An intelligent identification method and system applied to an effective switch of a power distribution network

Abstract

The application discloses an intelligent identification method and system applied to an effective switch of a power distribution network, relates to the field of automatic operation control of the power distribution network, and solves the technical problems that it is difficult to identify potential effective tie switches and invalid branch switches in the operation of the power distribution network and that it is difficult to adaptively perform equivalent merging on redundant switches; the application traces bidirectional substations through each switch and acquires a line influence factor; determines a running state factor of each switch based on a historical running state of each switch, determines effective tie switches and invalid branch switches based on the line influence factor and the running state factor; plans a plurality of time periods, identifies merging influence factors of each time period according to historical power consumption loads and environmental data, identifies series switch groups, and selectively combines the series switch groups into a single virtual switch based on the merging influence factors; and the application can eliminate topological redundant nodes and improve the calculation efficiency and stability of a subsequent reconstruction optimization algorithm.

Description

Technical Field

[0001] This invention relates to the field of automated operation control of distribution networks, specifically to an intelligent identification method and system for valid switches in distribution networks. Background Technology

[0002] With increasing demands for power supply reliability, most power distribution lines in various regions are now interconnected, forming a "hand-in-hand" power supply mode. While achieving interconnection and interoperability of the regional power distribution network and further optimizing the network structure and operation can solve the problem of a single line fault causing a wider power outage area, in actual operation, interconnection switches are only used for mutual switching during power supply maintenance. This unreasonable interconnection method can lead to a mismatch between the load center and the power source of the distribution lines, ultimately causing unnecessary losses, uneven load distribution, and excessive losses in the regional power distribution network.

[0003] Currently, most intelligent identification methods for effective switches in distribution networks struggle to identify potential effective tie switches and ineffective branch switches during network operation. They rely solely on binary judgments to determine effectiveness, which can easily lead to the omission of ineffective branch switches that appear to be effective tie switches in the topology but do not actually perform their tie function in operation, thus reducing the accuracy of identifying ineffective branch switches.

[0004] Therefore, this invention discloses an intelligent identification method and system for valid switches in power distribution networks, which solves the above-mentioned technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an intelligent identification method and system for effective switches in distribution networks, which solves the technical problems of difficulty in identifying potential effective tie switches and invalid branch switches in the operation of distribution networks, and difficulty in adaptively merging redundant switches.

[0006] To achieve the above objectives, a first aspect of the present invention provides an intelligent identification method for valid switches in a distribution network, comprising: Acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; Bidirectional substation tracing was performed on each switch to obtain the line impact factor; The operating status factors of each switch are determined based on the historical operating status of each switch, and the effective tie switches and ineffective branch switches are determined based on the line influence factors and operating status factors. The day is divided into several time periods. Based on historical electricity load and environmental data, the combined impact factors of each time period are identified. Series switch groups are identified, and series switches are selectively combined into a single virtual switch based on the combined impact factors.

[0007] Preferably, the acquisition of the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located, includes: The system extracts the number of days each switch has been in use, the number of times it has been activated, and the number of times it has been repaired since it was put into use from the switch status database. The target number of days is obtained by adding the number of days of use to the number of days of error elimination for new switches. The number of days of error elimination for new switches is a manually set number of days to eliminate the analysis error caused by the short number of days of use of new switches. It is generally set to 10 days. Divide the number of actions by the number of days of use to obtain the frequency of actions; divide the number of fault repairs by the number of days of use to obtain the frequency of fault repairs. Historical electricity load data for the area where the switch is located was extracted from the regional electricity load database at various time points, and the temperature and humidity of the area where the switch is located were obtained from the weather forecast platform.

[0008] Preferably, the step of performing bidirectional substation tracing of each switch and obtaining the line influence factor includes: A1: Sequentially extract switches, perform bidirectional path tracing on the extracted switches, and determine whether both ends of the bidirectional path tracing can be traced back to the substation; if yes, jump to A2; if no, mark the current switch as an invalid branch switch; the bidirectional path includes the incoming line and the outgoing line. A2: In the current bidirectional path tracing of the switch, the nearest substation on the incoming line to the current switch location is marked as Substation 1, and the nearest substation on the outgoing line to the current switch location is marked as Substation 2; the wire length between Substation 1 and the distribution network at the current switch location is marked as Length 1. Mark the wire length between substation two and the distribution network at the current switch location as length two. Based on length one and length two The line influence factor is determined by formula (1). ; The calculation formula (1) is: ; In the formula, The amplitude adjustment coefficient is determined based on the straight-line distance between the current switch and the nearest substation. The range of values ​​for is (0,1); and It is a proportional adjustment coefficient obtained through manual setting, and and All are not less than 0. ; The value is typically taken as 0.5. The typical value is 0.5; The length of the historical record and length two The average value.

