Power distribution network line loss reduction calculation method and system based on clustering algorithm

Abstract

The invention discloses a power distribution network line loss reduction calculation method and system based on a clustering algorithm. The method comprises the following steps: acquiring and processing initial data including a topological structure, a distribution transformer ledger, a meter bottom reading, power, voltage and switch displacement data; thirdly, identifying a transfer line through feature index extraction and a threshold value method, and determining an alternative supply line and a transfer time period; and for a to-be-diagnosed line, constructing a voltage similarity matrix by using a dynamic time warping algorithm, and identifying a transfer distribution transformer set by using density clustering. And then, a transfer distribution transformer set and a time period thereof are corrected in combination with a switch section proportion index. And finally, according to the optimized data, calculating and restoring the same-period line loss of the transfer supply line and the substitute supply line. By implementing the method provided by the invention, the accuracy of the synchronous line loss rate of the transfer line can be improved, and accurate calculation and reduction of the synchronous line loss rate can be realized, so that more effective power system management and energy-saving strategy implementation can be supported.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network, and particularly relates to a power distribution network line loss restoration calculation method and system based on a clustering algorithm. BACKGROUND

[0002] In the operation and management of modern power distribution networks, the line contemporaneous line loss rate is a key indicator for evaluating the operation status of the line. It not only reflects the loss in the process of power transmission, but also provides an important basis for optimizing the operation efficiency of the power system.

[0003] However, the current technology faces several significant challenges in dealing with contemporaneous line loss restoration calculation involving transfer lines. First, traditional transfer identification methods mainly rely on basic parameters such as power surges to make judgments. This method often shows a high misjudgment rate for cases with small load fluctuations, such as power transfer between some distribution transformers. This is because these traditional methods fail to fully utilize the transfer line loss rate, which reflects the matching relationship between load and power supply. The existing technical framework does not fully consider the time line loss rate and the combined transfer line loss rate, which are important indicators that can effectively demonstrate the characteristics of power load changes and the overall power supply efficiency after load transfer. Specifically, the time line loss rate can reveal the load characteristics of the line over time, while the combined transfer line loss rate can clearly reflect the overall power supply efficiency after power transfer. Due to the lack of application of these indicators, existing transfer identification methods lack important judgment dimensions. The current contemporaneous line loss rate calculation method is mainly based on static account information of equipment, without considering the dynamic changes of the distribution transformer set and the transfer period during the actual transfer process. In this case, the calculated contemporaneous line loss rate has a large deviation from the actual value, which cannot meet the needs of power companies in line loss assessment and energy saving optimization.

[0004] Therefore, it is necessary to design a new method to improve the accuracy of the contemporaneous line loss rate of the transfer line, and to accurately calculate and restore the contemporaneous line loss rate, in order to support more effective power system management and energy saving strategy implementation. SUMMARY

[0005] The present application aims to overcome the shortcomings of the prior art and provide a power distribution network line loss restoration calculation method and system based on a clustering algorithm.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solution: a power distribution network line loss restoration calculation method based on a clustering algorithm, comprising: Obtaining the topological structure of all lines, the distribution transformer account, the association relationship of lines and tie switches, and obtaining the table bottom data, power, voltage data and switch position data of all lines and the distribution transformers under the jurisdiction of the lines, to obtain initial data; Extract feature indicators from the initial data, and use index threshold method to identify line switching, define the line with switching as the to-be-diagnosed line, to obtain the switching line, the supplied line and the switching period; For the to-be-diagnosed line, calculate the voltage similarity matrix between all distribution transformer switching voltage curves in the switching period of the to-be-diagnosed line, and then cluster the voltage similarity matrix, and combine each cluster category with the voltage similarity index judgment of the supplied line to identify the switching distribution transformer set of the to-be-diagnosed line, to obtain the preliminary switching distribution transformer set; Based on the preliminary switching distribution transformer set, calculate the switching distribution transformer switch section proportion, and the total distribution transformer proportion index of the switching distribution transformer, and combine the switch variable position data to locate the upstream section switch and the downstream contact switch of the switching distribution transformer set, to correct the preliminary switching distribution transformer set and the switching period, to obtain the final switching distribution transformer set and the final switching period. Based on the final switching distribution transformer set and the final switching period, according to the table bottom data of the line and the distribution transformer, restore the synchronization line loss rate of the switching line and the supplied line.

[0007] Further technical solutions are as follows: the initial data is extracted to obtain feature indicators, and index threshold method is used to identify line switching, and the line with switching is defined as the to-be-diagnosed line, to obtain the switching line, the supplied line and the switching period, including: Based on the initial data, calculate the time line loss rate of all lines; When the time line loss rate of the line is negative for a certain duration, determine the information of all opposite lines connected to the line, and filter out the opposite line information whose time line loss rate is greater than a set threshold in the negative line loss period of the line, to obtain a filtering result; Calculate the switching line loss rate of the line and the opposite line in the filtering result in the negative line loss period of the line; When the switching line loss rate meets the requirements, the line is determined as the to-be-diagnosed line, and the opposite line is determined as the supplied line, and the negative line loss period is determined as the switching period, to obtain the switching line, the supplied line and the switching period.

[0008] Further technical solutions are as follows: for the to-be-diagnosed line, calculate the voltage similarity matrix between all distribution transformer switching voltage curves in the switching period of the to-be-diagnosed line, and then cluster the voltage similarity matrix, and combine each cluster category with the voltage similarity index judgment of the supplied line to identify the switching distribution transformer set of the to-be-diagnosed line, to obtain the preliminary switching distribution transformer set, including: The dynamic time warping algorithm is used to calculate a voltage similarity matrix between voltage curves of all distribution transformer transfer periods on the to-be-diagnosed line, and the density clustering algorithm is used to cluster the voltage similarity matrix, and a preliminary transfer distribution transformer set is obtained by combining each clustering category with a voltage similarity index judgment of the transfer line.

[0009] Further technical solutions are as follows: the dynamic time warping algorithm is used to calculate a voltage similarity matrix between voltage curves of all distribution transformer transfer periods on the to-be-diagnosed line, and the density clustering algorithm is used to cluster the voltage similarity matrix, and a preliminary transfer distribution transformer set is obtained by combining each clustering category with a voltage similarity index judgment of the transfer line, including: For the to-be-diagnosed line, voltage measurement data of the transfer line and the connected distribution transformer in the transfer period are extracted, and the voltage measurement data after sorting according to the measurement sending time is smoothed by using the moving translation method to eliminate voltage fluctuation noise, to form a smoothed voltage curve, and the DTW distance between the two smoothed voltage curves of each distribution transformer and the line is calculated by using the Manhattan distance; The DTW distance is converted into a similarity index, and a voltage similarity matrix is constructed; The DBSCAN clustering algorithm is used to cluster the similarity matrix, to obtain a plurality of clustering categories, the voltage curve of the transfer line in the transfer period is extracted, the correlation coefficient of all distribution transformer voltage curves in each category and the voltage curve of the transfer line is calculated and averaged as the category correlation coefficient, and the category with the maximum category correlation coefficient is selected, and the distribution transformers in the category are defined as the preliminary transfer distribution transformer set.

[0010] Further technical solutions are as follows: the calculation formula of the DTW distance is , wherein L is the warping path length, xm and ym are the sampling point indexes of the smoothed voltage curves , and d( ) is the Manhattan distance. The calculation formula of the similarity index is: Sij=exp(-α× ); wherein a is an adjustment coefficient, and the closer Sij is to 1, the higher the similarity of the two curves is. The voltage similarity matrix is M , wherein M is the total number of distribution transformers under the to-be-diagnosed line.

