A method and system for identifying station-user relationships in a distribution network

Abstract

The present invention discloses a method and system for identifying the relationship between stations and households in a distribution network. The method obtains distribution transformer voltage data and low-voltage user voltage data of the distribution network, and after data cleaning and data reconstruction of the distribution transformer voltage data and the low-voltage user voltage data, calculates the correlation coefficient between the low-voltage user and the ledger transformer, and judges whether the station-household relationship in the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. The method does not require manual identification, nor does it require reliance on dedicated identification equipment to use a carrier communication method or a pulse current method for identification. The method has a high degree of intelligence, and solves the technical problem that traditional manual identification and methods using dedicated station area identification equipment for station-household relationship identification have a low degree of intelligence and cannot meet the intelligent development needs of distribution network stations.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network data analysis, and in particular to a method and system for identifying station-user relationships in a distribution network. Background Art

[0002] The distribution network has a large number of users in the substation area and a complex network structure. When the user's wiring is changed or the power grid company carries out line reconstruction for load balancing, due to reasons such as untimely records or adjustments to rights and responsibilities, the records of the ownership relationship between the user's line end and the concentrator are inaccurate, the relationship between the station and the user does not match the actual situation, and it is difficult to define the substation ownership of users of ground cable power supply, and other substation file confusion problems.

[0003] Traditional methods for identifying the relationship between substations and users include manual identification and identification using dedicated substation identification equipment. The manual identification method requires power personnel to go to the site to check the ownership of substation users one by one, which is inefficient. The dedicated substation equipment identification method mainly uses carrier communication method or pulse current method. The carrier communication method has the problem of "cross-substation". The pulse current method cannot communicate in two directions and requires carrier communication method as auxiliary communication. In addition, there are safety hazards in the process of using current clamps to identify users in distribution substations. The cost of dedicated substation equipment is relatively high and it also cannot meet the intelligent development needs of distribution network substations. Therefore, it is necessary to provide a new method for identifying the relationship between substations and users in the distribution network to avoid the problems of the traditional method of identifying the relationship between substations and users and improve the intelligence level of the identification of the relationship between substations and users in the distribution network. Summary of the Invention

[0004] The embodiments of the present invention provide a method and system for identifying the relationship between distribution network stations and users, which are used to solve the technical problem that the traditional manual identification and the method of using dedicated substation identification equipment to identify the relationship between distribution network stations and users have low intelligence and cannot meet the intelligent development needs of distribution network substations.

[0005] In view of this, the present invention provides a method for identifying the relationship between distribution network stations and users, comprising the following steps:

[0006] S1. Obtain distribution network transformer voltage data and low-voltage user voltage data;

[0007] S2. Clean the distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes normalizing data outliers and filling in missing data values.

[0008] S3, reconstructing the distribution transformer voltage data and the low-voltage user voltage data after data cleaning to obtain a distribution transformer voltage time series and a low-voltage user voltage time series that meet a preset length;

[0009] S4. Calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series;

[0010] S5. Determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than a first threshold, the ledger's transformer-household relationship is correct.

[0011] Optionally, after step S5, the following steps are further included:

[0012] S6. If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second relevant transformer by more than the preset difference. At the same time, the most relevant transformer and the ledger transformer are consistent. If so, the ledger's station-household relationship is correct.

[0013] Optionally, after step S6, the following steps are further included:

[0014] S7. If the correlation coefficient between the low-voltage user and the most relevant transformer is not greater than the second threshold, or is less than the preset difference greater than the correlation coefficient of the second most relevant transformer, or the most relevant transformer is inconsistent with the ledger transformer, then calculate the voltage error caused by the low-voltage user being attributed to a specific transformer, select the transformer with the smallest error as the calculation result, and determine whether the transformer with the smallest error is consistent with the ledger transformer. If so, the transformer-to-household relationship in the ledger is correct.

