Low-voltage transformer area topology identification method based on self-adaptive sliding window

Abstract

The invention relates to the technical field of power distribution network automation, in particular to a low-voltage transformer area topology identification method based on a self-adaptive sliding window, and the method comprises the following steps: S1, broadcast timing; s2, acquiring forward active total electric energy data, and performing data preprocessing; s3, electric quantity data of 96 points are calculated based on the forward active total electric energy; s4, performing quartile calculation on the 96-point electric quantity data of the assessment table to obtain an upper quartile and a lower quartile of the electric quantity data; s5, determining electricity consumption data of all the measurement devices based on an adaptive window algorithm; s6, calculating daily electricity consumption Pi of each circuit breaker, and taking out all Pilt; and then, determining a meter box topology and a residual topology. According to the method, the electric quantity data used for data analysis are obtained through the self-adaptive sliding window technology, large electric quantity data are obtained by expanding the time interval for the interval period with small electric quantity, the influence of small electric quantity is avoided to a certain extent, and the problem that the identification accuracy of equipment with small electric quantity is not high is solved.

Description

Technical Field

[0001] The invention relates to the technical field of distribution network automation, and in particular to a low-voltage area topology identification method based on an adaptive sliding window. Background Art

[0002] Traditional low-voltage substation topology identification is mainly based on manual inspections, which is time-consuming, labor-intensive, inefficient, and unable to judge complex topological situations.

[0003] In view of the above situation, there have been many studies on the automatic identification of low-voltage area topology, such as signal injection technology. This technology has high hardware requirements, high cost, and has a certain impact on the power grid. In recent years, some low-voltage area topology technologies based on data analysis have been proposed. Compared with other methods, methods based on data analysis only require marketing data of various devices and do not require any other equipment. However, existing data analysis methods have great drawbacks. For example, a topology identification method for low-voltage area meter networks disclosed in publication number CN112087055A has low identification accuracy for meters with small electricity readings. Summary of the invention

[0004] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a low-voltage station area topology identification method based on an adaptive sliding window.

[0005] The technical solution adopted by the present invention to solve the technical problem is: a low voltage area topology identification method based on an adaptive sliding window comprises the following steps:

[0006] S1: Broadcast time calibration;

[0007] S2: Obtain the total forward active energy data and perform data preprocessing;

[0008] S3: Calculate 96 points of power data based on the total forward active power;

[0009] S4: Perform quartile calculation on the 96-point electricity consumption data in the assessment table to obtain the upper quartile and lower quartile of the electricity consumption data;

[0010] S5: Determine the power consumption data of all measuring devices based on the adaptive window algorithm;

[0011] S6: Calculate the daily power consumption P of each circuit breaker i , remove all P i <1kW·h circuit breaker, then determine the meter box topology and the remaining topology.

[0012] In S1, the time calibration command is broadcast to calibrate the time of the assessment table, intelligent circuit breaker and all user tables in the station area.

[0013] The clock synchronization error in S1 is less than 1 second.

[0014] In the S2, the total forward active energy of the assessment table, intelligent circuit breaker and all user meters in the substation area with a granularity of 1 minute is collected, and data preprocessing is performed.

[0015] The data preprocessing method in S2 is: if there is a power-off record, all data in this time period are deleted.

[0016] In the S3, the total forward active energy collected in the assessment table is subtracted every 15 minutes to obtain 96 points of electricity data with a granularity of 15 minutes.

[0017] In S4, the upper quartile is denoted as Q H , indicating that 1 / 4 of all observed values ​​are greater than it; the lower quartile is denoted as Q L , indicating that 1 / 4 of all observed values ​​are smaller than it.

[0018] The S5 comprises the following sub-steps:

[0019] S51: Get the total forward active energy of the assessment table with a granularity of 1 minute, and take the power consumption of the first 15 minutes as Q1. If Q H ≥Q1≥Q L , then the first window size I1=15; if Q1 L , then expand the window until the power consumption in this window is greater than Q L , record the window size I1 at this time; if Q1>Q H , then narrow the window until the power consumption in this window is less than Q H Or the window size is 1, record the window size I1 at this time, and calculate the power consumption of the first data point according to this window for all smart circuit breakers and electric meters;

[0020] S52: Take the power consumption in the window from I1 to (I1+15) as Q2. If Q H ≥Q2≥Q L , then the first window size I2=15; if Q2 L , then expand the window until the power consumption in this window is greater than Q L , record the window size i2 at this time; if Q2>Q H , then narrow the window until the power consumption in this window is less than Q H Or if the window size is 1, record the window size I2 at this time, and calculate the power consumption of the second data point for all smart circuit breakers and meters according to this window;

[0021] S53: repeat steps S51-S52 to calculate the power consumption of other windows in sequence until the last one, and record the number of windows as n;​​

[0022] S54: Obtain the n-point electricity data of the assessment table, all intelligent circuit breakers and electric meters in the substation.