[0009] Preferably, determining the operating state factor of each switch based on its historical operating state includes: B1: Extract the standard state factor range, and sequentially extract the frequency of switch actions. and frequency of fault repair The standard state factor range is determined based on the historical standard state factors of each switch in the distribution network. B2: Frequency of current switch operation When the frequency exceeds the threshold, the operating state factor of the current switch will be adjusted. The value is set to the maximum value within the standard state factor range; otherwise, jump to B2; where the frequency threshold is determined based on the percentile of the historical operation frequency of each switch in the distribution network; B2: Determine the current switch's operating state factor based on formula (2). ; The calculation formula (2) is: ; In the formula, This is the minimum value within the standard state factor range. This is the maximum value within the standard state factor range; and It is a proportional adjustment coefficient set according to the importance of the frequency of actions and the frequency of fault repairs, and , .

[0010] Preferably, the determination of valid tie switches and invalid branch switches based on line influence factors and operating status factors includes: Extract the circuit influence factors of the switches sequentially and operating state factors And the influence factors of the line and operating state factors The switching discrimination factor of the current switch is obtained by combining the proportions. ; When the switch discrimination factor When the value is greater than the standard discrimination factor, the current switch is marked as a valid contact switch; when the switch discrimination factor is greater than the standard discrimination factor, the current switch is marked as a valid contact switch. If the value is not greater than the standard discrimination factor, the current switch is marked as an invalid branch switch; the standard discrimination factor is determined based on the historical standard discrimination factors of manually selected valid connection switches and invalid branch switches.

[0011] Preferably, the division of a day into several time periods includes: At 11:59 every day, the historical electricity load of the current area for the previous n days is obtained, and the historical electricity load is integrated into several electricity load groups based on the time points; where n is obtained manually, and is generally taken as 30. The variance of historical electricity loads in each electricity load group is obtained sequentially. It is then determined whether the variance of the electricity load group is less than the electricity load variance threshold. If yes, the characteristic electricity load is calculated by averaging the historical electricity loads in the electricity load group. If no, the historical electricity load with the largest absolute difference from the average value in the electricity load group is removed, and the variance of the remaining data in the electricity load group is recalculated and re-judged. This process continues until the variance of the electricity load group is less than the electricity load variance threshold. Finally, the characteristic electricity load is calculated by averaging the remaining historical electricity loads in the electricity load group. The electricity load variance threshold is set manually. The characteristic electricity loads of each electricity load group are marked as the characteristic electricity loads at the corresponding time points; the characteristic electricity loads are integrated into characteristic load groups, and the average value of the data in the characteristic load group is marked as the characteristic average value; The time points when the characteristic electricity load is greater than the characteristic average value are marked as high load time points, and the time points when the characteristic electricity load is less than the characteristic average value are marked as low load time points. The next day is planned into several reference time periods with a fixed duration, and the proportion of high load time points and the proportion of low load time points in each reference time period are obtained. When the proportion of the first reference time period is greater than the proportion of the second reference time period, the corresponding reference time period is divided into 2m time periods. When the proportion of the first reference time period is not greater than the proportion of the second reference time period, the corresponding reference time period is divided into m time periods. Here, m is obtained manually, and is generally taken as 2. The fixed duration is obtained manually, and is generally taken as 30 minutes.

[0012] Preferably, the step of identifying the combined influencing factors for each time period based on historical electricity load and environmental data includes: Obtain the characteristic electricity load at each time point in each time period of the next day. Average temperature and average humidity ; compare the temperature over a historical period with the average value The same, and the humidity is the same as the average. The same time period is marked as the comparison time period; obtain the characteristic electricity load at each time point in the comparison time period. Based on characteristic electrical load and characteristic electrical loads The combined impact factor for each time period is determined by formula (3). ;in, Let each time point in the time period be a number, and The range of values ​​for is [1, ... ], This represents the maximum value of the time point numbers within the time period; To compare the numbering of each time point within the time period, and The range of values ​​for is [1, ... ], To compare the maximum value of the time point numbers within the time period; The calculation formula (3) is: ; In the formula, max() is the maximum value function, mode() is the mode function, and BTY is the standard power load set according to the area type and historical power consumption.