[0015] Optionally, after step S7, the following steps are further included:

[0016] S8. If the transformer with the smallest error in step S7 is inconsistent with the transformer in the ledger, calculate the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-to-user relationship has not been determined, and select the transformer corresponding to the nearest area. If the transformer corresponding to the nearest area is consistent with the transformer in the ledger, the transformer-to-user relationship in the ledger is correct. Among them, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-to-user relationship has not been determined is defined as 1 minus the correlation coefficient between the low-voltage user and the transformer in the area confirmed by the ledger.

[0017] Optionally, after step S8, the following steps are further included:

[0018] S9. If the transformer corresponding to the nearest area in step S8 is inconsistent with the transformer in the ledger, the correlation coefficient between low-voltage users and low-voltage users is calculated, and all confirmed users whose correlation coefficient with undetermined users is greater than the first threshold are screened. Each confirmed user votes for his or her own transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct.

[0019] Optionally, after step S9, the following steps are further included:

[0020] S10. If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, select the transformer with the largest correlation coefficient. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's transformer-user relationship is correct.

[0021] Optionally, after step S9, the following steps are further included:

[0022] S10. If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, select the transformer with the largest correlation coefficient. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's transformer-user relationship is correct.

[0023] Optionally, in step S4, the correlation coefficient between the low-voltage user and the distribution transformer is a Pearson correlation coefficient.

[0024] Optionally, in step S2, data cleaning further includes normalization processing and principal component analysis processing;

[0025] Normalization was performed using Z-score standardization.

[0026] The second aspect of the present invention further provides a distribution network station-user relationship identification system, comprising the following modules:

[0027] Data acquisition module, used to obtain distribution network distribution transformer voltage data and low-voltage user voltage data;

[0028] Data cleaning module, used to clean distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes data outlier standardization and data missing value filling.

[0029] The data reconstruction module is used to reconstruct the distribution transformer voltage data and low-voltage user voltage data after data cleaning to obtain the distribution transformer voltage time series and low-voltage user voltage time series that meet the preset length;

[0030] A correlation coefficient calculation module is used to calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series;

[0031] The station-household relationship identification module is used to determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than the first threshold, the station-household relationship of the ledger is correct.

[0032] Optionally, the platform-user relationship identification module is further used to:

[0033] If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, then determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second relevant transformer by more than the preset difference. At the same time, the most relevant transformer and the ledger transformer are consistent. If so, the ledger's station-household relationship is correct.

[0034] It can be seen from the above technical solutions that the embodiments of the present invention have the following advantages:

[0035] The distribution network station-user relationship identification method provided by the embodiment of the present invention obtains the distribution network distribution transformer voltage data and the low-voltage user voltage data, and after data cleaning and data reconstruction of the distribution transformer voltage data and the low-voltage user voltage data, calculates the correlation coefficient between the low-voltage user and the ledger transformer, and judges whether the station-user relationship of the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. It does not require manual identification, nor does it require reliance on special identification equipment to use carrier communication method or pulse current method for identification. It has a high degree of intelligence and solves the technical problem that the traditional manual identification and the method of using special station area identification equipment to identify the station-user relationship have low intelligence and cannot meet the intelligent development needs of the distribution network station area. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic flow chart of a method for identifying station-user relationships in a distribution network provided in an embodiment of the present invention;

[0037] Figure 2 Another flowchart of a method for identifying station-user relationships in a distribution network provided in an embodiment of the present invention;

[0038] Figure 3 The figure is a structural diagram of a distribution network station-user relationship identification system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0040] For easier understanding, see Figure 1 The present invention provides an embodiment of a method for identifying the relationship between distribution network stations and users, comprising the following steps:

[0041] S1. Obtain distribution network transformer voltage data and low-voltage user voltage data;

[0042] S2. Clean the distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes normalizing data outliers and filling in missing data values.

