Phase-to-household relationship identification method, device and electronic equipment for low-voltage distribution network

Abstract

The present application discloses a method, device and electronic device for identifying the relationship between phases and households in a low-voltage distribution network. It relates to the field of low-voltage distribution network topology identification. The method includes: collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line; determining the initial weight of the node current, calculating the initial regression coefficient and the initial adjustment residual value between the node current and the phase line current, updating the initial weight, calculating the regression coefficient and the adjustment residual value based on the updated weight, iterating the calculation multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold, and determining the regression coefficient of the last iteration as the target regression coefficient; determining the target regression coefficient between the target node and each phase line, determining the phase line corresponding to the maximum target regression coefficient as the target phase line to which the target node belongs, and obtaining the user set of each phase line. Through the present application, the problem of low accuracy in identifying the relationship between phases and households in low-voltage distribution networks in the related art is solved.

Description

Technical Field

[0001] The present application relates to the field of low-voltage distribution network topology identification, and more specifically, to a method, device and electronic equipment for identifying phase-to-household relationships in a low-voltage distribution network. Background Art

[0002] With the massive access of new energy sources, the randomness and uncertainty of new energy load fluctuations and the dispersion of load distribution make the topological structure of low-voltage distribution networks complex and difficult to identify, which affects the management and planning of low-voltage distribution networks and makes it difficult to carry out applications such as fault location, line loss management, and power theft detection. In the related art, the phase-to-household relationship identification methods of low-voltage distribution networks mainly include: model-driven methods, data-driven methods, and graph analysis. The model-driven method needs to combine real-time and historical data to estimate the network state through mathematical models to achieve topological identification; the data-driven method uses historical data and algorithms (such as neural networks) for topological identification; graph analysis uses graph theory principles to analyze network structures to effectively identify topology, etc. However, the phase-to-household relationship identification methods in the related art rely on models and historical data for analysis, which has certain errors and leads to low recognition accuracy.

[0003] With regard to the problem of low accuracy in identifying phase-to-household relationships in low-voltage distribution networks in related technologies, no effective solution has been proposed so far. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device and electronic equipment for identifying phase-to-household relationships in a low-voltage distribution network, so as to solve the problem of low accuracy in identifying phase-to-household relationships in a low-voltage distribution network in related technologies.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a method for identifying the relationship between phases and users of a low-voltage distribution network is provided. The method comprises: collecting the node current of each node on the target low-voltage distribution network, and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; for each phase line, determining the initial weight of the node current of each node, calculating the initial regression coefficient and initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and weighted least squares method, updating the initial weight by a preset weight function and the initial adjusted residual value, and updating the initial weight based on the updated The regression coefficient and the adjusted residual value are calculated based on the weights after the calculation, and the calculation is iterated multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than the preset threshold value, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; the target regression coefficient between the node current of the target node and the phase line current of each phase line is determined, and the phase line corresponding to the maximum target regression coefficient is determined as the target phase line to which the target node belongs, and the phase lines of all nodes are combined according to the phase line type to obtain the user set of each phase line.

[0006] Optionally, the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network are collected, and a node current matrix is ​​generated based on the node currents of all nodes, and the phase line current sequence of the phase line is generated based on the phase line current of each phase line, including: for each node, collecting T node currents of the target node at preset time intervals to generate a node current sequence of the target node, and generating a node current matrix based on the node current sequences of N nodes, wherein T and N are both positive integers; for each phase line, collecting T phase line currents of the target phase line at preset time intervals to generate a phase line current sequence of the target phase line.

[0007] Optionally, calculating the initial regression coefficient and initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weights and weighted least squares method includes: determining the initial weight matrix based on the initial weights of all nodes, inputting the initial weight matrix, the node current matrix and the phase line current sequence of the target phase line into the regression coefficient calculation formula to obtain the initial regression coefficient; calculating the fitting residual value of each node through the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line; and calculating the initial adjusted residual value based on the fitting residual value of each node.

[0008] Optionally, calculating the fitting residual value of each node through the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line includes: calculating the product of the initial regression coefficient and the node current matrix to obtain the current prediction value sequence of each node; calculating the difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node to obtain the fitting residual value of each node.

[0009] Optionally, calculating the initial adjusted residual value based on the fitted residual value of each node includes: determining the adjustment constant parameter and the preset parameter, and calculating the standard deviation of the error term based on the fitted residual values ​​of all nodes; calculating the product of the adjustment constant parameter, the preset parameter and the standard deviation of the error term, and calculating the ratio of the fitted residual value of each node to the product to obtain the initial adjusted residual value of the node.

[0010] Optionally, the weight of each iteration is calculated as follows: determine whether the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold; when the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold, determine the weight of the current iteration as a first preset value; when the absolute value of the adjusted residual value obtained in the previous iteration is less than the adjusted residual value threshold, calculate the square of the adjusted residual value to obtain a target value, calculate the square of the difference between the second preset value and the target value, and obtain the weight of the current iteration.

[0011] Optionally, the method also includes: determining the number of iterations of the regression coefficient, and judging whether the number of iterations is greater than or equal to an iteration threshold; when the number of iterations is less than the iteration threshold, executing the step of multiple iterative calculations; when the number of iterations is greater than or equal to the iteration threshold, and the difference between the regression coefficients calculated from two adjacent iterative calculations is greater than or equal to a preset threshold, stopping the iteration, and determining the regression coefficient obtained from the last iteration as the target regression coefficient between the node current of each user and the target phase line current.