[0023] The S6 comprises the following sub-steps:

[0024] S61: The daily power consumption P of each circuit breaker is obtained by solving the total active power energy of the circuit breaker with a 1-minute granularity i ;

[0025] S62: Remove all P i <1kW·h circuit breaker: This type of circuit breaker is connected to the subordinate electric meter through 485 lines, and the meter box topology is determined through the meter search function;

[0026] S63: Determine the remaining topology through a bottom-up power conservation algorithm.

[0027] The S63 includes the following sub-steps:

[0028] S631: Use the n-point electricity data of all electricity meters to fit the n-point electricity data of each intelligent circuit breaker, and solve the fitting coefficient by ridge regression. The optimization goal of the ridge regression model is:

[0029]

[0030] Among them, y is the power consumption vector of the circuit breaker, X is the power consumption matrix of all meters, λ is the regularity coefficient, is the coefficient matrix, β is the line loss correction coefficient;

[0031] S632: If the coefficient of the electric meter i corresponding to the circuit breaker j is greater than α, the electric meter i is assigned to the circuit breaker j and the electric meter i is deleted from the remaining list, where the value of α is 0.5 to 1.5;

[0032] S633: Determine the upper level circuit breaker of the electric meter, use the n-point electric quantity data of the determined circuit breaker and the unassigned electric meter to fit the n-point electric quantity data of the unassigned intelligent circuit breaker, and determine the assignment in the manner of S632;

[0033] S634: Repeat step S633 until all assignments are completed.

[0034] Compared with the prior art, the present invention has the following beneficial effects: the present application obtains the power data used for data analysis through the adaptive sliding window technology. For the interval period with low power consumption, the larger power data is obtained by expanding the time interval, which avoids the influence of low power consumption to a certain extent. For the meter box with low power consumption, the meter box circuit breaker and the corresponding electric meter are connected through the 485 line, and the meter box topology is sorted out through the table search function to solve the problem of low accuracy in identifying devices with low power consumption. The topology of other devices is sorted out through the bottom-up power conservation method. The present invention only requires the total positive active power of the metering equipment in the substation with a granularity of 1min, and no other equipment is required. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a flow chart of the present invention;

[0036] Figure 2 It is the division result of the total forward active energy with 1min granularity;

[0037] Figure 3 It is a topological structure diagram. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] Reference Figure 1-Figure 3 ,The low voltage area topology recognition method based on adaptive sliding window includes the following steps:

[0040] S1: Broadcast time synchronization: In S1, the time synchronization of the assessment table, intelligent circuit breaker and all user tables in the station area is carried out through the broadcast time synchronization command. The clock synchronization error in S1 is less than 1 second.

[0041] S2: Obtain the total forward active energy data and perform data preprocessing; S2 collects the total forward active energy of the assessment table, intelligent circuit breaker and all user tables in the substation with a granularity of 1 minute, and performs data preprocessing. The data preprocessing method in S2 is: if there is a power-off record, all data in this time period will be deleted.

[0042] S3: Calculate 96 points of electricity data based on the total forward active electric energy; in S3, the total forward active electric energy collected in the assessment table is subtracted every 15 minutes to obtain 96 points of electricity data with a granularity of 15 minutes.

[0043] S4: quartile calculation is performed on the 96-point electricity consumption data in the assessment table to obtain the upper quartile and lower quartile of the electricity consumption data; the upper quartile in S4 is recorded as Q H , indicating that 1 / 4 of all observed values ​​are greater than it; the lower quartile is denoted as Q L , indicating that 1 / 4 of all observed values ​​are smaller than it.

