The distribution network equipment data information X is a non-negative matrix, and the matrix decomposition is performed according to the following formula:

[0042]

[0043] stW≥0,H≥0

[0044] Among them, ||·|| F represents the Frobenius norm, W∈R m×r The column vectors are basis vectors, H∈R r×n The row vectors are coefficients of each variable in X on the basis vectors in W, r is the number of basis vectors, and in most cases r<min(m,n).

[0045] For the column vector xj, it is approximated by a non-negative linear combination under the basis vectors as follows,

[0046]

[0047] Among them, the r basis vectors wi (i=1, 2, ..., r) can be regarded as the skeleton of the data set X, and the r-dimensional coefficient vector hj describes the weight of the variable xj on the skeleton.

[0048] Based on the distribution network equipment data information X and column vector xj, the dimensionality reduction of the distribution network equipment data information is completed, and the information irrelevant to the equipment is filtered out. The distribution network equipment data information is projected into the same low-dimensional space after joint non-negative matrix decomposition, so that it shares the same set of basis vectors, as follows:

[0049]

[0050] stW≥0,H i ≥0, i=1, 2, 3

[0051] Among them, X i Represents the data set of each device in the distribution network, W∈R m×r The r column vectors in H form the basis of the same space shared by n data sets. i Represents the coefficient of the data variable under the basis.

[0052] Matrix H i The coefficient vectors in are used to select variables from each data set to construct the modal space, and the matrix H is calculated using the following formula i The z-score of the row vector element,

[0053]

[0054] Among them, μ i is the matrix H i The mean of the i-th row, σ i is the standard deviation of the i-th row, for H i , z ij>T indicates that the jth variable in the dataset X is a member variable of the modal space, and T>0 is a pre-given threshold.

[0055] Example 2

[0056] Reference Figure 2 , which is the second embodiment of the present invention. This embodiment is different from the first embodiment in that it provides an experimental test verification of a low-voltage distribution network equipment data modeling method combined with non-negative matrix decomposition, specifically including:

[0057] In order to better verify and illustrate the technical effects adopted in the method of the present invention, this embodiment selects a traditional data modeling method and the method of the present invention for comparative testing, compares the test results by means of scientific demonstration, and verifies the real effect of the method of the present invention.

[0058] Traditional data modeling methods are unable to model massive amounts of information and cannot accurately locate and identify equipment. In order to verify that the method of the present invention has higher computing efficiency and resource conservation than traditional methods, this embodiment will use traditional data modeling methods and the method of the present invention to conduct real-time testing and comparison on the data modeling of low-voltage distribution network equipment on the simulation platform.

[0059] Test environment: (1) Scrapy web crawler framework, Python, Http data packet, wireshark, service program;

[0060] (2) The simulation platform obtains the cloud database and finds 10,000 sets of historical data as the standard database test sample set;

[0061] (3) The machine learning algorithm of the traditional method and the joint non-negative matrix decomposition algorithm of the method of the present invention are respectively imported into the simulation software through MATLB coding for state simulation.

[0062] Reference Figure 2 The solid line is the curve output by the method of the present invention, and the dotted line is the curve output by the traditional method. Figure 2 It can be seen intuitively that the solid line and the dotted line show different trends with the increase of time. Compared with the dotted line, the solid line has been showing a stable upward trend in the early stage. Although it has declined in the later stage, the fluctuation is not large and it has been above the dotted line and maintained a certain distance. The dotted line shows a large fluctuation trend and is unstable. Therefore, the calculation efficiency of the solid line is always greater than that of the dotted line, which verifies the real effect of the method of the present invention.

[0063] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A low-voltage distribution network equipment data modeling method combined with non-negative matrix decomposition, characterized in that: include, The type, operating environment, asset identification, and electrical characteristic data of the primary equipment of distribution transformers, low-voltage junction boxes, and user meter boxes are stored and standardized in the form of a matrix; All data sets are projected into the same space using a joint non-negative matrix factorization method, and vectors with larger coefficients in the same basis vector in the space are considered heterogeneous vectors; The heterogeneous vectors constitute key information of the modal space characterization device; Based on heterogeneous vector information, the rapid identification of distribution network equipment information and the modeling of low-voltage distribution network equipment data information are completed; Encode the types of the distribution transformer, the low-voltage tap box, the user meter box primary equipment, and the asset identification information; The information is combined with the operating environment and electrical characteristic parameter information of the device, and converted into a matrix form for storage; Standardize and normalize all data information to complete data preprocessing; The non-negative matrix decomposition method is used to reduce the dimension of low-voltage distribution network equipment data information and filter out information irrelevant to the equipment; Based on the distribution network equipment data information X and the column vector Xj, the dimension of the distribution network equipment data information is reduced, and information irrelevant to the equipment is filtered out; The distribution network equipment data information is projected into the same low-dimensional space after joint non-negative matrix decomposition, so that they share the same set of basis vectors, as follows: s.t.w≥0,H i ≥0,i=1,2,3 Among them, X i Represents the data set of each device in the distribution network, w∈R m×r The r column vectors in H form the basis of the same space shared by n data sets. i Represents the coefficient of the data variable under the basis; Matrix H i The coefficient vectors in are used to select variables from each data set to construct the modal space, and the matrix H is calculated using the following formula i The z-score of the row vector element, Among them, μ i is the matrix H i The mean of the i-th row, σ i is the standard deviation of the i-th row, for H i , z ij >T indicates that the jth variable in the distribution network equipment data information X is a member variable of the modal space, and T>0 is a pre- A given threshold.

2. The low-voltage distribution network equipment data modeling method based on joint non-negative matrix decomposition according to claim 1 is characterized in that: include, The data information X of the distribution network equipment is a non-negative matrix, and the matrix decomposition is performed according to the following formula: stW≥0,H≥0 in,‖〃‖ F represents the Frobenius norm, W∈R m×r The column vectors are basis vectors, H∈R r×n The row vectors in the graph are the coefficients of each variable in X on the basis vectors in W, r is the number of basis vectors, and r <min(m,n)。 3. The low-voltage distribution network equipment data modeling method based on joint non-negative matrix decomposition according to claim 2 is characterized in that: include, For the column vector xj, it is approximated by a non-negative linear combination under the basis vectors as follows, Among them, the r basis vectors wi (i = 1, 2, ..., r) can be regarded as the skeleton of the distribution network equipment data information X, and the r-dimensional coefficient vector hj describes the weight of the variable xj on the skeleton.

Images