[0012] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a phase-user relationship identification device for a low-voltage distribution network is provided. The device includes: a collection unit, which is used to collect the node current of each node on the target low-voltage distribution network, and the phase line current of each phase line of the target low-voltage distribution network, and generate a node current matrix based on the node currents of all nodes, and generate a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; a calculation unit, which is used to determine the initial weight of the node current of each node for each phase line, calculate the initial regression coefficient and initial adjustment residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and weighted least squares method, update the initial weight by preset weight function and initial adjustment residual value, and The regression coefficient and the adjusted residual value are calculated based on the updated weight, and the calculation is iterated multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold value, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; a determination unit is used to determine the target regression coefficient between the node current of the target node and the phase line current of each phase line, and the phase line corresponding to the maximum target regression coefficient is determined as the target phase line to which the target node belongs, and the phase lines to which all nodes belong are combined according to the phase line type to obtain a user set for each phase line.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of the phase-to-household relationship identification method of the low-voltage distribution network described in each embodiment of the present application.

[0014] Through the present application, the following steps are adopted: collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; for each phase line, determining the initial weight of the node current of each node, calculating the initial regression coefficient and initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and weighted least squares method, updating the initial weight by a preset weight function and the initial adjusted residual value, and calculating the regression coefficient based on the updated weight and adjusting the residual value, iterating the calculation multiple times until the difference between the regression coefficients calculated by two adjacent iterations is less than the preset threshold, and determining the regression coefficient obtained by the last iteration as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained by each iteration is used to update the weight of the next iteration; determining the target regression coefficient between the node current of the target node and the phase line current of each phase line, determining the phase line corresponding to the maximum target regression coefficient as the target phase line to which the target node belongs, combining the phase lines to which all nodes belong according to the phase line type, and obtaining the user set of each phase line, solves the problem of low phase-to-user relationship recognition accuracy of the low-voltage distribution network in the related technology. The regression coefficient between the node current of the user and the phase line current of the phase line is calculated by the weighted least squares method through multiple iterations, so as to determine the phase line to which each user belongs according to the regression coefficient, thereby achieving the effect of improving the phase-to-user relationship recognition accuracy of the low-voltage distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 It is a flow chart of a method for identifying phase-to-household relationships in a low-voltage distribution network provided in an embodiment of the present application;

[0017] Figure 2 It is a flow chart of an optional method for identifying phase-to-household relationships in a low-voltage distribution network provided in an embodiment of the present application;

[0018] Figure 3 is a schematic diagram of the initial regression coefficient of phase A provided in an embodiment of the present application;

[0019] Figure 4 is a schematic diagram of the initial regression coefficient of phase B provided in an embodiment of the present application;

[0020] Figure 5Schematic diagram of the initial regression coefficient of phase C provided in an embodiment of the present application;

[0021] Figure 6 It is a schematic diagram of the regression coefficient of phase A after multiple iterative calculations provided in the embodiment of the present application;

[0022] Figure 7 It is a schematic diagram of the regression coefficient of phase B after multiple iterative calculations provided in the embodiment of the present application;

[0023] Figure 8 It is a schematic diagram of the regression coefficient of the C phase after multiple iterative calculations provided in the embodiment of the present application;

[0024] Fig. 9 is a schematic diagram of the topological relationship between a phase line and a user node subset provided according to an embodiment of the present application;

[0025] Fig.10 It is a schematic diagram of the comparison results of calculating the regression coefficient by the iterative weighted least squares method and the ordinary least squares method according to the embodiment of the present application;

[0026] Fig.11 is a schematic diagram of a phase-to-household relationship identification device for a low-voltage distribution network provided according to an embodiment of the present application;

[0027] Fig.12 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0029] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0031] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0032] It should be noted that the information collected is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of the relevant regions, take necessary confidentiality measures, do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0033] The present invention is described below in conjunction with preferred implementation steps. Figure 1 is a flow chart of a method for identifying a phase-to-household relationship in a low-voltage distribution network according to an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0034] Step S101, collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network.

[0035] Specifically, the target low-voltage distribution network can be a three-phase power supply mode, supplying power to users through three phase lines of phase A, phase B and phase C. Each user connects to any phase line of phase A, phase B and phase C to use electricity. The users connected to each phase line in the target low-voltage distribution network are identified, and the topological relationship of the target low-voltage distribution network is obtained. By identifying the topological relationship, the management efficiency of the low-voltage distribution network can be improved.

[0036] In order to identify the phase-to-user relationship of the target low-voltage distribution network, it is first necessary to collect the measured values ​​of the phase line current of each phase line and the node current of each user's corresponding node. During the collection, multiple sample data can be collected for each node and each phase line according to the preset time interval, and the node current matrix and phase line current sequence are generated according to the collected sample data.

[0037] Step S102, for each phase line, determine the initial weight of the node current of each node, calculate the initial regression coefficient and initial adjustment residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and weighted least squares method, update the initial weight by a preset weight function and the initial adjustment residual value, calculate the regression coefficient and the adjustment residual value based on the updated weight, and iterate the calculation multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold, and determine the regression coefficient obtained in the last iteration as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration.

[0038] Specifically, in the weighted least squares method, for data with heteroscedasticity, the influence of each data point can be adjusted by introducing weights so that the model fits the data better. The embodiment of the present application calculates the regression coefficient between the node current of each user and the phase line current between each phase line by weighted least squares method, thereby identifying the phase line to which each user belongs. After obtaining the node current matrix and the phase line current sequence of each phase line, an initial weight is set for each node current, for example, the initial weights of all node currents are set to 1, and an initial weight matrix is ​​generated. The weighted least squares method is used to calculate the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line, and the number of iterations is set to 0.