[0043] S3, reconstructing the distribution transformer voltage data and the low-voltage user voltage data after data cleaning to obtain a distribution transformer voltage time series and a low-voltage user voltage time series that meet a preset length;

[0044] S4. Calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series;

[0045] S5. Determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than a first threshold, the ledger's transformer-household relationship is correct.

[0046] It should be noted that, in the embodiment of the present invention, the distribution network distribution transformer voltage data and the low-voltage user voltage data are first obtained, and then the distribution network distribution transformer voltage data and the low-voltage user voltage data are cleaned. On the one hand, data cleaning includes data outlier normalization, normalizing the abnormal data to between its upper and lower limits, and on the other hand, filling in missing data values, filling in the gaps according to the distribution transformer data corresponding to the ledger. Data cleaning can also include normalization and principal component analysis of the data. Taking into account the inertia of low-voltage users' electricity consumption, the variance of the mean of the power data generally does not change significantly, and when distance is needed to measure similarity in classification and clustering algorithms, or when principal component analysis technology is used for dimensionality reduction, Z-score standardization has better performance. Therefore, Z-score standardization is used for normalization, that is:

[0047]

[0048] Among them, x * is the sample value after normalization, x is the sample value before normalization, μ is the mean of all sample data, and σ is the standard deviation of all sample data.

[0049] To address the problem of large data dimensions and resulting large computational complexity, dimensionality reduction technology can be used to reduce multidimensional voltage data into a few principal components for analysis. The reduced components can well express most of the information in the original multidimensional data. The specific process of principal component analysis is as follows:

[0050] (1) The voltage data of the low-voltage side of the transformer in the substation and the voltage data of the low-voltage user meter are combined into a matrix:

[0051]

[0052] in, Indicates the voltage data of the 1st to 1st transformers, Indicates the voltage data of the 1st to mth low-voltage users.

[0053] (2) Decentralize all voltage data:

[0054]

[0055] Among them, X' is the new sample matrix composed of all decentralized voltage data, X id is the d-th voltage data of the i-th type of voltage data.

[0056] (3) Calculate the covariance matrix C of the sample:

[0057] C=X'X' T

[0058] (4) Perform eigenvalue decomposition on the covariance matrix C.

[0059] (5) Take out the eigenvectors (w1,...w n' ), all eigenvectors are normalized to form the eigenvector matrix W.

[0060] (6) For each sample X in the sample set, transform it into a new sample Z = W T X, get the matrix after dimensionality reduction Among them, Z f =[z f1 ,z f2 ,...,z fn ] T ,f∈[1,l]∪[1,m],z f1 ~z fn is the element that represents the voltage value after dimensionality reduction.

[0061] Each row vector in the matrix Z can well express most of the information of the corresponding row vectors in the original matrix X (transformer and user meter voltage data). That is, the Z matrix can be used for classification analysis to realize user identification in the substation area.

[0062] After cleaning, the data is reconstructed, i.e., data segmentation and number segmentation. For each distribution transformer and each low-voltage user, hourly data is used. Based on 30 days per month and 24 points per day, a voltage time series with a length of 720 is obtained for each data indicator.

[0063] The correlation coefficient between low-voltage users and distribution transformers is calculated based on the distribution transformer voltage time series and the low-voltage user voltage time series. The Pearson correlation coefficient is used and the calculation formula is:

[0064]

[0065] Among them, Rij is the correlation coefficient between the voltage data of the i-th user and the voltage data of the j-th user, x i is the voltage vector of the i-th user, x j is the voltage vector of the jth user, is a unit row vector.

[0066] After calculating the correlation coefficient between low-voltage users and distribution transformers, a loose criterion is used to determine the correctness of the user-user relationship in the ledger. The loose criterion is: if the correlation coefficient between the low-voltage user and the transformer in the ledger is not less than a first threshold, the user-user relationship in the ledger is correct. The first threshold is 0.8.