[0011] Further technical solutions are as follows: based on the preliminary transfer supply distribution transformer set, a transfer supply distribution transformer switch section distribution transformer ratio and a switch section total distribution transformer ratio are calculated, upstream section switches and downstream tie switches of the transfer supply distribution transformer set are located in combination with switch transformer position data, the preliminary transfer supply distribution transformer set and the transfer supply time period are corrected to obtain a final transfer supply distribution transformer set and a final transfer supply time period, including: All tie information of the to-be-diagnosed line is acquired, all section switches are found as a section switch list from each tie switch as a last child node upwards; The section switch list is cycled, the total number of distribution transformers hung downstream of each switch is counted as a section total distribution transformer number, the number of distribution transformers hung downstream of each switch existing in the preliminary transfer supply distribution transformer set is counted as a preliminary transfer supply distribution transformer number, the total number of the preliminary transfer supply distribution transformers is counted as a preliminary transfer supply distribution transformer total number, a transfer supply distribution transformer switch section distribution transformer ratio and a section total distribution transformer ratio of each switch in the section switch list are calculated, and a section switch and a corresponding tie switch meeting a requirement are selected according to a sum of the two indexes to obtain a target section switch and a corresponding tie switch; Switch transformer position data corresponding to all tie information of the to-be-diagnosed line is acquired; When the switch transformer position data exists in the tie switch corresponding to the target section switch, the corresponding tie switch is selected as a final transfer supply information group, and when the switch transformer position data does not exist in the tie switch corresponding to the target section switch, the first-ranked tie switch is selected as the final transfer supply information group; The section switch in the final transfer supply information group is extracted as a transfer supply section switch, distribution transformers hung between the transfer supply section switch and the tie switch downstream of the transfer supply section switch are taken as a final transfer supply distribution transformer set, and a final transfer supply time period is determined.

[0012] Further technical solutions are as follows: based on the final transfer supply distribution transformer set and the final transfer supply time period, according to line and distribution transformer table bottom data, synchronous line loss rates of the transfer supply line and the substitute supply line are restored and calculated, including: Based on table bottom data of all lines and distribution transformers under the jurisdiction of the lines, in combination with the final transfer supply distribution transformer set and the final transfer supply time period, line forward electric quantity and line reverse electric quantity of the transfer supply line and the substitute supply line, original distribution transformer forward electric quantity and original distribution transformer reverse electric quantity corresponding to the two lines, transfer supply distribution transformer forward electric quantity and transfer supply distribution transformer reverse electric quantity data are calculated respectively to obtain intermediate calculation results; According to the intermediate calculation results, input electric quantity, output electric quantity and sold electric quantity of the transfer supply line and the substitute supply line are calculated, and synchronous line loss rates of the transfer supply line and the substitute supply line are calculated.

[0013] The application also provides a power distribution network line loss restoration calculation system based on a clustering algorithm, including: An acquisition unit is configured to acquire the topological structure of all lines, the transformer account, the line and contact switch association relationship, and acquire the table bottom data, power, voltage data and switch position data of all lines and the lines under jurisdiction to obtain initial data; An extraction unit is configured to extract feature indicators from the initial data, identify line switching by using an indicator threshold method, define the line with switching as a to-be-diagnosed line, and obtain a switching line, a substitute line and a switching period; An identification unit is configured to calculate a voltage similarity matrix between voltage curves of all transformers on the to-be-diagnosed line in the switching period, and perform clustering on the voltage similarity matrix, and identify a switching transformer set of the to-be-diagnosed line by combining each clustering category with a voltage similarity indicator of the substitute line to obtain a preliminary switching transformer set; A correction unit is configured to calculate a switching transformer proportion of a switch section, a total transformer proportion of a section and a switching transformer proportion indicator based on the preliminary switching transformer set, locate an upstream section switch and a downstream contact switch of the switching transformer set by combining switch position data, and correct the preliminary switching transformer set and the switching period to obtain a final switching transformer set and a final switching period; A restoration calculation unit is configured to restore calculate the same period line loss rate of the switching line and the substitute line according to the table bottom data of the line and the transformer based on the final switching transformer set and the final switching period.

[0014] Further technical solutions are as follows: the extraction unit comprises: A first calculation subunit is configured to calculate the time point line loss rate of all lines based on the initial data; A screening subunit is configured to determine all opposite line information associated with the line when the time point line loss rate of the line is negative loss for a certain time length, and screen out opposite line information with a time point line loss rate greater than a set threshold in the negative loss period of the line to obtain a screening result; A second calculation subunit is configured to calculate the switching line loss rate of the line and the opposite line in the screening result in the negative loss period of the line; A determination subunit is configured to determine the line as a to-be-diagnosed line and the opposite line as a substitute line when the switching line loss rate meets the requirements, and determine the negative loss period as a switching period to obtain a switching line, a substitute line and a switching period.

[0015] A further technical scheme is that the screening unit is used for calculating a voltage similarity matrix between voltage curves of all distribution transformer transfer periods on the to-be-diagnosed line by using a dynamic time warping algorithm, clustering the voltage similarity matrix by using a density clustering algorithm, and identifying a transfer distribution transformer set of the to-be-diagnosed line by combining each clustering category with a voltage similarity index judgment of the transfer line to obtain a preliminary transfer distribution transformer set.

[0016] Compared with the prior art, the present application has the following beneficial effects: the present application obtains detailed topological structures of all related lines, distribution transformer account books, contact switch correlation relationships and key operation parameters as initial data. Then, the initial data are subjected to feature index extraction, and an index threshold method is used to identify a line where transfer occurs, i.e., a to-be-diagnosed line. Next, the voltage similarity matrix of the distribution transformer of the to-be-diagnosed line at the transfer period is calculated and subjected to clustering analysis to identify a preliminary transfer distribution transformer set. Further, based on the preliminary transfer distribution transformer set, a specific proportion index is calculated and combined with switch position data to accurately locate the upstream and downstream key switch positions of the transfer distribution transformer set, so that the preliminary result is corrected to obtain a final transfer distribution transformer set and a corresponding transfer period. Finally, according to the table bottom data of the line and the distribution transformer, the contemporaneous line loss rate of the transfer line and the transfer line is accurately calculated and restored to provide strong data support and technical support for the optimized management of the power system. This method effectively improves the accuracy and reliability of the contemporaneous line loss rate of the transfer line in the power system, and helps to develop more scientific and reasonable energy-saving strategies.

[0017] The application will be further described below in conjunction with the drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0019] Figure 1 A flowchart of the power distribution network line loss restoration calculation method based on the clustering algorithm provided by the embodiment of the present application is shown. Figure 2 A schematic block diagram of the power distribution network line loss restoration calculation system based on the clustering algorithm provided by the embodiment of the present application is shown. Figure 3 A schematic block diagram of the computer device provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0020] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be clearly and completely described in the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of the present application.

[0021] It should be understood that the terms "include" and "comprise" as used in the specification and the appended claims indicate the presence of the described features, integers, steps, operations, elements, and / or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0022] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0023] It should be further understood that the term "and / or" as used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations thereof.

[0024] Please refer to Figure 1 , Figure 1 The flowchart of the power distribution network line loss restoration calculation method based on a clustering algorithm provided by the embodiments of the present application. The power distribution network line loss restoration calculation method based on a clustering algorithm is applied to a server, accurately identifies and calculates the contemporaneous line loss rate of the transfer supply line in the power distribution network, thereby supporting more effective power system management and energy saving strategy implementation. First, the initial data is obtained and processed, including the topological structure, the distribution transformer account, etc., and the line transfer supply identification is performed by using the index threshold method, and the line to be diagnosed and its time period are determined. Second, the dynamic time warping algorithm and the density clustering algorithm are used to construct the voltage similarity matrix to identify the preliminary transfer supply distribution transformer set. Then, the transfer supply distribution transformer set and its time period are corrected in combination with the switch segmentation proportion index, and the final transfer supply distribution transformer set and time period are further located. Finally, the contemporaneous line loss rate of the transfer supply line and the substitute supply line is restored and calculated based on the optimized data. This method significantly improves the accuracy of the contemporaneous line loss rate of the transfer supply line and the precision of the calculation by fine data processing and advanced algorithm application, and provides a solid foundation for effective management of the power system.