[0039] The preset weight function can be a bisquare weight function (a function used for data fitting and regression analysis). The node current weight is updated by the preset weight function and the adjusted error value of each iteration, and the regression coefficient of each iteration is recalculated. The adjusted residual value of the next iteration is calculated according to the calculation formula of the adjusted residual; the iterative calculation is repeated until the difference between the estimated value of the regression coefficient in two consecutive iterations is less than the preset threshold value, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line. For example, if the estimated value of any regression coefficient is , m is the target phase line, j is the jth user, and the change range in two consecutive iterations is less than the preset threshold ,Right now , g is the g-th iteration, then the target regression coefficient between the node current of each node and the phase current of the target phase line m is output .

[0040] It should be noted that for the general multiple linear regression model Y=Xβ+e, when the error term e t (t=1,…,T) satisfies the following formula:

[0041] (1)

[0042] Where T represents the Tth sample point collected, represents the expectation of the error term, represents the covariance of the error term, represents the variance of the error term, and the standard deviation of the error term , …, If they are not completely equal, the model is said to have heteroscedasticity. The root cause of heteroscedasticity is that the data errors of some abnormal sample points are too large, resulting in abnormal probability distribution of regression residuals. When the model has heteroscedasticity, the weighted least squares method can be considered to estimate the regression coefficient. The node currents of users in the embodiment of the present application belong to different phase lines, so the weighted least squares method can be used to estimate the regression coefficient.

[0043] The regression coefficient is calculated as follows:

[0044] (2)

[0045] Among them, w i is the weight value of the i-th sampling point of each given user, x i1 represents the node current of the i-th sampling point of the first user, N represents the total number of users using the target low-voltage distribution network, and the weighted least squares estimation is to find the estimated value of the regression coefficient m makes Q w To reach the minimum value, the weight matrix W is recorded as:

[0046] (3)

[0047] The calculation formula of the regression coefficient of each user's node current to the target phase line m is as follows:

[0048]

[0049] Where X represents the node current matrix, X′ represents the transpose of X, and Y represents the phase current sequence of the target phase m. Select the bisquare weight function and calculate the weight W of the g+1th iteration g+1 . Calculate the regression coefficient for the g+1th iteration .

[0050] Step S103, determine the target regression coefficient between the node current of the target node and the phase line current of each phase line, determine the phase line corresponding to the maximum target regression coefficient as the target phase line belonging to the target node, combine the phase lines belonging to all nodes according to the phase line type, and obtain the user set of each phase line.

[0051] Specifically, different phase line currents are used as dependent variables, and the above process is repeated to obtain the target regression coefficient of each user relative to each phase line; for the node current of each target node, the phase line corresponding to the maximum regression coefficient among the target regression coefficients of all phase lines is the target phase line to which the target node belongs, and all phase lines are traversed to determine the node subset of the user corresponding to each phase line; the detailed information of all phase line-user node subsets is found, and the phase line-user node subset (that is, the user set of each phase line) represents the phase where any user in the data set is located, which can reflect the topological relationship of the target low-voltage distribution network.

[0052] Output each phase line-user node subset, where the phase line represents a phase in a group of outgoing lines, the node represents the user connected to the phase line, and the phase line-user node subset represents the phase where any user in the data set is located, which can reflect the topological relationship of the target low-voltage distribution network.

[0053] The method for identifying the relationship between phases and users of a low-voltage distribution network provided in an embodiment of the present application collects the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generates a node current matrix based on the node currents of all nodes, and generates a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; for each phase line, the initial weight of the node current of each node is determined, and the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line are calculated based on the initial weight and weighted least squares method, the initial weight is updated by a preset weight function and the initial adjusted residual value, and based on the updated weight Calculate the regression coefficient and adjust the residual value, iterate the calculation multiple times until the difference between the regression coefficients calculated by two adjacent iterations is less than the preset threshold, and determine the regression coefficient obtained by the last iteration as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained by each iteration is used to update the weight of the next iteration; determine the target regression coefficient between the node current of the target node and the phase line current of each phase line, determine the phase line corresponding to the maximum target regression coefficient as the target phase line to which the target node belongs, combine the phase lines to which all nodes belong according to the phase line type, and obtain the user set of each phase line, thereby solving the problem of low phase-to-user relationship recognition accuracy of the low-voltage distribution network in the related technology. The regression coefficient between the user's node current and the phase line current of the phase line is calculated by the weighted least squares method through multiple iterations, so as to determine the phase line to which each user belongs according to the regression coefficient, thereby achieving the effect of improving the phase-to-user relationship recognition accuracy of the low-voltage distribution network.

[0054] In order to calculate the regression coefficient of each node current relative to the phase line current, it is necessary to first determine the node current matrix and the phase line current sequence. Optionally, in the phase-to-user relationship identification method of the low-voltage distribution network provided in the embodiment of the present application, the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network are collected, and a node current matrix is ​​generated based on the node currents of all nodes, and the phase line current sequence of the phase line is generated based on the phase line current of each phase line, including: for each node, collecting T node currents of the target node at preset time intervals to generate a node current sequence of the target node, and generating a node current matrix based on the node current sequences of N nodes, wherein T and N are both positive integers; for each phase line, collecting T phase line currents of the target phase line at preset time intervals to generate a phase line current sequence of the target phase line.