[0013] Preferably, the identification of the series switch group includes: Integrate continuously connected switches in the distribution network that are physically connected and have no load or transformer in between into a series switch group.

[0014] Preferably, the selective combination of series switches into a single virtual switch based on the merging influence factor includes: Extracting and merging impact factors When the combined influence factors When the combined impact factor is greater than the standard, all series switches whose sum of impedances is less than the standard impedance are combined into a single virtual switch; when the combined impact factor... When the sum of the impedances of all switches is not greater than the standard merging influence factor, the series switches with impedances not less than the standard impedance are combined into a single virtual switch; where the standard merging influence factor and the standard impedance are both obtained through empirical settings.

[0015] A second aspect of the present invention provides an intelligent identification system for valid switches in a distribution network, comprising: a switch determination module, and a data acquisition module and a redundancy merging module connected to the switch determination module; The data acquisition module is used to acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; wherein, the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; The switch determination module is used to perform bidirectional substation tracing for each switch and obtain the line influence factor; determine the operating status factor of each switch based on its historical operating status; and determine the valid tie switches and invalid branch switches based on the line influence factor and the operating status factor. The redundant merging module is used to divide a day into several time periods, identify the merging impact factors of each time period based on historical electricity load and environmental data, identify series switch groups, and selectively combine series switches into a single virtual switch based on the merging impact factors.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention obtains the historical operating status of switches, as well as historical power load and environmental data of the area where the switches are located; performs bidirectional substation tracing for each switch and obtains line influence factors; determines the operating status factors of each switch based on its historical operating status; and determines effective tie switches and ineffective branch switches based on the line influence factors and operating status factors. It divides a day into several time periods, identifies the merging influence factors for each time period based on historical power load and environmental data, identifies series switch groups, and selectively combines series switches into a single virtual switch based on the merging influence factors. This solves the technical problems of difficulty in identifying potential effective tie switches and ineffective branch switches in distribution network operation, and difficulty in adaptively merging redundant switches. It can eliminate topological redundancy nodes and improve the computational efficiency and stability of subsequent reconfiguration optimization algorithms.

[0017] 2. This invention proposes a method for obtaining line influence factors through bidirectional substation tracing. By tracing the path bidirectionally, it can accurately determine whether both the incoming and outgoing ends of a switch can be traced back to a substation, thereby identifying invalid branch switches, promptly discovering and addressing potential problems in the power system, and improving system reliability and security. Furthermore, the line influence factor LZ calculation model comprehensively considers the differences in line length between the incoming and outgoing ends and their impact on grid operation. By constructing a multi-dimensional evaluation system, it can effectively identify invalid branch switches that appear to be effective interconnecting switches in the topology but do not actually perform their interconnecting function in operation. This data-driven analysis method not only improves the accuracy of identifying invalid branch switches but also provides a scientific basis for the optimized operation of the distribution network. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the operation steps of the present invention; Figure 2 This is a schematic diagram illustrating the operational steps for obtaining the line influence factor according to the present invention; Figure 3 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0020] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figure 1 The first aspect of this invention provides an intelligent identification method for valid switches in a distribution network, comprising: Acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; Bidirectional substation tracing was performed on each switch to obtain the line impact factor; The operating status factors of each switch are determined based on the historical operating status of each switch, and the effective tie switches and ineffective branch switches are determined based on the line influence factors and operating status factors. The day is divided into several time periods. Based on historical electricity load and environmental data, the combined impact factors of each time period are identified. Series switch groups are identified, and series switches are selectively combined into a single virtual switch based on the combined impact factors.

[0022] This application obtains the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located, including: The system extracts the number of days each switch has been in use, the number of times it has been activated, and the number of times it has been repaired since it was put into use from the switch status database. The target number of days is obtained by adding the number of days of use to the number of days of error elimination for new switches. The number of days of error elimination for new switches is a manually set number of days to eliminate the analysis error caused by the short number of days of use of new switches. It is generally set to 10 days. Divide the number of actions by the number of days of use to obtain the frequency of actions; divide the number of fault repairs by the number of days of use to obtain the frequency of fault repairs. Historical electricity load data for the area where the switch is located was extracted from the regional electricity load database at various time points, and the temperature and humidity of the area where the switch is located were obtained from the weather forecast platform.