[0067] The distribution network station-user relationship identification method provided by the embodiment of the present invention obtains the distribution network distribution transformer voltage data and the low-voltage user voltage data, and after data cleaning and data reconstruction of the distribution transformer voltage data and the low-voltage user voltage data, calculates the correlation coefficient between the low-voltage user and the ledger transformer, and judges whether the station-user relationship of the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. It does not require manual identification, nor does it require reliance on special identification equipment to use carrier communication method or pulse current method for identification. It has a high degree of intelligence and solves the technical problem that the traditional manual identification and the method of using special station area identification equipment to identify the station-user relationship have low intelligence and cannot meet the intelligent development needs of the distribution network station area.

[0068] See also Figure 2 In one embodiment, step S5 may further include:

[0069] S6. If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second relevant transformer by more than the preset difference. At the same time, the most relevant transformer and the ledger transformer are consistent. If so, the ledger's station-household relationship is correct.

[0070] After step S5, if the correlation coefficient between the low-voltage user and the ledger transformer is not less than the first threshold, it can be determined that the ledger's household relationship is correct. However, if the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, the correlation coefficient criterion can be used for further judgment, that is, when the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold (taken as 0.7), and is greater than the correlation coefficient of the second relevant transformer by more than a preset difference (taken as 0.08), and the most relevant transformer is consistent with the ledger transformer, then the ledger's household relationship is considered correct.

[0071] See also Figure 2 In one embodiment, after step S6, the following steps may also be included:

[0072] S7. If the correlation coefficient between the low-voltage user and the most relevant transformer is not greater than the second threshold, or is less than the preset difference greater than the correlation coefficient of the second most relevant transformer, or the most relevant transformer is inconsistent with the ledger transformer, then calculate the voltage error caused by the low-voltage user being attributed to a specific transformer, select the transformer with the smallest error as the calculation result, and determine whether the transformer with the smallest error is consistent with the ledger transformer. If so, the transformer-to-household relationship in the ledger is correct.

[0073] After step S6, if the transformer with the smallest error is consistent with the transformer in the ledger, it can be determined that the transformer-household relationship in the ledger is correct. However, if the transformer with the smallest error is inconsistent with the transformer in the ledger, the error criterion can be used for further judgment, that is, the voltage error caused by the low-voltage user being attributed to a specific transformer is calculated, and the transformer with the smallest error is selected as the calculation result. It is judged whether the transformer with the smallest error is consistent with the transformer in the ledger. If so, the transformer-household relationship in the ledger is correct.

[0074] The voltage error caused by the low-voltage user being attributed to a specific transformer is calculated based on the distribution transformer voltage time series and the low-voltage user voltage time series. The calculation formula is:

[0075]

[0076] Among them, e ij is the error between the i-th voltage time series and the j-th voltage time series, x i is the voltage time series of the i-th low-voltage user, y j is the j-th distribution transformer voltage time series, diag(x i ) is, and the elements on the diagonal are x i , a diagonal matrix whose off-diagonal elements are all 0.

[0077] See also Figure 2 In one embodiment, after step S7, the following steps may also be included:

[0078] S8. If the transformer with the smallest error in step S7 is inconsistent with the transformer in the ledger, calculate the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-to-user relationship has not been determined, and select the transformer corresponding to the nearest area. If the transformer corresponding to the nearest area is consistent with the transformer in the ledger, the transformer-to-user relationship in the ledger is correct. Among them, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-to-user relationship has not been determined is defined as 1 minus the correlation coefficient between the low-voltage user and the transformer in the area confirmed by the ledger.

[0079] After step S7, if the transformer corresponding to the nearest area is consistent with the transformer in the ledger, it can be determined that the station-household relationship in the ledger is correct. However, if the transformer corresponding to the nearest area is inconsistent with the transformer in the ledger, a distance criterion can be used for further judgment, that is, the distance between the area where the correct ledger has been confirmed and the low-voltage user whose station-household relationship has not been determined (referring to the user for whom the station-household relationship in the ledger cannot be determined to be correct after steps S5, S6 and S7) is calculated, and the transformer corresponding to the nearest area is selected. If the transformer corresponding to the nearest area is consistent with the transformer in the ledger, the station-household relationship in the ledger is correct.