[0055] For example, the preset time interval can be 15 minutes, the collection cycle is two days, each node collects a total of T = 192 sample points (that is, node current), each node current sequence contains 192 sample point data of the target node, and the node current sequences of the collected N nodes are combined to form a node current matrix X. Each phase line also collects T sample points, and each phase line current sequence contains 192 phase line current data of the target phase line.

[0056] The embodiment of the present application collects multiple data of node current and phase line current, calculates the regression coefficient through multiple sampling data, and ensures the accuracy of the calculation result.

[0057] After determining the node current matrix and the phase line current sequence, the initial regression coefficient and the initial adjusted residual value are calculated by the weighted least squares method. Optionally, in the phase-to-household relationship identification method of the low-voltage distribution network provided in the embodiment of the present application, the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line are calculated based on the initial weights and the weighted least squares method, including: determining the initial weight matrix based on the initial weights of all nodes, inputting the initial weight matrix, the node current matrix and the phase line current sequence of the target phase line into the regression coefficient calculation formula to obtain the initial regression coefficient; calculating the fitting residual value of each node by the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line; and calculating the initial adjusted residual value based on the fitting residual value of each node.

[0058] Specifically, the calculation formula of the initial regression coefficient between the node current of each node and the phase line current of the target phase line m is as follows:

[0059]

[0060] Where X represents the node current matrix, X′ represents the transpose of X, and Y represents the phase current sequence of the target phase line m. Therefore, by inputting the phase current sequence Y of each phase line a , Y b , Y c , user's node current matrix X, initial weight matrix W 0 The initial regression coefficient of each node relative to each phase line can be calculated. , the calculation formula for adjusting the residual value is as follows:

[0061] (6)

[0062] in, represents the adjusted residual value, represents the fitting residual value of the node current, K represents the adjustment constant parameter, S represents the standard deviation of the error term, Indicates preset parameters.

[0063] The embodiment of the present application calculates the initial regression coefficient and the initial adjusted residual value as the data for calculating the regression coefficient in the next iteration.

[0064] Optionally, in the phase-to-user relationship identification method of the low-voltage distribution network provided in an embodiment of the present application, the fitting residual value of each node is calculated through the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line, including: calculating the product of the initial regression coefficient and the node current matrix to obtain the current prediction value sequence of each node; calculating the difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node to obtain the fitting residual value of each node.

[0065] Specifically, the fitting residual value is the difference between the observed value and the fitting value, and the observed value is also the measured phase line current sequence. The fitting value refers to the value calculated based on the product of the regression coefficient and the independent variable. Therefore, by calculating the product of the initial regression coefficient and the node current matrix, the current prediction value sequence of each node can be obtained. By calculating the difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node, the fitting residual value of each node is obtained. The embodiment of the present application determines the adjustment residual value by calculating the fitting residual value.

[0066] Optionally, in the phase-to-user relationship identification method of the low-voltage distribution network provided in an embodiment of the present application, calculating the initial adjusted residual value based on the fitted residual value of each node includes: determining the adjustment constant parameter and the preset parameter, and calculating the standard deviation of the error term based on the fitted residual values ​​of all nodes; calculating the product of the adjustment constant parameter, the preset parameter and the standard deviation of the error term, and calculating the ratio of the fitted residual value of each node to the product to obtain the initial adjusted residual value of the node.

[0067] Specifically, according to the calculation formula of the adjusted residual value, the preset parameters h in tt is the t-th diagonal element of the hat matrix H of the node current matrix; K is an adjustment constant, and lowering the adjustment constant will increase the downweight assigned to large residuals; increasing the adjustment constant will reduce the downweight assigned to large residuals; S is an estimate of the standard deviation of the error term, S = MAD / 0.6745. MAD is the absolute deviation of the median of the fitted residual values ​​of all samples, and the constant 0.6745 makes the estimate of the normal distribution unbiased. The embodiment of the present application calculates the initial adjusted residual value by fitting the residual value, and then updates the weight of the next iteration based on the initial adjusted residual value.

[0068] The weight of each iteration is updated according to the adjusted residual value obtained in the previous iteration. Optionally, in the phase-to-household relationship identification method of the low-voltage distribution network provided in the embodiment of the present application, the weight of each iteration is calculated as follows: determine whether the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold; when the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold, determine the weight of the current iteration as the first preset value; when the absolute value of the adjusted residual value obtained in the previous iteration is less than the adjusted residual value threshold, calculate the square of the adjusted residual value to obtain the target value, calculate the square of the difference between the second preset value and the target value, and obtain the weight of the current iteration.

[0069] Specifically, when the model fitting of the regression coefficient is calculated by weighted least squares method, different weights need to be assigned according to the degree of data variation. A larger weight is assigned to data with a smaller degree of variation and more accurate measurement, and a smaller weight is assigned to data with a larger degree of variation and more accurate measurement, so that the residual sum of squares of the weighted regression model is minimized; the residual value based on the fitting of the tth sample Set the preset weight function wt(·) to affect the weight of each iteration by adjusting the residual value of the sample point. The weight update calculation formula is as follows:

[0070] (7)

[0071] Among them, the preset value can be 0, and the adjustment residual value threshold can be set to 1. It should be noted that the embodiment of the present application does not involve the situation where the absolute value of the adjustment residual value is greater than or equal to the adjustment residual value threshold.