[0025] Figure 1 The flowchart of the power distribution network line loss restoration calculation method based on a clustering algorithm provided by the embodiments of the present application. As Figure 1As shown, the method includes the following steps S110 to S150.

[0026] S110, acquire the topology of all lines, transformer account, line and contact switch association, and acquire the table bottom data of all lines and the lines under jurisdiction transformer, power, voltage data and switch variable position data, to get the initial data.

[0027] In this embodiment, the initial data refers to a series of key information collected through a specific data acquisition process, which is crucial for subsequent analysis and calculation of the same period line loss rate of lines in the distribution network. Specifically, the initial data includes the following aspects: Transformer account data: This includes detailed equipment information such as transformer ID, name, operation status, line, type, resource ID, asset ID, city, and operation and maintenance unit. These information provides a basis for identifying and classifying each transformer in the power system.

[0028] Table bottom data: Acquire the positive and negative daily frozen table bottom data of all line head measurement tables and the measurement tables of the transformers under jurisdiction (including the data of the current day and the previous day), as well as the 15-minute sampling interval of the transformer measurement table positive and negative frozen table bottom data. These data are essential for calculating power consumption and generating accurate power reports.

[0029] Power and voltage data: Obtain 15-minute sampling interval power and voltage measurement data from the power information collection system and D5000 system, especially total active power and A-phase voltage data. These real-time monitoring data help analyze the operation state of the power system and provide necessary input for line loss calculation.

[0030] Switch variable position data: Obtain detailed information about sectionalizing switches and contact switches from the D5200 system, including switch ID, name, type, operation status, variable position time, and variable position type (opening / closing). This kind of data is important for understanding the dynamic changes of the power network and the impact of switch operation on the structure of the power grid.

[0031] Line topology: Obtain the connection position relationship between all switches and transformers hung on each line from the PMS system. Understanding this topology helps determine the location of each component in the power system and its connection mode, which is very important for analyzing fault points and optimizing power grid configuration.

[0032] In summary, obtaining the topology of all lines, distribution transformer account, and the association between lines and tie switches refers to collecting the physical connection and logical relationship data between all switches and distribution transformers connected on each line from the power management system (PMS), including but not limited to the basic account information of each distribution transformer (such as distribution transformer ID, name, operation status, etc.), the membership relationship between each line and its subordinate distribution transformer, and the detailed account information of all sectional switches and tie switches (including switch ID, name, type, etc.), and establishing a comprehensive model reflecting the accurate connection relationship and operation state of the power grid equipment through these information, so as to facilitate subsequent analysis of the operation efficiency of the power system and optimization of the power grid configuration. This process ensures a comprehensive understanding of the power grid structure and provides necessary data support for accurate line loss calculation and other power grid management activities, and the purpose is to build a comprehensive and detailed initial data set for subsequent accurate line loss calculation and power grid management. These data constitute the basis for in-depth analysis, support more effective power system management and implementation of energy-saving strategies.

[0033] S120, extracting feature indicators from the initial data, and using index threshold method to identify line transfer supply, defining the line where transfer supply occurs as the line to be diagnosed to obtain transfer supply line, substitute supply line and transfer supply period.

[0034] In this embodiment, step S120 aims to determine whether there is line transfer supply in the power system through a series of calculation and analysis methods. The process includes extracting key feature indicators from the original monitoring data and applying pre-set threshold standards to determine whether transfer supply has occurred. Specifically: Transfer supply line: refers to a line that is suspected to cause abnormal loss due to transfer supply behavior when the line has continuous negative loss (line loss rate less than 0%).

[0035] Substitute supply line: refers to other lines that are in contact with the line with continuous negative loss and show high loss (line loss rate greater than 6%) characteristics at the same time, which may be due to increased loss caused by taking over the transferred power.

[0036] Transfer supply period: refers to the time period during which the above-mentioned negative loss line and its substitute supply line have transfer supply behavior, i.e. the continuous time period during which the line loss rate is negative.

[0037] Line to be diagnosed: refers to the line that is preliminarily judged to cause abnormal line loss due to transfer supply, which needs to be further checked for its operation state and power flow.

[0038] In an embodiment, the above-mentioned step S120 can include steps S121-S124.

[0039] S121, calculating the line loss rate of all lines at the moment based on the initial data.

[0040] In the embodiment, the time point line loss rate refers to the ratio of the power loss of each line in a specific time interval calculated according to the input power, output power and sold power at the sampling point time t. It is one of the key indicators for evaluating the efficiency of power grid operation.

[0041] S122, when the time point line loss rate of the line is negative loss for a certain duration, determine all the information of the opposite end line connected with the line, and filter out the opposite end line information whose time point line loss rate in the negative loss period of the line is greater than the set threshold value to obtain a filtering result.

[0042] In the embodiment, the filtering result refers to finding those lines showing significant high loss (line loss rate exceeding 6%) characteristics in the same period by analyzing other lines connected by the tie switch. These lines are considered as potential replacement lines because they may accept power supply from the negative loss line.

[0043] S123, calculate the replacement line loss rate of the line and the opposite end line in the filtering result in the negative loss period of the line; S124, when the replacement line loss rate meets the requirements, determine the line as the line to be diagnosed and the opposite end line as the replacement line, and take the negative loss period as the replacement period to obtain the replacement line, the replacement line and the replacement period.

[0044] If the calculated replacement line loss rate is in the normal range (0%-6%), it is confirmed that the original negative loss line is the line to be diagnosed, and the corresponding high loss line is the replacement line, and the time is recorded as the replacement period to provide basis for further troubleshooting. Otherwise, if the replacement line loss rate is not in the normal range, it is considered that no effective replacement behavior occurs on the same day, and the current research and judgment process is ended.

[0045] In the embodiment, the time point line loss rate of all lines is calculated, and the calculation formula is as follows: The time point input power is calculated, and the unit is kilowatt-hour: wherein QBi represents the active power value of the line at the sampling point time t, n represents the total number of devices connected to the line, and Pi represents the active power value of device k at the sampling point time t. Here, QBi is required to be greater than 0, and Pk,i value is required to be less than 0.

[0046] The time point output power is calculated, and the unit is kilowatt-hour: wherein QBi represents the active power value of the line at the sampling point time t. Here, QBi value is required to be less than 0.

[0047] The time point sold power is calculated, and the unit is kilowatt-hour: Wherein n represents the total number of devices connected to the line, Pk,i represents the active power value of device k on the line at sampling point time t, and the value of Pk,i is required to be greater than 0.

[0048] Calculate the line loss rate at the time: .

[0049] If the line has a negative loss for more than 1 hour (defined as a line loss rate less than 0%, and a negative loss for more than 1 hour is defined as a line loss rate greater than or equal to 4 consecutive time line loss rates), find all the information of the opposite line connected to the line, calculate the time line loss rate of the opposite line using the above formula, and select the opposite line information whose time line loss rate is in the high loss (defined as a line loss rate greater than 6%) during the negative loss period of the line. Then calculate the transfer supply line loss rate between the line and the opposite line during the negative loss period of the line, and the calculation formula is as follows: Calculate the transfer supply input power, unit: kilowatt-hour: Wherein QBi(t) represents the active power value of the line at sampling point time t, QBj(t) represents the active power value of the opposite line at sampling point time t, c represents the number of 15-minute sampling points in the negative loss period of the line, n represents the total number of devices connected to the line, m represents the total number of devices connected to the opposite line, Pk,i(t) represents the active power value of device k on the line at sampling point time t, and Pk,j(t) represents the active power value of device k on the opposite line at sampling point time t. Here, the Q-related power including QBi(t) and QBj(t) is required to be greater than 0, and the P-related power including Pk,i(t) and Pk,j(t) is required to be less than 0.