[0023] It should be noted that the switch status database is used to store the number of days of use, number of operations, and number of fault repairs for each switch in the distribution network.

[0024] It should be noted that, regardless of whether it is a new switch or an old switch, the number of days for error elimination of the new switch needs to be added to obtain the target number of days.

[0025] Please see Figure 2 This application performs bidirectional substation tracing for each switch and obtains the line impact factor, including: A1: Sequentially extract switches, perform bidirectional path tracing on the extracted switches, and determine whether both ends of the bidirectional path tracing can be traced back to the substation; if yes, jump to A2; if no, mark the current switch as an invalid branch switch; the bidirectional path includes the incoming line and the outgoing line. A2: In the current bidirectional path tracing of the switch, the nearest substation on the incoming line to the current switch location is marked as Substation 1, and the nearest substation on the outgoing line to the current switch location is marked as Substation 2; the wire length between Substation 1 and the distribution network at the current switch location is marked as Length 1. Mark the wire length between substation two and the distribution network at the current switch location as length two. Based on length one and length two The line influence factor is determined by formula (1). ; The calculation formula (1) is: ; In the formula, The amplitude adjustment coefficient is determined based on the straight-line distance between the current switch and the nearest substation. The range of values ​​for is (0,1); and It is a proportional adjustment coefficient obtained through manual setting, and and All are not less than 0. ; The value is typically taken as 0.5. The typical value is 0.5; The length of the historical record and length two The average value.

[0026] It is worth noting that this step proposes a method for obtaining line impact factors through bidirectional substation tracing. By tracing the path bidirectionally, it is possible to accurately determine whether both the incoming and outgoing ends of a switch can be traced back to a substation, thereby identifying invalid branch switches, promptly discovering and addressing potential problems in the power system, and improving system reliability and security. Furthermore, the line impact factor LZ calculation model comprehensively considers the differences in line length between the incoming and outgoing ends and their impact on grid operation. By constructing a multi-dimensional evaluation system, it can effectively identify invalid branch switches that appear to be effective interconnecting switches in the topology but do not actually perform their interconnecting function in operation. This data-driven analysis method not only improves the accuracy of identifying invalid branch switches but also provides a scientific basis for the optimized operation of the distribution network.

[0027] It should be noted that the order in which the switches are extracted is based on the manually assigned numbering order.

[0028] It should be noted that in the bidirectional path tracing of the extracted switches, the tracing is performed according to the two ends of the circuit in which the switches are connected, such as tracing the substation along the incoming line and tracing the substation along the outgoing line.

[0029] It should be noted that the nearest substation on the incoming line to the current switch location is marked as Substation 1, and the nearest substation on the outgoing line to the current switch location is marked as Substation 2. The nearest substation to the current switch location is determined based on the length of the power lines in the distribution network.

[0030] This application determines the operating state factors of each switch based on its historical operating state, including: B1: Extract the standard state factor range, and sequentially extract the frequency of switch actions. and frequency of fault repair The standard state factor range is determined based on the historical standard state factors of each switch in the distribution network. B2: Frequency of current switch operation When the frequency exceeds the threshold, the operating state factor of the current switch will be adjusted. The value is set to the maximum value within the standard state factor range; otherwise, jump to B2; where the frequency threshold is determined based on the percentile of the historical operation frequency of each switch in the distribution network; B2: Determine the current switch's operating state factor based on formula (2). ; The calculation formula (2) is: ; In the formula, This is the minimum value within the standard state factor range. This is the maximum value within the standard state factor range; and It is a proportional adjustment coefficient set according to the importance of the frequency of actions and the frequency of fault repairs, and , .

[0031] It is worth noting that this step proposes a switch importance assessment method based on the frequency of operation and the frequency of fault maintenance. By comprehensively considering both the frequency of switch operation and the frequency of fault maintenance, this step constructs an assessment system that can effectively identify the importance of switchgear. The frequency of operation reflects the activity level of the switch in the power system; the more operations, the more critical the switch's role in the system. The frequency of fault maintenance reflects the maintenance status of the switchgear; the higher the maintenance frequency, the more attention the switch receives during operation and the more focus it requires. By comprehensively analyzing these two key indicators, the operating status factor of the switchgear can be obtained, thus indirectly reflecting its importance.