[0072] This embodiment reduces the impact on the regression coefficient by eliminating abnormal points or differential weighting, thereby improving the accuracy of low-voltage distribution network topology identification, which is beneficial to the management and regulation of low-voltage distribution networks and provides important data support for applications such as low-voltage distribution network fault location, line loss management, and power theft detection.

[0073] In order to prevent the calculation of the regression coefficient from taking too long, the target regression coefficient is determined by setting an iteration number threshold. Optionally, in the phase-to-user relationship identification method of the low-voltage distribution network provided in an embodiment of the present application, the method also includes: determining the number of iterations of the regression coefficient, and judging whether the number of iterations is greater than or equal to the iteration number threshold; when the number of iterations is less than the iteration number threshold, performing multiple iterative calculation steps; when the number of iterations is greater than or equal to the iteration number threshold, and the difference between the regression coefficients of two adjacent iterative calculations is greater than or equal to a preset threshold, stopping the iteration, and determining the regression coefficient obtained from the last iteration as the target regression coefficient between the node current of each user and the target phase line current.

[0074] Specifically, in order to prevent the iterative calculation of the regression coefficient from falling into an infinite loop, an iteration number threshold can be set artificially. In multiple iterative calculations of the regression coefficient, if the number of iterations is less than the iteration number threshold before reaching the iteration number threshold, the iteration is stopped and the regression coefficient obtained by the last iteration is output. If the difference between the regression coefficients calculated by two adjacent iterations of all previous iterations is greater than or equal to the preset threshold after reaching the iteration number threshold, the iteration is stopped when the number of iterations reaches the iteration number threshold, and the regression coefficient obtained by the last iteration is output as the target regression coefficient. The embodiment of the present application avoids the iterative calculation of the regression coefficient from falling into an infinite loop by setting the iteration number threshold.

[0075] According to another embodiment of the present application, this embodiment provides an optional method for identifying a phase-to-household relationship in a low-voltage distribution network. Figure 2 FIG. 1 is a flow chart of an optional method for identifying the relationship between households in a low-voltage distribution network provided in an embodiment of the present application. Figure 2 As shown, the method includes:

[0076] S201: Collect sample data, input the current measurement value vector of each phase line, the user node current measurement value matrix, and the initial weight matrix.

[0077] Specifically, a low-voltage distribution network test system with a total number of 63 users is established, among which phase A users are numbered 1-21, phase B users are numbered 22-42, and phase C users are numbered 43-63. The collection time interval is 15 minutes, the time period is 2 days, and there are a total of T=192 time sections. The A, B, and C phase buses on the low-voltage side of the distribution transformer are used as balance nodes, and multi-section power flow simulation calculations are performed to obtain the real current data of each node, that is, the phase line current measurement value vector and the user node current measurement value matrix, and an initial weight matrix with the weight of each sample being 1 is set.

[0078] S202: assigning initial weights to the sample points, and calculating the estimated values ​​of the regression coefficients and the adjusted residual values ​​based on the weighted least squares method, and setting the number of iterations g=0;

[0079] Specifically, the calculation formula of the regression coefficient is as follows:

[0080] (8)

[0081] Among them, w i is the weight value of the i-th sampling point of each given user, x i1 represents the node current of the i-th sampling point of the first user, N represents the total number of users using the target low-voltage distribution network, and the weighted least squares estimation is to find the estimated value m of the regression coefficient so that Q w To reach the minimum value, the weight matrix W is recorded as:

[0082] (9)

[0083] The calculation formula of the regression coefficient of each user's node current to the target phase line m is as follows:

[0084]

[0085] Wherein, X represents the node current matrix, X′ represents the transpose of X, and Y represents the phase line current sequence of the target phase line m.

[0086] For the first iteration, the estimated least squares regression coefficients of the 63 nodes for the ABC three-phase are as follows: Figure 3 , Figure 4 and Figure 5 As shown, Figure 3 It is a schematic diagram of the initial regression coefficient of phase A provided according to the embodiment of the present application. Figure 4 is a schematic diagram of the initial regression coefficient of phase B provided in the embodiment of the present application, Figure 5 It is a schematic diagram of the initial regression coefficient of phase C provided according to an embodiment of the present application.

[0087] S203: According to the calculation formula of the adjusted residual and the estimated value of the regression coefficient obtained after the g-th iteration, the adjusted residual value of the g+1-th iteration is obtained.

[0088] Specifically, the calculation formula for adjusting the residual value is as follows:

[0089] (11)

[0090] in, represents the adjusted residual value, represents the fitting residual value of the node current, K represents the adjustment constant parameter, S represents the standard deviation of the error term, Represents a preset parameter. The fitting residual value is the difference between the observed value and the fitting value, and the observed value is also the measured phase line current sequence. The fitting value refers to the value calculated based on the product of the regression coefficient and the independent variable. Therefore, by calculating the product of the initial regression coefficient and the node current matrix, the current prediction value sequence of each node can be obtained. By calculating the difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node, the fitting residual value of each node is obtained. The embodiment of the present application determines the adjustment residual value by calculating the fitting residual value.

[0091] S204: Select a weight function and calculate the weight of the g+1th iteration.

[0092] Specifically, the weight update calculation formula is as follows:

[0093] (12)

[0094] S205: Calculate the estimated value of the regression coefficient of the g+1th iteration;

[0095] Specifically, calculate the weight W of the g+1th iteration g+1 . Calculate the regression coefficient for the g+1th iteration .