[0050] Calculate the transfer supply output power, unit: kilowatt-hour: Wherein c represents the number of 15-minute sampling points in the negative loss period of the line, QBi(t) represents the active power value of the line at sampling point time t, and QBj(t) represents the active power value of the opposite line at sampling point time t. Here, the Q-related power including QBi(t) and QBj(t) is required to be less than 0.

[0051] Calculate the transfer supply power, unit: kilowatt-hour: Wherein c represents the number of 15-minute sampling points in the negative loss period of the line, n represents the total number of devices connected to the line, m represents the total number of devices connected to the opposite line, Pk,i(t) represents the active power value of device k on the line at sampling point time t, and Pk,j(t) represents the active power value of device k on the opposite line at sampling point time t. Here, the P-related power including Pk,i(t) and Pk,j(t) is required to be greater than 0.

[0052] Calculate the transfer supply line loss rate: .

[0053] If the transfer supply line loss rate is normal (between 0% and 6%), the line is considered to be a to-be-diagnosed line for transfer supply, the opposite end line is considered to be a supply line, and the to-be-diagnosed line loss period is considered to be a transfer supply period, otherwise, the to-be-diagnosed line does not occur transfer supply on the same day and directly ends the judgment.

[0054] S130, for the to-be-diagnosed line, calculating a voltage similarity matrix between voltage curves of all distribution transformer transfer supply periods on the to-be-diagnosed line, and then clustering the voltage similarity matrix, combining each cluster category with the voltage similarity index of the supply line to judge and identify the transfer supply distribution transformer set of the to-be-diagnosed line, to obtain a preliminary transfer supply distribution transformer set.

[0055] In the embodiment, step S130 aims to identify the transfer supply distribution transformer set in the to-be-diagnosed line by dynamic time warping (DTW) algorithm and density-based spatial clustering of applications with noise (DBSCAN) algorithm. Specifically: Voltage similarity matrix: refers to an M size matrix, where M is the total number of all distribution transformers (abbreviated as "distribution transformers") under the to-be-diagnosed line. Each element Sij in the matrix represents the similarity between the voltage curves of the i th distribution transformer and the j th distribution transformer, which is converted based on the DTW distance. The closer the similarity value is to 1, the more similar the two voltage curves are.

[0056] Preliminary transfer supply distribution transformer set: refers to a group of distribution transformers that are considered to have participated in the transfer supply process after the above analysis. This set is determined according to the similarity of the voltage curve and the correlation with the voltage curve of the supply line.

[0057] Specifically, for the to-be-diagnosed line, the dynamic time warping algorithm is used to calculate the voltage similarity matrix between the voltage curves of all distribution transformer transfer supply periods on the to-be-diagnosed line, and then the density-based spatial clustering of applications with noise algorithm is used to cluster the voltage similarity matrix, combined with each cluster category and the voltage similarity index of the supply line to judge and identify the transfer supply distribution transformer set of the to-be-diagnosed line, to obtain a preliminary transfer supply distribution transformer set.

[0058] In an embodiment, the above step S130 can include steps S131-S133.

[0059] S131, for the to-be-diagnosed line, extracting voltage measurement data of the transfer supply line and the connected distribution transformer within the transfer supply period, and according to the measurement uploading time sorting, using the moving average method to smooth the voltage for eliminating the voltage fluctuation noise, forming the smoothed voltage curve, and using the Manhattan distance to calculate the DTW distance between the two smoothed voltage curves of each distribution transformer and the line.

[0060] In the embodiment, the calculation formula of the DTW distance is where L is the regular path length, xm, ym are the smoothed voltage curve sample point index, d( ) is the Manhattan distance.

[0061] Firstly, the voltage measurement data of the to-be-diagnosed line and all distribution transformers connected thereto in the time period of power transfer are extracted from the monitoring system and sorted according to the measurement time. Then, the moving average method and other techniques are applied to pre-process the voltage data, so as to reduce noise interference and obtain more accurate voltage curves. Then, the DTW algorithm is used to calculate the distance between each pair of distribution transformers and the smoothed line voltage curve, so as to serve as the basis for measuring the similarity therebetween.

[0062] In S132, the DTW distance is converted into a similarity index, and a voltage similarity matrix is constructed.

[0063] In this embodiment, the calculation formula of the similarity index is: Sij=exp(-α× ); where a is an adjustment coefficient, and the closer Sij is to 1, the higher the similarity of the two curves is.

[0064] This step involves converting the DTW distance calculated in the foregoing into a similarity index between 0 and 1. Here, a is an adjustment coefficient, which is used to ensure that the similarity index falls within an appropriate range. Based on the similarity index of all pairs, a similarity matrix is constructed. Each element in the matrix represents the voltage curve similarity between a distribution transformer and another distribution transformer. Since the similarity is symmetrical, the matrix is also a symmetrical matrix, and the elements on the diagonal are all 1 (because the similarity of each distribution transformer to itself is the highest). The voltage similarity matrix constructed finally can intuitively show the voltage curve similarity degree between various distribution transformers.

[0065] In S133, the DBSCAN clustering algorithm is used to cluster the similarity matrix, a plurality of clustering categories are obtained, the voltage curve of the power supply line in the power transfer period is extracted, the correlation coefficient of all distribution transformer voltage curves in each category and the voltage curve of the power supply line is calculated and averaged as the category correlation coefficient, the category with the largest category correlation coefficient is selected, and the distribution transformers in the category are defined as the preliminary power transfer distribution transformer set.

[0066] In this embodiment, the voltage similarity matrix is M M similarity matrix, where M is the total number of distribution transformers under the to-be-diagnosed line.

[0067] In the final step, the DBSCAN algorithm is applied to the previously constructed voltage similarity matrix for clustering analysis, thereby grouping distribution transformers with similar voltage patterns into the same category. For each cluster category, the Pearson correlation coefficient between the voltage curves of the internal distribution transformers and the supply line voltage curve is further calculated, and the average value is taken as the correlation coefficient of the category. Finally, the category with the highest correlation coefficient is selected, and it is considered that the distribution transformers contained in this category are the most likely to participate in the transfer behavior, i.e., the preliminary transfer distribution transformer set. This method not only improves the accuracy of fault location, but also provides a direction for further analysis.

[0068] S140, based on the preliminary transfer distribution transformer set, calculate the transfer distribution transformer proportion of switch section, section total distribution transformer proportion of transfer distribution transformer proportion index, combined with switch position data, locate the upstream section switch and downstream tie switch of the transfer distribution transformer set, to correct the preliminary transfer distribution transformer set and the transfer period, to get the final transfer distribution transformer set and the final transfer period.

[0069] In this embodiment, step S140 aims to further accurately locate the actual transfer distribution transformers in the line to be diagnosed and their corresponding time periods through a series of detailed analysis steps. Specifically: Final transfer distribution transformer set: refers to a group of distribution transformers (abbreviated as "distribution transformers") that are confirmed to have participated in the transfer process after all analysis steps. This set is derived based on the preliminary transfer distribution transformer set through more accurate analysis and verification steps.

[0070] Final transfer period: refers to the specific time period during which the transfer phenomenon occurs in the line to be diagnosed based on switch position data and other related information. This helps to clarify the exact time range of the fault or operation event. In an embodiment, the above step S140 can include steps S141-S145.

[0071] S141, obtain all tie information of the line to be diagnosed, take each tie switch as the last child node, and find all section switches as a section switch list.

[0072] In this embodiment, in this step, first collect all tie information of the line to be diagnosed, and take each tie switch as the starting point to trace back to its belonging section switch, forming a list containing all related section switches. This list will be used for subsequent statistical analysis.

[0073] Specifically, collect information of all tie switches on the line to be diagnosed, including their positions, states and connection relationships. For each tie switch, trace back to the nearest section switch to create a list containing all related section switches. Use database queries or specialized software tools to automate this process to improve efficiency and accuracy.