[0044] S5: Determine the power consumption data of all measuring devices based on the adaptive window algorithm; S5 includes the following sub-steps:

[0045] S51: Get the total forward active energy of the assessment table with a granularity of 1 minute, and take the power consumption of the first 15 minutes as Q1. If Q H ≥Q1≥Q L , then the first window size I1=15; if Q1 L , then expand the window until the power consumption in this window is greater than Q L , record the window size I1 at this time; if Q1>Q H , then narrow the window until the power consumption in this window is less than Q H Or the window size is 1, record the window size I1 at this time, and calculate the power consumption of the first data point according to this window for all smart circuit breakers and electric meters;

[0046] S52: Take the power consumption in the window from I1 to (I1+15) as Q2. If Q H ≥Q2≥Q L , then the first window size I2=15; if Q2 L , then expand the window until the power consumption in this window is greater than Q L , record the window size i2 at this time; if Q2>Q H , then narrow the window until the power consumption in this window is less than Q H Or if the window size is 1, record the window size I2 at this time, and calculate the power consumption of the second data point for all smart circuit breakers and meters according to this window;

[0047] S53: Repeat steps S51-S52 to calculate the power consumption of other windows in sequence until the last one, and record the number of windows as n;

[0048] S54: Obtain the n-point electricity data of the assessment table, all intelligent circuit breakers and electric meters in the substation.

[0049] The division results of the total forward active energy with 1-min granularity are as follows: Figure 2 shown.

[0050] S6: Calculate the daily power consumption P of each circuit breaker i , remove all P i ​​<1kW·h circuit breaker, then determine the meter box topology and the remaining topology.

[0051] The S6 comprises the following sub-steps:

[0052] S61: The daily power consumption P of each circuit breaker is obtained by solving the total active power energy of the circuit breaker with a 1-minute granularity i ;

[0053] S62: Remove all P i <1kW·h circuit breaker: This type of circuit breaker is connected to the subordinate electric meter through 485 lines, and the meter box topology is determined through the meter search function;

[0054] S63: Determine the remaining topology through a bottom-up power conservation algorithm.

[0055] The S63 includes the following sub-steps:

[0056] S631: Use the n-point electricity data of all electricity meters to fit the n-point electricity data of each intelligent circuit breaker, and solve the fitting coefficient by ridge regression. The optimization goal of the ridge regression model is:

[0057]

[0058] Among them, y is the power consumption vector of the circuit breaker, X is the power consumption matrix of all meters, λ is the regularity coefficient, is the coefficient matrix, β is the line loss correction coefficient. The recommended value of β in this application is between 1 and 1.05. When the slip time is short, considering the metering rounding, the recommended value can be enlarged to 0.95-1.1. In the actual operation process, multiple regression processing can be enlarged;

[0059] S632: If the coefficient of the electric meter i corresponding to the circuit breaker j is greater than α, the electric meter i is assigned to the circuit breaker j and the electric meter i is deleted from the remaining list, where the value of α is 0.5 to 1.5;

[0060] S633: Determine the upper level circuit breaker of the electric meter, use the n-point electric quantity data of the determined circuit breaker and the unassigned electric meter to fit the n-point electric quantity data of the unassigned intelligent circuit breaker, and determine the assignment in the manner of S632;

[0061] S634: Repeat step S633 until all assignments are completed.

[0062] The specific topological structure in this application is as follows Figure 3 As shown, it includes energy controller, circuit breaker and electric meter. The daily power consumption of circuit breaker 9 is less than 1kW·h, so 485 lines are used to connect the subordinate electric meters, and the meter box topology is determined by the table search function. The remaining topology is sorted from bottom to top as shown in the figure.

[0063] The above descriptions are only optional embodiments of the present invention, and are not intended to limit the patent scope of the present invention. All equivalent structural changes made using the contents of the present invention's specification and drawings, or directly / indirectly applied in other related technical fields, are included in the patent protection scope of the present invention.

Claims

1. A low voltage area topology identification method based on an adaptive sliding window, characterized in that: The steps include: S1: Broadcast time calibration; S2: Obtain the total forward active energy data and perform data preprocessing; S3: Calculate 96 points of power data based on the total forward active power; S4: Perform quartile calculation on the 96-point electricity consumption data in the assessment table to obtain the upper quartile and lower quartile of the electricity consumption data; S5: Determine the power consumption data of all measuring devices based on the adaptive window algorithm; S6: Calculate the daily power consumption P of each circuit breaker i , remove all P i <1kW·h circuit breaker, then determine the meter box topology and the remaining topology.