[0096] S206: Repeat S203 to S205. If the variation of the estimated value of any regression coefficient between two consecutive iterations is less than a certain preset threshold, the iterative weighted least squares estimated value of the regression coefficient is output;

[0097] Specifically, the regression coefficient of each iteration is recalculated, and the adjusted residual value of the next iteration is calculated according to the calculation formula of the adjusted residual; the iterative calculation is repeated until the difference between the estimated value of the regression coefficient in two consecutive iterations is less than the preset threshold, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line. For example, if the estimated value of any regression coefficient is , m is the target phase line, j is the jth user, and the change range in two consecutive iterations is less than the preset threshold ,Right now , g is the g-th iteration, then the target regression coefficient between the node current of each node and the phase current of the target phase line m is output .

[0098] S207: Using different phase line currents as dependent variables, a series of estimated values ​​of regression coefficients are obtained.

[0099] Specifically, Figure 6 It is a schematic diagram of the regression coefficient of phase A after multiple iterative calculations provided in the embodiment of the present application. Figure 7 is a schematic diagram of the regression coefficient of phase B after multiple iterative calculations provided in the embodiment of the present application, Figure 8 It is a schematic diagram of the regression coefficient of the C phase after multiple iterative calculations provided in the embodiment of the present application.

[0100] S208: Determine whether the user node is grouped into the phase line-user node subset according to the estimated value relationship of the regression coefficient between the user node and the phase line.

[0101] Specifically, according to Figure 6 , Figure 7 and Figure 8 For each node, the regression coefficient for each of the three phases ABC takes the maximum value, and the phase sequence corresponding to the maximum value is the phase line (phase line) corresponding to the node (user node). For example, for user node 1, the final estimated values ​​of the regression coefficients corresponding to the three phases ABC are 1.69, 0.15, and 0.89, respectively, so it is determined that user node 1 belongs to phase A.

[0102] S209: All phase lines are traversed to obtain each phase line-user node subset.

[0103] Fig. 9 is a schematic diagram of the topological relationship between the phase line and the user node subset provided in an embodiment of the present application, such as Fig. 9 As shown, the phase line represents a phase in a group of outgoing lines, the user node represents the user connected to the connected line, and the phase line-user node subset represents the phase where any user in the data set is located, which can reflect the topological relationship of the group of outgoing lines.

[0104] S210: Output each phase line-user node subset as the topological relationship of the group of outgoing lines.

[0105] Specifically, the real data is: users 1-21 are connected to phase A, users 22-42 are connected to phase B, and users 43-63 are connected to phase C. Suppose method 1 M1: iterative weighted least squares method, which is the optional phase-user relationship identification method provided in this embodiment; method 2 M2: ordinary least squares method. By comparing the experimental results of the two different methods, it is concluded that the iterative weighted least squares method is more superior. Repeat 5 groups of experiments, each group of experiments sets the M1 and M2 methods for comparison, and all are carried out according to steps S201-S210. Fig.10 It is a schematic diagram of the comparison results of calculating the regression coefficient by the iterative weighted least squares method and the ordinary least squares method provided in the embodiment of the present application.

[0106] It can be seen from simulation calculations that compared with the ordinary least squares method, the topology recognition accuracy of the iterative weighted least squares method can be improved with a high probability, and the accuracy can be maintained above 95%.

[0107] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0108] The embodiment of the present application also provides a phase-to-household relationship identification device for a low-voltage distribution network. It should be noted that the phase-to-household relationship identification device for a low-voltage distribution network in the embodiment of the present application can be used to execute the phase-to-household relationship identification method for a low-voltage distribution network provided in the embodiment of the present application. The following introduces the phase-to-household relationship identification device for a low-voltage distribution network provided in the embodiment of the present application.

[0109] Fig.11 Schematic diagram of a phase-to-household relationship identification device for a low-voltage distribution network provided according to an embodiment of the present application. Fig.11 As shown, the device comprises:

[0110] The acquisition unit 1101 is used to collect the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generate a node current matrix based on the node currents of all nodes, and generate a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network;

[0111] A calculation unit 1102 is used to determine the initial weight of the node current of each node for each phase line, calculate the initial regression coefficient and the initial adjustment residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and the weighted least squares method, update the initial weight by a preset weight function and the initial adjustment residual value, calculate the regression coefficient and the adjustment residual value based on the updated weight, perform multiple iterations until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold value, and determine the regression coefficient obtained in the last iteration as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration;

[0112] The determination unit 1103 is used to determine the target regression coefficient between the node current of the target node and the phase line current of each phase line, determine the phase line corresponding to the maximum target regression coefficient as the target phase line belonging to the target node, and combine the phase lines belonging to all nodes according to the phase line type to obtain a user set for each phase line.

[0113] The phase-user relationship identification device of the low-voltage distribution network provided in the embodiment of the present application collects the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network through the collection unit 1101, and generates a node current matrix based on the node currents of all nodes, and generates a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; the calculation unit 1102 determines the initial weight of the node current of each node for each phase line, calculates the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and the weighted least squares method, updates the initial weight through the preset weight function and the initial adjusted residual value, calculates the regression coefficient and the adjusted residual value based on the updated weight, and iterates the calculation multiple times until the regression coefficients calculated in two adjacent iterations are equal to the value between the regression coefficients calculated in two adjacent iterations. The difference is less than a preset threshold value, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; a determination unit 1103 determines the target regression coefficient between the node current of the target node and the phase line current of each phase line, and determines the phase line corresponding to the maximum target regression coefficient as the target phase line to which the target node belongs, and combines the phase lines to which all nodes belong according to the phase line type to obtain the user set of each phase line, thereby solving the problem of low phase-to-user relationship recognition accuracy of the low-voltage distribution network in the related art, and calculating the regression coefficient between the user's node current and the phase line current of the phase line by the weighted least squares method of multiple iterations, thereby determining the phase line to which each user belongs according to the regression coefficient, thereby achieving the effect of improving the phase-to-user relationship recognition accuracy of the low-voltage distribution network.