[0080] See also Figure 2 In one embodiment, after step S8, the following steps may also be included:

[0081] S9. If the transformer corresponding to the nearest area in step S8 is inconsistent with the transformer in the ledger, the correlation coefficient between low-voltage users and low-voltage users is calculated, and all confirmed users whose correlation coefficient with undetermined users is greater than the first threshold are screened. Each confirmed user votes for his or her own transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct.

[0082] After step S8, if the transformer corresponding to the nearest area is consistent with the ledger transformer, it can be determined that the ledger's transformer-household relationship is correct. However, if the transformer corresponding to the nearest area is inconsistent with the ledger transformer, a hand-in-hand criterion can be used for further judgment, that is, all confirmed users whose correlation coefficient with the undetermined users is greater than the first threshold are screened, and each confirmed user votes for his or her own subordinate transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the ledger transformer, the ledger's transformer-household relationship is correct.

[0083] See also Figure 2 In one embodiment, after step S9, the following steps may also be included:

[0084] S10. If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, select the transformer with the largest correlation coefficient. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's transformer-user relationship is correct.

[0085] After step S9, if the transformer with the highest vote is consistent with the ledger transformer, it can be determined that the ledger's transformer-household relationship is correct. However, if the transformer with the highest vote is inconsistent with the ledger transformer, a single-handed criterion can be used for further judgment, that is, for each undetermined user, the transformer with the largest correlation coefficient is selected. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's transformer-household relationship is correct.

[0086] The distribution network station-user relationship identification method provided in the present invention provides 6 criteria for judgment. The 6 criteria are used in a sequential judgment manner. The order of the criteria is: 1. Loose criterion, 2. Correlation coefficient criterion, 3. Error criterion, 4. Distance criterion, 5. Hand-in-hand criterion, 6. Single-hand criterion. Those skilled in the art can select the number of criteria in sequence according to actual needs. As long as the current criterion is met, the judgment is stopped and the result is returned: the ledger is correct; if the current criterion is not met, the judgment is continued; if the last criterion is reached and the ledger is still not correct, the ledger is judged to be wrong, and the result is returned: the ledger is wrong. The station-user relationship identification based on multi-dimensional criteria can reduce the sample size requirements on the basis of ensuring the accuracy of the algorithm, and obtain prediction and analysis results that are more in line with the actual project.

[0087] For easier understanding, see Figure 3 The present invention provides an embodiment of a distribution network station-user relationship identification system, including the following modules:

[0088] The data acquisition module 301 is used to obtain the distribution network distribution transformer voltage data and the low voltage user voltage data;

[0089] Data cleaning module 302 is used to clean distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes normalizing data outliers and filling missing values. Data cleaning also includes normalization and principal component analysis, with normalization using Z-score standardization.

[0090] The data reconstruction module 303 is used to reconstruct the distribution transformer voltage data and the low-voltage user voltage data after data cleaning to obtain the distribution transformer voltage time series and the low-voltage user voltage time series that meet the preset length;

[0091] The correlation coefficient calculation module 304 is used to calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series. The correlation coefficient is the Pearson correlation coefficient.

[0092] The station-user relationship identification module 305 is used to determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than the first threshold, the station-user relationship of the ledger is correct.

[0093] The platform-user relationship identification module 305 is also used to:

[0094] If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, then determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second relevant transformer by more than the preset difference. At the same time, the most relevant transformer and the ledger transformer are consistent. If so, the ledger's station-household relationship is correct.

[0095] The platform-user relationship identification module 305 is also used to:

[0096] If the correlation coefficient between the low-voltage user and the most relevant transformer is not greater than the second threshold, or is less than the preset difference greater than the correlation coefficient of the second most relevant transformer, or the most relevant transformer is inconsistent with the ledger transformer, then the voltage error caused by the low-voltage user being attributed to a specific transformer is calculated, and the transformer with the smallest error is selected as the calculation result. It is determined whether the transformer with the smallest error is consistent with the ledger transformer. If so, the transformer-to-household relationship in the ledger is correct.