[0074] S142, cycle the segment switch list, count the total number of distribution transformers hanging downstream of each switch as the segment total distribution transformer number, the number of distribution transformers hanging downstream of the switch existing in the preliminary transfer distribution transformer set as the switch downstream preliminary transfer distribution transformer number, and the total number of the preliminary transfer distribution transformer set as the preliminary transfer distribution transformer total number, calculate the transfer distribution transformer proportion of each switch in the segment switch list segment distribution transformer proportion index, and the segment total distribution transformer proportion of transfer distribution transformer proportion index, select the segment switch whose sum of the two indexes ranks in line with the requirements and record the corresponding tie switch, to obtain the target segment switch and the corresponding tie switch.

[0075] In this step, detailed statistical analysis is performed on each segment switch, including calculating the total number of distribution transformers hanging downstream of each segment switch (Cb), the number of distribution transformers belonging to the preliminary transfer distribution transformer set (Cc), and the total number of the preliminary transfer distribution transformer set (Ca). Then, the transfer distribution transformer proportion of each segment switch in the segment distribution transformer proportion index (P1) and the segment total distribution transformer proportion of transfer distribution transformer proportion index (P2) are calculated. The segment switches whose sum of the two indexes rank in the top three are selected, and the tie switches associated therewith are recorded as target objects for the next step of analysis.

[0076] S143, obtain the switch position data corresponding to all tie information of the to-be-diagnosed line.

[0077] In this embodiment, the operation records or position data of all tie switches on the to-be-diagnosed line are collected. These data are crucial for identifying specific transfer behaviors, as they can directly reflect which tie switches have undergone state changes at a specific time.

[0078] S144, when the switch position data exists in the tie switches corresponding to the target segment switch, the corresponding tie switch is selected as the final transfer information group; when the switch position data does not exist in the tie switches corresponding to the target segment switch, the tie switch ranked first is selected as the final transfer information group.

[0079] In this embodiment, the final transfer information group refers to the set of tie switches determined to be responsible for performing power transfer tasks based on the operation records or position data of the target segment switch and its corresponding tie switches. This set is used to identify the switches that actually perform power transfer operations and their related time periods, in order to optimize power grid management and improve power supply reliability.

[0080] ​​According to the previously selected target sectionalizing switch and its corresponding tie switch, check whether there is switch position data. If there is, it is considered as the final transfer information group; if not, the highest ranked tie switch is selected as the final transfer information group by default. This step ensures that reasonable inferences can be made even in the absence of direct evidence.

[0081] S145, extract the sectionalizing switch in the final transfer information group as the transfer sectionalizing switch, and the distribution transformers hanging between the tie switches downstream of the transfer sectionalizing switch as the final transfer distribution transformer set, and determine the final transfer period.

[0082] In this embodiment, finally, based on the information of the final transfer information group, it is determined which sectionalizing switch is the real transfer sectionalizing switch, and all the distribution transformers hanging between the sectionalizing switch and the corresponding tie switch downstream of the sectionalizing switch are identified as the final transfer distribution transformer set. At the same time, the specific transfer period, i.e. the final transfer period, is determined by using the switch position data.

[0083] In summary, step S140 accurately locates the actual distribution transformers involved in the transfer and their operation time by a series of systematic analysis processes, starting from the preliminary transfer distribution transformer set, combined with actual operation data, thereby providing a more accurate transfer analysis result.

[0084] S150, based on the final transfer distribution transformer set and the final transfer period, according to the table bottom data of the lines and distribution transformers, restore the contemporaneous line loss rate of the transfer line and the replacement line.

[0085] In this embodiment, contemporaneous line loss rate refers to the percentage of the difference between the amount of electricity transmitted by the transmission line in the power system and the actual amount of electricity sold in a specific period of time (e.g. one day or one month) to the total input electricity in that period. It reflects the energy loss caused by various losses during the transmission of electricity in the network, and is one of the important indicators for measuring the efficiency of the power grid.

[0086] In an embodiment, the above step S150 can include steps S151-S152.

[0087] S151, based on the table bottom data of all lines and the distribution transformers under the lines, combined with the final transfer distribution transformer set and the final transfer period, calculate the line forward and reverse power of the transfer line and the replacement line, the corresponding original distribution transformer forward and reverse power of the two lines, the transfer distribution transformer forward and reverse power data, respectively, to obtain intermediate calculation results.

[0088] In this embodiment, the intermediate calculation results refer to various power data obtained through the above steps, which specifically include: forward and reverse power of the transfer line and the replacement line; Forward and reverse power of all distribution transformers connected to each of the transfer line and the replacement line; In particular, for the distribution transformers within the set of transfer distribution transformers involved in the transfer, the forward and reverse power during the transfer start to end time period.

[0089] These intermediate calculation results provide necessary data support for the subsequent steps.

[0090] S152, according to the intermediate calculation results, calculate the input power, output power and sold power of the transfer line and the replacement line, and calculate the contemporaneous line loss rate of the transfer line and the replacement line.

[0091] According to the intermediate calculation results obtained in step S151, further calculate the input power (such as Q zin and Q din ), output power (such as Q zout and Q dout ) and sold power (such as Q zs and Q ds ) of each line.

[0092] Then, the contemporaneous line loss rate is calculated using the formula, for example, for the transfer line using the formula , and for the replacement line using .

[0093] Finally, by repeating the above steps for all relevant lines, the contemporaneous line loss rate of each transfer line and replacement line in the entire power grid system can be restored and calculated, thereby helping power grid managers to assess the energy loss condition of the system, optimize the power grid structure and operation strategy, and improve energy utilization efficiency.

[0094] In this embodiment, based on the obtained total indication data of lines and distribution transformers, combined with the set of transfer distribution transformers and the transfer time period data, the following power is calculated respectively: Forward power of the transfer line: ; wherein PAPzc is the total indication of forward active power of the transfer line on the day, and PAPzr is the total indication of forward active power of the transfer line on the previous day.

[0095] Forward power of the replacement line: ; wherein PAPdc is the total indication of forward active power of the replacement line on the day, and PAPdr is the total indication of forward active power of the replacement line on the previous day.

[0096] Reverse power of the transfer line: ; wherein RAPzc is the total indication of reverse active power of the transfer line on the day, and RAPzr is the total indication of reverse active power of the transfer line on the previous day.

[0097] Reverse active power of the replacement supply line: ; wherein RAPdc is the total number of reverse active power of the replacement supply line on the day, and RAPdr is the total number of reverse active power of the replacement supply line on the previous day.

[0098] Positive active power of the original distribution transformer of the replacement supply line: ; ; wherein n is the total number of distribution transformers connected in the account information under the replacement supply line, PAPk,tzc is the total number of positive active power of the distribution transformer k on the day under the replacement supply line, and PAPk,tzr is the total number of positive active power of the distribution transformer k on the previous day under the replacement supply line.

[0099] Positive active power of the original distribution transformer of the replacement supply line: ; wherein m is the total number of distribution transformers connected in the account information under the replacement supply line, PAPk,tdc is the total number of positive active power of the distribution transformer k on the day under the replacement supply line, and PAPk,tdr is the total number of positive active power of the distribution transformer k on the previous day under the replacement supply line.

[0100] Reverse active power of the original distribution transformer of the replacement supply line: ; wherein n is the total number of distribution transformers connected in the account information under the replacement supply line, RAPk,tzc is the total number of reverse active power of the distribution transformer k on the day under the replacement supply line, and RAPk,tzr is the total number of reverse active power of the distribution transformer k on the previous day under the replacement supply line.

[0101] Reverse active power of the original distribution transformer of the replacement supply line: ; wherein m is the total number of distribution transformers connected in the account information under the replacement supply line, RAPk,tdc is the total number of reverse active power of the distribution transformer k on the day under the replacement supply line, and RAPk,tdr is the total number of reverse active power of the distribution transformer k on the previous day under the replacement supply line.