2. The low voltage area topology identification method based on adaptive sliding window according to claim 1 is characterized in that: In S1, the time calibration command is broadcast to calibrate the time of the assessment table, intelligent circuit breaker and all user tables in the station area.

3. The low voltage area topology identification method based on adaptive sliding window according to claim 1 or 2, characterized in that: The clock synchronization error in S1 is less than 1 second.

4. The method for identifying low voltage area topology based on adaptive sliding window according to claim 1 or 2, characterized in that: In the S2, the total forward active energy of the assessment table, intelligent circuit breaker and all user meters in the substation area with a granularity of 1 minute is collected, and data preprocessing is performed.

5. The method for identifying low voltage area topology based on adaptive sliding window according to claim 4, characterized in that: The data preprocessing method in S2 is: if there is a power-off record, all data in this time period are deleted.

6. The method for identifying low voltage area topology based on adaptive sliding window according to claim 5, characterized in that: In the S3, the total forward active energy collected in the assessment table is subtracted every 15 minutes to obtain 96 points of electricity data with a granularity of 15 minutes.

7. The method for identifying low voltage area topology based on adaptive sliding window according to claim 6, characterized in that: In S4, the upper quartile is denoted as Q H , indicating that 1 / 4 of all observed values ​​are greater than it; the lower quartile is denoted as Q L , indicating that 1 / 4 of all observed values ​​are smaller than it.

8. The method for identifying low voltage area topology based on adaptive sliding window according to claim 7, characterized in that: The S5 comprises the following sub-steps: S51: Get the total forward active energy of 1-minute granularity in the assessment table, and take the power consumption of the first 15 minutes as Q1. If Q H ≥Q1≥Q L , then the first window size I1=15; if Q1 L , then expand the window until the power consumption in this window is greater than Q L , record the window size I1 at this time; if Q1>Q H , then narrow the window until the power consumption in this window is less than Q H Or the window size is 1, record the window size I1 at this time, and calculate the power consumption of the first data point according to this window for all smart circuit breakers and electric meters;​ S52: Take the power consumption in the window from I1 to (I1+15) as Q2. If Q H ≥Q2≥Q L , then the first window size I2=15; if Q2 L , then expand the window until the power consumption in this window is greater than Q L , record the window size I2 at this time; if Q2>Q H , then narrow the window until the power consumption in this window is less than Q H Or if the window size is 1, record the window size I2 at this time, and calculate the power consumption of the second data point for all smart circuit breakers and meters according to this window;​ S53: repeat steps S51-S52 to calculate the power consumption of other windows in sequence until the last one, and record the number of windows as n; S54: Obtain the n-point electricity data of the assessment table, all intelligent circuit breakers and electric meters in the substation.

9. The method for identifying low voltage area topology based on adaptive sliding window according to claim 8, characterized in that: The S6 comprises the following sub-steps: S61: The daily power consumption P of each circuit breaker is obtained by solving the total active power energy of the circuit breaker with a 1-minute granularity i ; S62: Remove all P i For circuit breakers with a capacity of less than 1kW·h, such circuit breakers are connected to subordinate electric meters through 485 lines, and the meter box topology is determined through the meter search function; S63: Determine the remaining topology through a bottom-up power conservation algorithm.

10. The method for identifying low voltage area topology based on adaptive sliding window according to claim 9, characterized in that: The S63 includes the following sub-steps: S631: Use the n-point electricity data of all electricity meters to fit the n-point electricity data of each intelligent circuit breaker, and solve the fitting coefficient by ridge regression. The optimization goal of the ridge regression model is: Among them, y is the power consumption vector of the circuit breaker, X is the power consumption matrix of all meters, λ is the regularity coefficient, is the coefficient matrix, β is the line loss correction coefficient; S632: If the coefficient of the electric meter i corresponding to the circuit breaker j is greater than α, the electric meter i is assigned to the circuit breaker j and the electric meter i is deleted from the remaining list, where the value of α is 0.5 to 1.5; S633: Determine the upper level circuit breaker of the electric meter, use the n-point electric quantity data of the determined circuit breaker and the unassigned electric meter to fit the n-point electric quantity data of the unassigned intelligent circuit breaker, and determine the assignment in the manner of S632; S634: Repeat step S633 until all assignments are completed.

Images