[0097] The platform-user relationship identification module 305 is also used to:

[0098] If the transformer with the smallest error is inconsistent with the transformer in the ledger, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-user relationship has not been determined is calculated, and the transformer corresponding to the nearest area is selected. If the transformer corresponding to the nearest area is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct. Among them, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose transformer-user relationship has not been determined is defined as 1 minus the correlation coefficient between the low-voltage user and the transformer in the area confirmed by the ledger.

[0099] The platform-user relationship identification module 305 is also used to:

[0100] If the transformer corresponding to the nearest area is inconsistent with the transformer in the ledger, the correlation coefficient between the low-voltage users is calculated, and all confirmed users whose correlation coefficient with the undetermined users is greater than the first threshold are screened. Each confirmed user votes for his or her affiliated transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct.

[0101] The platform-user relationship identification module 305 is also used to:

[0102] If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, the transformer with the largest correlation coefficient is selected. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's platform-user relationship is correct.

[0103] The distribution network station-user relationship identification system provided by the embodiment of the present invention obtains the distribution network distribution transformer voltage data and the low-voltage user voltage data, and after data cleaning and data reconstruction of the distribution transformer voltage data and the low-voltage user voltage data, calculates the correlation coefficient between the low-voltage user and the ledger transformer, and judges whether the station-user relationship in the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. It does not require manual identification, nor does it require reliance on dedicated identification equipment to use carrier communication method or pulse current method for identification. It has a high degree of intelligence and solves the technical problem that traditional manual identification and the use of dedicated station area identification equipment for station-user relationship identification methods have low intelligence and cannot meet the intelligent development needs of distribution network stations.

[0104] The distribution network station-customer relationship identification system provided in the embodiment of the present invention is used to execute the distribution network station-customer relationship identification method in the aforementioned embodiment. Its working principle is the same as the distribution network station-customer relationship identification method in the aforementioned embodiment, and can achieve the same technical effect, which will not be repeated here.

[0105] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying the relationship between distribution network stations and users, characterized in that: The following steps are involved: S1. Obtain distribution network transformer voltage data and low-voltage user voltage data; S2. Clean the distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes normalizing data outliers and filling in missing data values. S3, reconstructing the distribution transformer voltage data and the low-voltage user voltage data after data cleaning to obtain a distribution transformer voltage time series and a low-voltage user voltage time series that meet a preset length; S4. Calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series; S5. Determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than a first threshold, the ledger's transformer-household relationship is correct. After step S5, the following steps are also included: S6. If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second most relevant transformer by at least a preset difference. If the most relevant transformer is consistent with the ledger transformer, then the ledger's transformer-to-user relationship is correct. After step S6, the following steps are also included: S7. If the correlation coefficient between the low-voltage user and the most relevant transformer is not greater than the second threshold, or is not greater than the correlation coefficient of the second most relevant transformer by more than a preset difference, or the most relevant transformer is inconsistent with the recorded transformer, then the voltage error caused by attributing the low-voltage user to a specific transformer is calculated, and the transformer with the smallest error is selected as the calculation result. It is then determined whether the transformer with the smallest error is consistent with the recorded transformer. If so, the transformer-to-user relationship in the recorded transformer is correct. After step S7, the following steps are also included: S8. If the transformer with the smallest error in step S7 is inconsistent with the ledger transformer, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose relationship with the user is not determined is calculated, and the transformer corresponding to the nearest area is selected. If the transformer corresponding to the nearest area is consistent with the ledger transformer, the user relationship of the ledger is correct. The distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose relationship with the user is not determined is defined as 1 minus the correlation coefficient between the low-voltage user and the transformer in the area where the ledger is confirmed; After step S8, the following steps are also included: S9. If the transformer corresponding to the nearest area in step S8 is inconsistent with the transformer in the ledger, the correlation coefficient between low-voltage users and low-voltage users is calculated, and all confirmed users whose correlation coefficient with undetermined users is greater than the first threshold are screened. Each confirmed user votes for his or her affiliated transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct. After step S9, the following steps are also included: S10. If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, select the transformer with the largest correlation coefficient. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's transformer-user relationship is correct.