[0032] This step not only provides data support for the optimized operation of the power system but also offers a scientific basis for equipment maintenance and upgrade strategies. In practical applications, distinguishing between effective tie switches and ineffective branch switches can effectively improve the reliability and economy of the power system. Switchgear with frequent operations and high maintenance requirements can be prioritized for focused monitoring; while switchgear with fewer operations and good operating conditions can be given less attention, thereby optimizing resource allocation and improving operation and maintenance efficiency.

[0033] It should be noted that the standard state factor range is determined based on the historical standard state factors of each switch in the distribution network. When the historical standard state factors of each switch in the distribution network are missing or not available, the initial state factor range is used instead of the standard state factor range.

[0034] It should be noted that the frequency threshold is determined based on the percentile of the historical operation frequency of each switch in the distribution network. The percentile is the value at a specific percentile in the historical operation frequency. For example, the operation frequency of the most recent recorded operation of each switch is obtained, and the operation frequency is sorted in ascending order. If the manually set percentile is 90%, then the data at the 90% position is the percentile of the operation frequency. If there is no data at the 90% position, then the data closest to the 90% position is taken as the percentile of the operation frequency.

[0035] It should be noted that, and It is a proportional adjustment coefficient set according to the importance of the frequency of actions and the frequency of fault repairs. Because: What is multiplied is the frequency of the current switch operation. ,and The multiplier is the fault repair frequency, which indicates the current switch maintenance status. In power distribution networks, the operating frequency of a switch is significantly higher than its maintenance frequency. More data leads to higher analysis accuracy. Therefore, the impact data on operating state factors obtained from operating frequency analysis is significantly more accurate than that from maintenance frequency analysis. Thus, this invention focuses on the frequency of operating actions. The set proportional adjustment coefficient is greater than the fault repair frequency. The proportional adjustment coefficient.

[0036] This application determines effective tie switches and ineffective branch switches based on line influence factors and operating status factors, including: Extract the circuit influence factors of the switches sequentially and operating state factors And the influence factors of the line and operating state factors The switching discrimination factor of the current switch is obtained by combining the proportions. ; When the switch is distinguished factor When the value is greater than the standard discrimination factor, the current switch is marked as a valid contact switch; when the switch discrimination factor is greater than the standard discrimination factor, the current switch is marked as a valid contact switch. If the value is not greater than the standard discrimination factor, the current switch is marked as an invalid branch switch; the standard discrimination factor is determined based on the historical standard discrimination factors of manually selected valid connection switches and invalid branch switches.

[0037] It should be noted that the influence factors of the aforementioned lines... and operating state factors In the process of proportional integration, the proportions are manually set, such as the line influence factor. The corresponding ratio is 0.3, operating status factor. The corresponding ratio is 0.7.

[0038] It should be noted that the standard discrimination factor is determined based on the historical standard discrimination factors of manually selected effective contact switches and ineffective branch switches. The manually selected effective contact switches and ineffective branch switches are manually selected and do not require standard discrimination factor identification; they are guaranteed to be effective contact switches and ineffective branch switches.

[0039] This application divides a day into several time periods, including: At 11:59 every day, the historical electricity load of the current area for the previous n days is obtained, and the historical electricity load is integrated into several electricity load groups based on the time points; where n is obtained manually, and is generally taken as 30. The variance of historical electricity loads in each electricity load group is obtained sequentially. It is then determined whether the variance of the electricity load group is less than the electricity load variance threshold. If yes, the characteristic electricity load is calculated by averaging the historical electricity loads in the electricity load group. If no, the historical electricity load with the largest absolute difference from the average value in the electricity load group is removed, and the variance of the remaining data in the electricity load group is recalculated and re-judged. This process continues until the variance of the electricity load group is less than the electricity load variance threshold. Finally, the characteristic electricity load is calculated by averaging the remaining historical electricity loads in the electricity load group. The electricity load variance threshold is set manually. The characteristic electrical loads of each electrical load group are marked as the characteristic electrical loads at the corresponding time points; the characteristic electrical loads are integrated into characteristic load groups, and the average value of the data in the characteristic load group is marked as the characteristic average value; The time points when the characteristic electricity load is greater than the characteristic average value are marked as high load time points, and the time points when the characteristic electricity load is less than the characteristic average value are marked as low load time points. The next day is planned into several reference time periods with a fixed duration, and the proportion of high load time points and the proportion of low load time points in each reference time period are obtained. When the proportion of the first reference time period is greater than the proportion of the second reference time period, the corresponding reference time period is divided into 2m time periods. When the proportion of the first reference time period is not greater than the proportion of the second reference time period, the corresponding reference time period is divided into m time periods. Here, m is obtained manually, and is generally taken as 2. The fixed duration is obtained manually, and is generally taken as 30 minutes.