[0114] Optionally, in the phase-to-user relationship identification device of the low-voltage distribution network provided in the embodiment of the present application, the collection unit 1101 includes: a first collection module, used for collecting T node currents of the target node at preset time intervals for each node, generating a node current sequence of the target node, and generating a node current matrix based on the node current sequences of N nodes, wherein T and N are both positive integers; a second collection module, used for collecting T phase line currents of the target phase line at preset time intervals for each phase line, generating a phase line current sequence of the target phase line.

[0115] Optionally, in the phase-to-household relationship identification device of the low-voltage distribution network provided in the embodiment of the present application, the calculation unit 1102 includes: an input module, used to determine the initial weight matrix based on the initial weights of all nodes, and input the initial weight matrix, the node current matrix and the phase line current sequence of the target phase line into the regression coefficient calculation formula to obtain the initial regression coefficient; a first calculation module, used to calculate the fitting residual value of each node through the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line; and a second calculation module, used to calculate the initial adjusted residual value based on the fitting residual value of each node.

[0116] Optionally, in the phase-to-user relationship identification device of the low-voltage distribution network provided in an embodiment of the present application, the first calculation module includes: a first calculation sub-module, used to calculate the product of the initial regression coefficient and the node current matrix to obtain a current prediction value sequence for each node; a second calculation sub-module, used to calculate the difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node to obtain a fitting residual value for each node.

[0117] Optionally, in the phase-to-household relationship identification device of the low-voltage distribution network provided in the embodiment of the present application, the second calculation module includes: a first determination submodule, used to determine the adjustment constant parameters and preset parameters, and calculate the standard deviation of the error term based on the fitting residual values ​​of all nodes; a third calculation submodule, used to calculate the product of the adjustment constant parameters, the preset parameters and the standard deviation of the error term, and calculate the ratio of the fitting residual value of each node to the product to obtain the initial adjusted residual value of the node.

[0118] Optionally, in the phase-to-household relationship identification device of the low-voltage distribution network provided in the embodiment of the present application, the device also includes: a first judgment unit, used to judge whether the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold; a weight determination unit, used to determine the weight of the current iteration as a first preset value when the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold; a weight calculation unit, used to calculate the square of the adjusted residual value to obtain a target value when the absolute value of the adjusted residual value obtained in the previous iteration is less than the adjusted residual value threshold, and calculate the square of the difference between the second preset value and the target value to obtain the weight of the current iteration.

[0119] Optionally, in the phase-to-household relationship identification device for a low-voltage distribution network provided in an embodiment of the present application, the device also includes: a second judgment unit, used to determine the number of iterations of the regression coefficient, and judge whether the number of iterations is greater than or equal to an iteration number threshold; an execution unit, used to execute multiple iterative calculation steps when the number of iterations is less than the iteration number threshold; a regression coefficient determination unit, used to stop the iteration when the number of iterations is greater than or equal to the iteration number threshold, and the difference between the regression coefficients of two adjacent iterative calculations is greater than or equal to a preset threshold, and determine the regression coefficient obtained from the last iteration as the target regression coefficient between the node current of each user and the target phase line current.

[0120] The phase-to-household relationship identification device of the low-voltage distribution network includes a processor and a memory. The above-mentioned acquisition unit 1101, calculation unit 1102 and determination unit 1103 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0121] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the accuracy of the household relationship recognition of the low-voltage distribution network can be improved by adjusting the kernel parameters.

[0122] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0123] An embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, a method for identifying a phase-to-user relationship in a low-voltage distribution network is implemented.

[0124] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes a method for identifying a phase-to-user relationship in a low-voltage distribution network when the program is running.

[0125] Fig.12 is a schematic diagram of an electronic device provided according to an embodiment of the present application. Fig.12As shown, the electronic device 1201 includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the following steps are implemented: collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; for each phase line, determining the initial weight of the node current of each node, calculating the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and the weighted least squares method, and The weight function and the initial adjusted residual value are used to update the initial weight, and the regression coefficient and the adjusted residual value are calculated based on the updated weight. The calculation is repeated multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than the preset threshold. The regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; the target regression coefficient between the node current of the target node and the phase line current of each phase line is determined, and the phase line corresponding to the maximum target regression coefficient is determined as the target phase line to which the target node belongs. The phase lines of all nodes are combined according to the phase line type to obtain the user set of each phase line. The device in this article can be a server, PC, PAD, mobile phone, etc.

[0126] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize the user using the target low-voltage distribution network; for each phase line, determining the initial weight of the node current of each node, calculating the initial regression coefficient and the initial adjusted residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and the weighted least squares method, and The initial weights are updated based on the number and the initial adjusted residual value, and the regression coefficient and the adjusted residual value are calculated based on the updated weights. The calculation is iterated multiple times until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold value, and the regression coefficient obtained in the last iteration is determined as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; the target regression coefficient between the node current of the target node and the phase line current of each phase line is determined, and the phase line corresponding to the maximum target regression coefficient is determined as the target phase line to which the target node belongs, and the phase lines of all nodes are combined according to the phase line type to obtain the user set of each phase line.