2. The method for identifying the relationship between distribution network stations and users according to claim 1, characterized in that: In step S4, the correlation coefficient between the low-voltage user and the distribution transformer is the Pearson correlation coefficient.

3. The method for identifying the relationship between distribution network stations and users according to claim 1, characterized in that: In step S2, data cleaning also includes normalization processing and principal component analysis processing; Normalization was performed using Z-score standardization.

4. A distribution network station-user relationship identification system, characterized in that: Includes the following modules: Data acquisition module, used to obtain distribution network distribution transformer voltage data and low-voltage user voltage data; Data cleaning module, used to clean distribution transformer voltage data and low-voltage user voltage data. Data cleaning includes data outlier standardization and data missing value filling. The data reconstruction module is used to reconstruct the distribution transformer voltage data and low-voltage user voltage data after data cleaning to obtain the distribution transformer voltage time series and low-voltage user voltage time series that meet the preset length; A correlation coefficient calculation module is used to calculate the correlation coefficient between the low-voltage user and the distribution transformer based on the distribution transformer voltage time series and the low-voltage user voltage time series; The station-household relationship identification module is used to determine whether the ledger is correct based on the correlation coefficient between the low-voltage user and the ledger transformer. If the correlation coefficient between the low-voltage user and the ledger transformer is not less than a first threshold, the station-household relationship of the ledger is correct; The platform-user relationship identification module is also used to: If the correlation coefficient between the low-voltage user and the ledger transformer is less than the first threshold, then determine whether the correlation coefficient between the low-voltage user and the most relevant transformer is greater than the second threshold, and is greater than the correlation coefficient of the second relevant transformer by more than the preset difference. At the same time, the most relevant transformer is consistent with the ledger transformer. If so, the ledger's transformer-household relationship is correct. If the correlation coefficient between the low-voltage user and the most relevant transformer is not greater than the second threshold, or is not greater than the correlation coefficient of the second most relevant transformer by more than the preset difference, or the most relevant transformer is inconsistent with the ledger transformer, then the voltage error caused by the low-voltage user being attributed to a specific transformer is calculated, and the transformer with the smallest error is selected as the calculation result. It is determined whether the transformer with the smallest error is consistent with the ledger transformer. If so, the transformer-to-household relationship in the ledger is correct. If the transformer with the smallest error is inconsistent with the transformer in the ledger, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose relationship with the user has not been determined is calculated, and the transformer corresponding to the nearest area is selected. If the transformer corresponding to the nearest area is consistent with the transformer in the ledger, the user relationship in the ledger is correct. Among them, the distance between the area where the ledger has been confirmed to be correct and the low-voltage user whose relationship with the user has not been determined is defined as 1 minus the correlation coefficient between the low-voltage user and the transformer in the area confirmed by the ledger; If the transformer corresponding to the nearest area is inconsistent with the transformer in the ledger, the correlation coefficient between low-voltage users and low-voltage users is calculated, and all confirmed users whose correlation coefficient with undetermined users is greater than the first threshold are screened. Each confirmed user votes for his or her affiliated transformer, and the transformer with the highest vote wins. If the transformer with the highest vote is consistent with the transformer in the ledger, the transformer-user relationship in the ledger is correct. If the transformer with the highest vote is inconsistent with the ledger transformer, then for each undetermined user, the transformer with the largest correlation coefficient is selected. If the transformer with the largest correlation coefficient is consistent with the ledger transformer, the ledger's platform-user relationship is correct.

Images