[0102] Positive active power of the replacement distribution transformer: ; wherein x is the total number of distribution transformers in the replacement distribution transformer set Dt, PAPk,tc is the total number of positive active power of the distribution transformer k on the day corresponding to the end time of the replacement under the replacement distribution transformer set Dt, and PAPk,tr is the total number of positive active power of the distribution transformer k on the day corresponding to the start time of the replacement under the replacement distribution transformer set Dt.

[0103] Reverse active power of the replacement distribution transformer: ; wherein x is the total number of distribution transformers in the replacement distribution transformer set Dt, RAPk,tc is the total number of reverse active power of the distribution transformer k on the day corresponding to the end time of the replacement under the replacement distribution transformer set Dt, and RAPk,tr is the total number of reverse active power of the distribution transformer k on the day corresponding to the start time of the replacement under the replacement distribution transformer set Dt.

[0104] According to the data calculated above, the input power, output power, and sold power of the replacement supply line are calculated, and the same period line loss rate of the two lines is further calculated:

[0105] Input power of the transfer line: ; Output power of the transfer line: ; Sold power of the transfer line: ; According to the input power, output power and sold power of the transfer line calculated in the above steps, the contemporaneous line loss rate of the transfer line is further calculated: ; Input power of the substitute line: ; Output power of the substitute line: ; Sold power of the substitute line: ; According to the input power, output power and sold power of the substitute line calculated in the above steps, the contemporaneous line loss rate of the substitute line is further calculated: ; The above steps are repeated until the contemporaneous line loss rates of all transfer lines and substitute lines are calculated.

[0106] The method of the embodiment can accurately locate the specific transfer area by combining the switch action information. This method provides strong technical support for accurate identification of line transfer and accurate calculation of contemporaneous line loss rate. Traditional methods often cannot accurately identify the transfer area, resulting in inaccurate line loss calculation. The method of the present application effectively solves this problem.

[0107] In the embodiment, a method for identifying line transfer using DTW (Dynamic Time Warping) and DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is proposed. The combination of these two algorithms not only improves the accuracy of line transfer device positioning, but also greatly shortens the identification time. Compared with existing technical means, the method of the embodiment can conveniently identify line transfer devices at a faster speed and with higher accuracy, greatly improving work efficiency and reliability.

[0108] The method of the embodiment uses a relatively simple calculation method, and the principle behind it is very clear, easy to understand and implement. This enables it to achieve accurate restoration calculation of the line loss rate of the transfer line, helping operation and maintenance personnel to quickly locate the real line loss abnormal line. In addition, the method provides an effective tool for line loss anomaly management, which helps to improve the operation efficiency and service quality of the power grid. Due to its simple implementation and significant effect, it has broad application prospects.

[0109] In summary, the method of the embodiment has excellent ability in improving the management efficiency of the distribution network and reducing line loss. It not only optimizes the traditional line loss calculation method, but also greatly improves the accuracy and speed of line transfer equipment identification by introducing advanced clustering algorithms and techniques, which is of great significance to improving the overall performance of the power system.

[0110] The above distribution network line loss restoration calculation method based on a clustering algorithm obtains detailed topological structures of all related lines, transformer substation accounts, associated relationships of tie switches, and key operating parameters as initial data. Then, the initial data is subjected to feature index extraction, and an index threshold method is used to identify lines that have occurred transfer, i.e., the line to be diagnosed. Then, the voltage similarity matrix of the transformer substation of the line to be diagnosed during the transfer period is calculated and subjected to clustering analysis to identify a preliminary transfer transformer substation set. Further, based on the preliminary transfer transformer substation set, a specific proportion index is calculated and combined with switch position data to accurately locate the upstream and downstream key switch positions of the transfer transformer substation set, thereby correcting the preliminary result to obtain the final transfer transformer substation set and its corresponding transfer period. Finally, according to the table bottom data of the lines and transformer substations, the contemporaneous line loss rate of the transfer and supply lines is accurately calculated and restored, providing strong data support and technical support for the optimization management of the power system. This method effectively improves the accuracy and reliability of the contemporaneous line loss rate of the transfer line in the power system, which helps to develop more scientific and reasonable energy-saving strategies.

[0111] Figure 2 is a schematic block diagram of a distribution network line loss restoration calculation system 300 based on a clustering algorithm provided by an embodiment of the present application. As Figure 2 As shown in the above distribution network line loss restoration calculation method based on a clustering algorithm, the present application also provides a distribution network line loss restoration calculation system 300 based on a clustering algorithm. The distribution network line loss restoration calculation system 300 based on a clustering algorithm includes units for executing the above distribution network line loss restoration calculation method based on a clustering algorithm. The system can be configured in a desktop computer, a tablet computer, a laptop computer, and the like. Specifically, please refer to Figure 2The power distribution network line loss restoration calculation system 300 based on the clustering algorithm comprises an acquisition unit 301, an extraction unit 302, an identification unit 303, a correction unit 304, and a restoration calculation unit 305.

[0112] The acquisition unit 301 is configured to acquire the topological structure of all lines, the transformer account, the association relationship between the lines and the contact switches, and acquire the table bottom display data, power, voltage data and switch position data of all lines and the lines under jurisdiction, to obtain initial data; the extraction unit 302 is configured to extract characteristic indexes from the initial data, and identify line transfer by using the index threshold method, define the line where transfer occurs as a line to be diagnosed, to obtain a transfer line, a supply line and a transfer period; the identification unit 303 is configured to calculate the voltage similarity matrix between all transformer voltage curves in the transfer period on the line to be diagnosed, and then cluster the voltage similarity matrix, and identify the transfer transformer set of the line to be diagnosed by combining each cluster category with the voltage similarity index of the supply line, to obtain a preliminary transfer transformer set; the correction unit 304 is configured to calculate the switch section proportion of the transfer transformer and the section total transformer proportion of the transfer transformer based on the preliminary transfer transformer set, and locate the upstream section switch and the downstream contact switch of the transfer transformer set by combining the switch position data, to correct the preliminary transfer transformer set and the transfer period, to obtain a final transfer transformer set and a final transfer period; and the restoration calculation unit 305 is configured to restore the contemporaneous line loss rate of the transfer line and the supply line according to the table bottom display data of the line and the transformer based on the final transfer transformer set and the final transfer period.

[0113] In an embodiment, the extraction unit 302 comprises: A first calculation subunit is configured to calculate the time point line loss rate of all lines based on the initial data; a screening subunit is configured to determine all opposite line information associated with the line when the time point line loss rate of the line is negative loss for a certain period of time, and screen out the opposite line information whose time point line loss rate is greater than a set threshold in the negative loss period of the line, to obtain a screening result; a second calculation subunit is configured to calculate the transfer line loss rate of the line and the opposite line in the screening result in the negative loss period of the line; and a determination subunit is configured to determine the line as a line to be diagnosed and the opposite line as a supply line when the transfer line loss rate meets the requirements, and determine the negative loss period as a transfer period, to obtain a transfer line, a supply line and a transfer period.

[0114] In an embodiment, the screening unit is configured to, for the to-be-diagnosed line, calculate a voltage similarity matrix between voltage curves of all distribution transformers in the to-be-diagnosed line during the period of distribution transformer switching, and perform clustering on the voltage similarity matrix by using a density clustering algorithm, and identify a set of distribution transformers for switching of the to-be-diagnosed line by combining a voltage similarity index judgment of each cluster category and the distribution transformer switching line, to obtain a preliminary set of distribution transformers for switching.