[0040] It is worth noting that this step, through the analysis and processing of historical electricity load data, can significantly improve the efficiency and accuracy of power system operation. First, by analyzing the historical electricity load data from the previous n days, the patterns and trends of electricity load changes can be effectively identified, thus providing reliable data support for load forecasting. Second, by calculating and filtering the variance of historical electricity loads, outlier data can be eliminated, ensuring data stability and reliability, thereby improving the accuracy of characteristic electricity loads. Comparing characteristic electricity loads with characteristic average values ​​can effectively distinguish between high and low load time points, providing a scientific basis for subsequent time period division.

[0041] In the process of dividing the time period, this step dynamically adjusts the number of time periods based on the ratio of high-load to low-load time points, enabling more flexible adaptation to the changing characteristics of electricity load. For example, dividing periods with a large proportion of high-load time points into more sub-time periods allows for more precise capture of load fluctuation details; while dividing periods with a large proportion of low-load time points into fewer sub-time periods reduces computational complexity and improves overall efficiency. This dynamic division method not only improves the accuracy of load forecasting but also provides more refined support for the optimal allocation and scheduling of power system resources.

[0042] It should be noted that, among the historical loads in the current load group that have the largest absolute difference from the average value, the average value is the average value of the historical loads retained in the current variance judgment step.

[0043] It should be noted that, except for the historical loads in the load group with the largest absolute value of the difference from the average value, if the historical load with the largest absolute value of the difference from the average value in the load group has both the largest and smallest historical loads, the smallest historical load will be removed first.

[0044] It should be noted that if, after removing 90% of the historical electricity load data, the variance of the remaining historical electricity load is still not less than the electricity load variance threshold, then the average value of the original historical electricity load of the electricity load group is used as the characteristic electricity load.

[0045] This application identifies combined influencing factors for each time period based on historical electricity load and environmental data, including: Obtain the characteristic electricity load at each time point in each time period of the next day. Average temperature and average humidity ; compare the temperature over a historical period with the average value The same, and the humidity is the same as the average. The same time period is marked as the comparison time period; obtain the characteristic electricity load at each time point in the comparison time period. Based on characteristic electrical load and characteristic electrical loads The combined impact factor for each time period is determined by formula (3). ;in, Let each time point in the time period be a number, and The range of values ​​for is [1, ... ], This represents the maximum value of the time point numbers within the time period; To compare the numbering of each time point within the time period, and The range of values ​​for is [1, ... ], To compare the maximum value of the time point numbers within the time period; The calculation formula (3) is: ; In the formula, max() is the maximum value function, mode() is the mode function, and BTY is the standard power load set according to the area type and historical power consumption.

[0046] This application identifies series switch groups, including: Integrate continuously connected switches in the distribution network that are physically connected and have no load or transformer in between into a series switch group.

[0047] In this application, series switches are selectively combined into a single virtual switch based on the combined impact factor, including: Extracting and merging impact factors When merging influence factors When the combined impact factor is greater than the standard, all series switches whose sum of impedances is less than the standard impedance are combined into a single virtual switch; when the combined impact factor... When the sum of the impedances of all switches is not greater than the standard merging influence factor, the series switches with impedances not less than the standard impedance are combined into a single virtual switch; where the standard merging influence factor and the standard impedance are both obtained through empirical settings.

[0048] It should be noted that when series switches are combined into a single virtual switch, the equivalent impedance of the single virtual switch is the sum of the impedances of all switches in the series switch group, and the state variables are managed uniformly, such as the closed and open states are synchronized.