[0127] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0128] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0129] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0130] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0131] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0132] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0133] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0134] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0135] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0136] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for identifying phase-to-household relationships in a low-voltage distribution network, characterized in that: include: Collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the nodes are used to characterize users using the target low-voltage distribution network; For each phase line, determine the initial weight of the node current of each node, calculate the initial regression coefficient and initial adjustment residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and weighted least squares method, update the initial weight by a preset weight function and the initial adjustment residual value, calculate the regression coefficient and the adjustment residual value based on the updated weight, perform multiple iterations until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold, determine the regression coefficient obtained in the last iteration as the target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; Determine the target regression coefficient between the node current of the target node and the phase line current of each phase line, determine the phase line corresponding to the maximum target regression coefficient as the target phase line to which the target node belongs, combine the phase lines to which all nodes belong according to the phase line type, and obtain a user set for each phase line.

2. The method according to claim 1, characterized in that Collecting the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generating a node current matrix based on the node currents of all nodes, and generating a phase line current sequence of the phase line based on the phase line current of each phase line includes: For each node, T node currents of the target node are collected at preset time intervals to generate a node current sequence of the target node, and the node current matrix is ​​generated based on the node current sequences of N nodes, where T and N are both positive integers; For each phase line, T phase line currents of the target phase line are collected at preset time intervals to generate a phase line current sequence of the target phase line.

3. The method according to claim 1, characterized in that Calculating the initial regression coefficient and the initial adjustment residual value between the node current of each node and the phase line current of the target phase line based on the initial weight and the weighted least square method includes: An initial weight matrix is ​​determined based on the initial weights of all nodes, and the initial weight matrix, the node current matrix and the phase line current sequence of the target phase line are input into a regression coefficient calculation formula to obtain the initial regression coefficient; Calculating the fitting residual value of each node through the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line; The initial adjusted residual value is calculated based on the fitting residual value of each node.

4. The method according to claim 3, characterized in that Calculating the fitting residual value of each node by using the initial regression coefficient, the node current matrix and the phase line current sequence of the target phase line includes: Calculating the product of the initial regression coefficient and the node current matrix to obtain a current prediction value sequence for each node; The difference between the phase line current sequence of the target phase line and the current prediction value sequence of each node is calculated to obtain the fitting residual value of each node.

5. The method according to claim 3, characterized in that: Calculating the initial adjusted residual value based on the fitting residual value of each node includes: Determine the adjustment constant parameters and preset parameters, and calculate the standard deviation of the error term based on the fitting residual values ​​of all nodes; The product of the adjustment constant parameter, the preset parameter and the standard deviation of the error term is calculated, and the ratio of the fitting residual value of each node to the product is calculated to obtain the initial adjustment residual value of the node.

6. The method according to claim 1, characterized in that The weights for each iteration are calculated as follows: Determine whether the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold; When the absolute value of the adjusted residual value obtained in the previous iteration is greater than or equal to the adjusted residual value threshold, the weight of the current iteration is determined as a first preset value; When the absolute value of the adjusted residual value obtained in the previous iteration is less than the adjusted residual value threshold, the square of the adjusted residual value is calculated to obtain the target value, and the square of the difference between the second preset value and the target value is calculated to obtain the weight of the current iteration.

7. The method according to claim 1, characterized in that The method further comprises: Determine the number of iterations of the regression coefficient, and judge whether the number of iterations is greater than or equal to a threshold number of iterations; When the number of iterations is less than the iteration number threshold, the steps of multiple iterative calculations are performed; when the number of iterations is greater than or equal to the iteration number threshold, and the difference between the regression coefficients calculated in two adjacent iterative calculations is greater than or equal to a preset threshold, the iteration is stopped, and the regression coefficient obtained from the last iteration is determined as the target regression coefficient between the node current of each user and the target phase line current.

8. A phase-to-household relationship identification device for a low-voltage distribution network, characterized in that: include: A collection unit, used to collect the node current of each node on the target low-voltage distribution network and the phase line current of each phase line of the target low-voltage distribution network, and generate a node current matrix based on the node currents of all nodes, and generate a phase line current sequence of the phase line based on the phase line current of each phase line, wherein the node is used to characterize a user using the target low-voltage distribution network; A calculation unit, for determining, for each phase line, an initial weight of a node current of each node, calculating an initial regression coefficient and an initial adjusted residual value between the node current of each node and the phase line current of a target phase line based on the initial weight and a weighted least square method, updating the initial weight by a preset weight function and the initial adjusted residual value, and calculating the regression coefficient and the adjusted residual value based on the updated weight, performing multiple iterations of calculations until the difference between the regression coefficients calculated in two adjacent iterations is less than a preset threshold value, and determining the regression coefficient obtained in the last iteration as a target regression coefficient between the node current of each node and the phase line current of the target phase line, wherein the adjusted residual value obtained in each iteration is used to update the weight of the next iteration; A determination unit is used to determine a target regression coefficient between a node current of a target node and a phase line current of each phase line, determine a phase line corresponding to a maximum target regression coefficient as a target phase line to which the target node belongs, and combine the phase lines to which all nodes belong according to the phase line type to obtain a user set for each phase line.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for identifying the relationship between households in a low-voltage distribution network as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method for identifying the relationship between households in a low-voltage distribution network as described in any one of claims 1 to 7.

Images