[0115] In an embodiment, the screening unit includes: The distance calculation sub-unit is configured to, for the to-be-diagnosed line, extract voltage measurement data of the distribution transformer switching line and the connected distribution transformer during the period of distribution transformer switching, perform smoothing processing on the voltage by using a moving translation method to eliminate voltage fluctuation noise according to the sorted measurement uploading time, form a smoothed voltage curve, and calculate a DTW distance between the two smoothed voltage curves of each distribution transformer and the line by using Manhattan distance; the matrix construction sub-unit is configured to convert the DTW distance into a similarity index, and construct a voltage similarity matrix; and the clustering sub-unit is configured to perform clustering on the similarity matrix by using a DBSCAN clustering algorithm, obtain a plurality of cluster categories, extract a voltage curve of the distribution transformer switching line during the period of distribution transformer switching, calculate a correlation coefficient of all distribution transformer voltage curves in each category and the distribution transformer switching line voltage curve and average the correlation coefficient as a category correlation coefficient, and select a category with the largest category correlation coefficient, and define the distribution transformers in the category as a preliminary set of distribution transformers for switching.

[0116] In an embodiment, the correction unit 304 includes: The information acquisition subunit is configured to acquire all tie information of the to-be-diagnosed line, take each tie switch as a last sub-node, and find all sectional switches as a sectional switch list; the switch determination subunit is configured to loop the sectional switch list, count a total number of distribution transformers hung under each switch as a sectional total distribution transformer number, count a number of distribution transformers hung under each switch existing in the preliminary transfer supply distribution transformer set as a preliminary transfer supply distribution transformer number, count a total number of the preliminary transfer supply distribution transformer set as a preliminary transfer supply distribution transformer total number, calculate a transfer supply distribution transformer occupancy ratio index of each switch in the sectional switch list, and calculate a sectional total distribution transformer occupancy transfer supply distribution transformer ratio index; the target sectional switch and the corresponding tie switch are obtained by selecting a sectional switch whose sum of the two indexes meets a requirement and recording the corresponding tie switch; the data acquisition subunit is configured to acquire switch position data corresponding to all tie information of the to-be-diagnosed line; the selection subunit is configured to select a corresponding tie switch as a final transfer supply information group when the switch position data exists in the tie switch corresponding to the target sectional switch, and select a tie switch ranked first as the final transfer supply information group when the switch position data does not exist in the tie switch corresponding to the target sectional switch; the extraction subunit is configured to extract a sectional switch in the final transfer supply information group as a transfer supply sectional switch, extract distribution transformers hung between the transfer supply sectional switch and a tie switch downstream of the transfer supply sectional switch as a final transfer supply distribution transformer set, and determine a final transfer supply time period.

[0117] In an embodiment, the reduction calculation unit 305 includes: The intermediate calculation subunit is configured to calculate line forward power and line reverse power of the transfer supply line and the substitute supply line, corresponding original distribution transformer forward power and original distribution transformer reverse power of the two lines, transfer supply distribution transformer forward power and transfer supply distribution transformer reverse power data, respectively, based on table bottom data of all lines and distribution transformers under the jurisdiction of the lines, in combination with the final transfer supply distribution transformer set and the final transfer supply time period, to obtain an intermediate calculation result. The synchronous line loss rate calculation subunit is configured to calculate input power, output power, and power sold of the transfer supply line and the substitute supply line, and calculate a synchronous line loss rate of the transfer supply line and the substitute supply line, according to the intermediate calculation result.

[0118] It should be noted that those skilled in the art can clearly understand the specific implementation process of the above-mentioned power distribution network line loss reduction calculation system 300 and each unit based on the clustering algorithm, which can refer to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, it will not be repeated here.

[0119] The above-mentioned power distribution network line loss reduction calculation system 300 based on the clustering algorithm can be implemented in the form of a computer program, which can run on a computer device as shown in the accompanying drawings. Figure 3 The computer device shown in the accompanying drawings.

[0120] Please refer to Figure 3 , Figure 3 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device 500 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0121] Please refer to Figure 3 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, wherein the memory can include a non-volatile storage medium 503 and an internal memory 504.

[0122] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions which, when executed, can cause the processor 502 to perform a power distribution network line loss restoration calculation method based on a clustering algorithm.

[0123] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0124] The internal memory 504 provides an environment for the running of the computer program 5032 in the non-volatile storage medium 503, which, when executed by the processor 502, can cause the processor 502 to perform a power distribution network line loss restoration calculation method based on a clustering algorithm.

[0125] The network interface 505 is configured to perform network communication with other devices. Those skilled in the art can understand that Figure 3 The structure shown in the above

[0126] The processor 502 is configured to run the computer program 5032 stored in the memory to implement all steps of the power distribution network line loss restoration calculation method based on a clustering algorithm described above.

[0127] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0128] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments.

[0129] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program is executed by a processor to make the processor execute all steps of the above-mentioned power distribution network line loss restoration calculation method based on a clustering algorithm.

[0130] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer-readable storage media that can store program codes.

[0131] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of each example have been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0132] In several embodiments provided by the present application, it should be understood that the disclosed system and method can be implemented in other manners. For example, the division of the system embodiments is merely illustrative. For example, the division of the units can be divided in another manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. The disclosed embodiments of the present application are not limited to the described division.

[0133] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the system embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0134] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part of the prior art that makes a contribution, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application.

[0135] The above description is merely a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for calculating line losses in distribution networks based on clustering algorithms, characterized in that, include: Obtain the topology of all lines, transformer ledgers, and the relationship between lines and tie switches. Also obtain the meter readings, power and voltage data, and switch position data of all lines and their subordinate transformers to obtain the initial data. Feature indicators are extracted from the initial data, and the indicator threshold method is used to identify line transfer. Lines that have been transferred are defined as lines to be diagnosed, so as to obtain the transferred lines, the substitute lines, and the transfer time periods. For the line to be diagnosed, calculate the voltage similarity matrix between the voltage curves of all distribution transformers switching power supply periods on the line to be diagnosed, then cluster the voltage similarity matrix, and combine the voltage similarity index of each cluster category with the power supply line to identify the set of distribution transformers switching power supply for the line to be diagnosed, so as to obtain a preliminary set of distribution transformers switching power supply. Based on the preliminary set of transfer transformers, the proportion of transfer transformers to switch sections and the proportion of total transformers in sections to transfer transformers are calculated. Combined with switch position data, the upstream section switches and downstream tie switches of the transfer transformer set are located to correct the preliminary set of transfer transformers and the transfer time period, so as to obtain the final set of transfer transformers and the final transfer time period. Based on the final set of distribution transformers and the final transfer period, the synchronous line loss rate of the transfer lines and the substitute lines is calculated according to the meter readings of the lines and transformers.

2. The method for calculating distribution network line losses based on clustering algorithm according to claim 1, characterized in that, The process of extracting feature indicators from the initial data and using the indicator threshold method to identify line transfers, defining lines that have experienced transfers as lines to be diagnosed, and obtaining the transferred lines, substitute lines, and transfer periods, includes: Calculate the moment-based line loss rate for all lines based on the initial data; If the line loss rate is negative for a certain period of time, then all peer line information connected to the line is determined, and peer line information whose line loss rate is greater than a set threshold during the negative loss period of the line is filtered out to obtain the filtering results. Calculate the transfer line loss rate between the line and the peer line in the screening results during the negative loss period of the line; When the transfer line loss rate meets the requirements, the line is determined to be the line to be diagnosed and the opposite line is the line to be supplied on behalf of another. The period of negative line loss is taken as the transfer period to obtain the transfer line, the line to be supplied on behalf of another, and the transfer period.

3. The method for calculating distribution network line losses based on clustering algorithm according to claim 1, characterized in that, For the line to be diagnosed, a voltage similarity matrix is ​​calculated between the voltage curves of all distribution transformers on the line during the switching period. Then, the voltage similarity matrix is ​​clustered, and the set of switching distribution transformers for the line to be diagnosed is identified by combining the voltage similarity index of each cluster with that of the switching line, thus obtaining a preliminary set of switching distribution transformers, including: For the line to be diagnosed, the voltage similarity matrix between the voltage curves of all distribution transformers switching power supply periods on the line to be diagnosed is calculated using the dynamic time warping algorithm. Then, the voltage similarity matrix is ​​clustered using the density clustering algorithm. By combining the voltage similarity index of each cluster category with that of the line to be supplied, the set of distribution transformers switching power supply for the line to be diagnosed is identified, so as to obtain a preliminary set of distribution transformers switching power supply.