[0049] Please see Figure 3 A second aspect of the present invention provides an intelligent identification system for valid switches in a distribution network, comprising: a switch determination module, and a data acquisition module and a redundancy merging module connected to the switch determination module; Data acquisition module: used to acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; Switch determination module: used to perform bidirectional substation tracing for each switch and obtain line influence factors; determine the operating status factors of each switch based on its historical operating status; and determine valid tie switches and invalid branch switches based on line influence factors and operating status factors. Redundancy merging module: It is used to divide a day into several time periods, identify the merging impact factors of each time period based on historical power load and environmental data, identify series switch groups, and selectively combine series switches into a single virtual switch based on the merging impact factors.

[0050] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0051] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A smart identification method for valid switches in a power distribution network, characterized in that, include: Acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; Bidirectional substation tracing was performed on each switch to obtain the line impact factor; The operating status factors of each switch are determined based on the historical operating status of each switch, and the effective tie switches and ineffective branch switches are determined based on the line influence factors and operating status factors. The day is divided into several time periods. Based on historical electricity load and environmental data, the combined impact factors of each time period are identified. Series switch groups are identified, and series switches are selectively combined into a single virtual switch based on the combined impact factors.

2. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The acquisition of the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located, includes: The system extracts the number of days each switch has been in use, the number of times it has been activated, and the number of times it has been repaired since it was put into use from the switch status database. The target number of days is obtained by adding the number of days of new switch error elimination to the number of days of new switch use. The number of days of new switch error elimination is used to eliminate the analysis error caused by the short number of days of new switch use. Divide the number of actions by the number of days of use to obtain the frequency of actions; divide the number of fault repairs by the number of days of use to obtain the frequency of fault repairs. Historical electricity load data for the area where the switch is located was extracted from the regional electricity load database at various time points, and the temperature and humidity of the area where the switch is located were obtained from the weather forecast platform.

3. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The process of bidirectional substation tracing of each switch and obtaining line impact factors includes: A1: Sequentially extract switches, perform bidirectional path tracing on the extracted switches, and determine whether both ends of the bidirectional path tracing can be traced back to the substation; if yes, jump to A2; if no, mark the current switch as an invalid branch switch; the bidirectional path includes the incoming line and the outgoing line. A2: In the current bidirectional path tracing of the switch, the nearest substation on the incoming line to the current switch location is marked as Substation 1, and the nearest substation on the outgoing line to the current switch location is marked as Substation 2; the wire length between Substation 1 and the distribution network at the current switch location is marked as Length 1. Mark the wire length between substation two and the distribution network at the current switch location as length two. Based on length one and length two The line influence factor is determined by formula (1). ; The calculation formula (1) is: ； In the formula, The amplitude adjustment coefficient is determined based on the straight-line distance between the current switch and the nearest substation. The range of values ​​for is (0,1); and It is a proportional adjustment coefficient, and and All are not less than 0. ; The length of the historical record and length two The average value.

4. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The determination of the operating state factor for each switch based on its historical operating state includes: B1: Extract the standard state factor range, and sequentially extract the frequency of switch actions. and frequency of fault repair The standard state factor range is determined based on the historical standard state factors of each switch in the distribution network. B2: Frequency of current switch operation When the frequency exceeds the threshold, the operating state factor of the current switch will be adjusted. The value is set to the maximum value within the standard state factor range; otherwise, jump to B2; where the frequency threshold is determined based on the percentile of the historical operation frequency of each switch in the distribution network; B2: Determine the current switch's operating state factor based on formula (2). ; The calculation formula (2) is: ； In the formula, This is the minimum value within the standard state factor range. This is the maximum value within the standard state factor range; and It is a proportional adjustment coefficient set according to the importance of the frequency of actions and the frequency of fault repairs, and , .

5. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The determination of valid tie switches and invalid branch switches based on line influence factors and operating status factors includes: Extract the circuit influence factors of the switches sequentially and operating state factors And the influence factors of the line and operating state factors The switching discrimination factor of the current switch is obtained by combining the proportions. ; When the switch discrimination factor When the value is greater than the standard discrimination factor, the current switch is marked as a valid contact switch; when the switch discrimination factor is greater than the standard discrimination factor, the current switch is marked as a valid contact switch. If the value is not greater than the standard discrimination factor, the current switch is marked as an invalid branch switch; the standard discrimination factor is determined based on the historical standard discrimination factors of valid tie switches and invalid branch switches.

6. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The plan of dividing a day into several time periods includes: At 11:59 every day, the historical electricity load of the current area for the previous n days is obtained, and the historical electricity load is integrated into several electricity load groups based on the time points. The variance of historical electricity loads in each electricity load group is obtained sequentially. It is then determined whether the variance of the electricity load group is less than the electricity load variance threshold. If yes, the average value of the historical electricity loads in the electricity load group is calculated to obtain the characteristic electricity load. If no, the historical electricity load with the largest absolute value of the difference from the average value in the electricity load group is removed. The variance of the remaining data in the electricity load group is recalculated and the variance is re-determined. This process continues until the variance of the electricity load group is less than the electricity load variance threshold. Finally, the average value of the remaining historical electricity loads in the electricity load group is calculated to obtain the characteristic electricity load. The characteristic electricity loads of each electricity load group are marked as the characteristic electricity loads at the corresponding time points; the characteristic electricity loads are integrated into characteristic load groups, and the average value of the data in the characteristic load group is marked as the characteristic average value; The time points when the characteristic electricity load is greater than the characteristic average value are marked as high load time points, and the time points when the characteristic electricity load is less than the characteristic average value are marked as low load time points. The next day is planned into several reference time periods with a fixed duration, and the proportion of high load time points and the proportion of low load time points in each reference time period are obtained. When the proportion of the first reference time period is greater than the proportion of the second reference time period, the corresponding reference time period is divided into 2m time periods. When the proportion of the first reference time period is not greater than the proportion of the second reference time period, the corresponding reference time period is divided into m time periods.

7. The intelligent identification method for valid switches in a distribution network according to claim 6, characterized in that, The method for identifying combined influencing factors for each time period based on historical electricity load and environmental data includes: Obtain the characteristic electricity load at each time point in each time period of the next day. Average temperature and average humidity ; compare the temperature over a historical period with the average value The same, and the humidity is the same as the average. The same time period is marked as the comparison time period; obtain the characteristic electricity load at each time point in the comparison time period. Based on characteristic electrical load and characteristic electrical loads The combined impact factor for each time period is determined by formula (3). ;in, Let be the number of each time point in the time period, and The range of values ​​for is [1, ... ], This represents the maximum value of the time point numbers within the time period; To compare the numbering of each time point within the time period, and The range of values ​​for is [1, ... ], To compare the maximum value of the time point numbers within the time period; The calculation formula (3) is: ； In the formula, max() is the maximum value function, mode() is the mode function, and BTY is the standard power load set according to the area type and historical power consumption.

8. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The identification of the series switch group includes: Integrate continuously connected switches in the distribution network that are physically connected and have no load or transformer in between into a series switch group.

9. The intelligent identification method for valid switches in a distribution network according to claim 1, characterized in that, The selective combination of series switches into a single virtual switch based on the merging influence factor includes: Extracting and merging impact factors When the combined influence factors When the combined impact factor is greater than the standard, all series switches whose sum of impedances is less than the standard impedance are combined into a single virtual switch; when the combined impact factor... When the sum of the impedances of all switches is not greater than the standard merging influence factor, the series switches that are connected in series and whose sum of impedances is not less than the standard impedance are combined into a single virtual switch.

10. An intelligent identification system for valid switches in a distribution network, used to operate the intelligent identification method for valid switches in a distribution network as described in any one of claims 1 to 9, characterized in that, include: A switch determination module, and a data acquisition module and a redundancy merging module connected to the switch determination module; The data acquisition module is used to acquire the historical operating status of the switch, as well as the historical power load and environmental data of the area where the switch is located; wherein, the historical operating status includes the frequency of operation and the frequency of fault repair, and the environmental data includes temperature and humidity; The switch determination module is used to perform bidirectional substation tracing for each switch and obtain the line influence factor; determine the operating status factor of each switch based on its historical operating status; and determine the valid tie switches and invalid branch switches based on the line influence factor and the operating status factor. The redundant merging module is used to divide a day into several time periods, identify the merging impact factors of each time period based on historical electricity load and environmental data, identify series switch groups, and selectively combine series switches into a single virtual switch based on the merging impact factors.

Images