4. The method for calculating distribution network line losses based on clustering algorithm according to claim 3, characterized in that, For the line to be diagnosed, a voltage similarity matrix is ​​calculated between the voltage curves of all distribution transformers switching power supply periods on the line to be diagnosed using a dynamic time warping algorithm. Then, a density clustering algorithm is used to cluster the voltage similarity matrix. Combining the voltage similarity index of each cluster category with that of the line to be supplied, the set of distribution transformers switching power supply for the line to be diagnosed is identified to obtain a preliminary set of distribution transformers switching power supply, including: For the line to be diagnosed, the voltage measurement data of the transfer line and the connected distribution transformer during the transfer period are extracted and sorted according to the measurement transmission time. The sorted voltage measurement data are smoothed by eliminating voltage fluctuation noise using the moving translation method to form a smoothed voltage curve. The Manhattan distance is used to calculate the DTW distance between the two smoothed voltage curves of each distribution transformer and the line. The DTW distance is converted into a similarity index, and a voltage similarity matrix is ​​constructed. Using the DBSCAN clustering algorithm, the similarity matrix is ​​clustered to obtain several cluster categories. The voltage curves of the power supply line during the power transfer period are extracted. The correlation coefficients between the voltage curves of all distribution transformers in each category and the voltage curves of the power supply line are calculated and averaged as the category correlation coefficient. The category with the largest category correlation coefficient is selected, and the distribution transformers under the category are defined as the initial power transfer distribution transformer set.

5. The method for calculating distribution network line losses based on clustering algorithm according to claim 4, characterized in that, The formula for calculating the DTW distance is as follows: Where L is the normalized path length, and xm and ym are the smoothed voltage curves, respectively. The sampling point index, d( () represents the distance to Manhattan; The formula for calculating the similarity index is: Sij = exp(-α × ); where α is the adjustment coefficient, and the closer Sij is to 1, the higher the similarity between the two curves; The voltage similarity matrix is ​​M M is a similarity matrix, where M is the total number of distribution transformers under the line to be diagnosed.

6. The method for calculating distribution network line losses based on clustering algorithm according to claim 1, characterized in that, Based on the initial set of transfer transformers, the percentage of transfer transformers in a switch section and the percentage of total transformers in a section in the transfer transformer ratio are calculated. Combined with switch position data, the upstream section switches and downstream tie switches of the transfer transformer set are located to correct the initial set of transfer transformers and the transfer time period, resulting in the final set of transfer transformers and the final transfer time period. This includes: Obtain all contact information of the line to be diagnosed, and use each contact switch as the last child node to search upwards for all sectional switches to form a sectional switch list. The cyclic segmented switch list is used to count the total number of downstream distribution transformers connected to each switch as the total number of distribution transformers in the segment, the number of downstream distribution transformers connected to the switch that exist in the initial transfer distribution transformer set as the initial transfer distribution transformer number downstream of the switch, and the total number of the initial transfer distribution transformer set as the initial transfer distribution transformer total number. The transfer distribution transformer ratio of each switch in the segmented switch list and the ratio of the total distribution transformer in the segment to the transfer distribution transformer are calculated. The segmented switches that meet the requirements for the sum of the two indicators are selected and the corresponding tie switches are recorded to obtain the target segmented switches and their corresponding tie switches. Obtain the switch position data corresponding to all communication information of the line to be diagnosed; If the switch change data exists in the connecting switch corresponding to the target segment switch, the corresponding connecting switch is selected as the final transfer information group. If the switch change data does not exist in the connecting switch corresponding to the target segment switch, the first-ranked connecting switch is selected as the final transfer information group. Extract the sectionalizing switches from the final transfer information group as transfer sectionalizing switches, and the distribution transformers connected downstream of the transfer sectionalizing switches to the tie switches as the final transfer distribution transformer set, and determine the final transfer time period.

7. The method for calculating distribution network line losses based on clustering algorithm according to claim 6, characterized in that, Based on the final set of distribution transformers and the final transfer period, and according to the meter readings of the lines and transformers, the synchronous line loss rate of the transfer lines and the substitute lines is calculated, including: Based on the meter readings of all lines and their subordinate distribution transformers, combined with the final set of transfer distribution transformers and the final transfer period, the forward and reverse electricity of the transfer lines and the replacement supply lines, the forward and reverse electricity of the original distribution transformers corresponding to the two lines, the forward electricity of the transfer distribution transformers, and the reverse electricity of the transfer distribution transformers are calculated respectively to obtain intermediate calculation results. Based on the intermediate calculation results, the input power, output power, and sales power of the transfer line and the substitute line are calculated, and the synchronous line loss rate of the transfer line and the substitute line is calculated.

8. A distribution network line loss restoration calculation system based on clustering algorithm, characterized in that, include: The acquisition unit is used to acquire the topology of all lines, distribution transformer ledgers, the relationship between lines and tie switches, and to acquire the meter readings, power, voltage data and switch position data of all lines and their subordinate distribution transformers in order to obtain the initial data. The extraction unit is used to extract feature indicators from the initial data and use the indicator threshold method to identify line transfer. The line that has been transferred is defined as the line to be diagnosed, so as to obtain the transferred line, the substitute line, and the transfer time period. The identification unit is used to calculate the voltage similarity matrix between the voltage curves of all distribution transformers switching power supply periods on the line to be diagnosed, and then cluster the voltage similarity matrix. By combining the voltage similarity index of each cluster category with the voltage similarity index of the line to be supplied, the set of distribution transformers switching power supply for the line to be diagnosed is identified to obtain a preliminary set of distribution transformers switching power supply. The correction unit is used to calculate the proportion of the transfer distribution transformers to the switch sections and the proportion of the total distribution transformers in the sections to the transfer distribution transformers based on the preliminary transfer distribution transformer set. Combined with the switch position data, it locates the upstream section switch and the downstream tie switch of the transfer distribution transformer set to correct the preliminary transfer distribution transformer set and the transfer time period to obtain the final transfer distribution transformer set and the final transfer time period. The restoration calculation unit is used to restore and calculate the synchronous line loss rate of the transfer line and the substitute line based on the final transfer transformer set and the final transfer period, according to the meter reading data of the line and the transformer.

9. The distribution network line loss restoration calculation system based on clustering algorithm according to claim 8, characterized in that, The extraction unit includes: The first calculation subunit is used to calculate the moment line loss rate of all lines based on the initial data; The filtering subunit is used to determine all peer line information connected to the line when the line has a negative line loss rate for a continuous period of time, and to filter out peer line information whose line loss rate is greater than a set threshold during the negative line loss period, so as to obtain the filtering result. The second calculation subunit is used to calculate the transfer line loss rate between the line and the peer line in the screening results during the negative line loss period. A sub-unit is defined to determine the line to be diagnosed and the opposite line to be supplied when the transfer line loss rate meets the requirements, and to use the negative loss period of the line as the transfer period, so as to obtain the transfer line, the supply line, and the transfer period.

10. The distribution network line loss restoration calculation system based on clustering algorithm according to claim 8, characterized in that, The screening unit is used to calculate the voltage similarity matrix between the voltage curves of all distribution transformers switching power supply periods on the line to be diagnosed using a dynamic time warping algorithm, and then use a density clustering algorithm to cluster the voltage similarity matrix. By combining the voltage similarity index of each cluster category with that of the line to be supplied, the set of distribution transformers switching power supply for the line to be diagnosed is identified to obtain a preliminary set of distribution transformers switching power supply.

Images