User electric energy meter missort identification method and device based on electric power big data and medium

Abstract

The invention discloses a user electric energy meter missort identification method and device based on electric power big data and a medium, and belongs to the technical field of electric power information. The method comprises the following steps: determining a user set of electric energy meters in the same meter box based on power user information in a power master station database; obtaining active power of each user in the user set at a plurality of different time points every day in n days before and after meter replacement, and daily power consumption of each user in n days before and after meter replacement; calculating an active power correlation coefficient between each user and each user according to the active power of each user before meter changing and the active power of each user after meter changing; calculating an electricity consumption fluctuation coefficient of each user before and after meter replacement according to the daily electricity consumption; and identifying the electric energy meter missort pair of the user set according to the active power correlation coefficient and the electric quantity fluctuation coefficient. According to the method, the electric energy meter missort pair can be accurately identified by analyzing the electric power big data, the operation and maintenance cost is reduced, and the identification efficiency of the electric energy meter missort is improved.

Description

Technical Field

[0001] This application belongs to the field of power information technology, and in particular relates to a method, device and medium for identifying user electricity meters based on power big data. Background Technology

[0002] In the electricity metering system for low-voltage users of the power grid, residential communities generally adopt a one-meter-per-household metering model, meaning each household is equipped with one electricity meter to record its electricity consumption. The meters of multiple adjacent households are usually installed together in the same meter box in the building's distribution room. However, electricity meters have a limited lifespan and need to be replaced periodically. During replacement, the problem of meter mismatch can easily occur, mainly manifested as errors in record-keeping or incorrect wiring, leading to confusion in user billing. For example, user A might pay user B's electricity bill, and vice versa.

[0003] Currently, power grid companies primarily use manual inspection to resolve cross-connection issues. Maintenance personnel need to communicate with users and conduct on-site inspections, using the power-off function of the electricity meter to disconnect the power. Users then check if their own household has also lost power. If their household has not lost power but neighboring households have, cross-connection can be identified. However, this method requires a significant investment of manpower for on-site communication and operation, resulting in high maintenance costs. Furthermore, the efficiency of this door-to-door inspection process is limited, making the overall inspection process slow and unable to meet the actual need for quickly resolving cross-connection issues. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a user electricity meter cross-connection identification system, method, and apparatus based on big data of electricity, so as to improve the identification efficiency of cross-connection of electricity meters and reduce operation and maintenance costs.

[0005] Firstly, this application provides a method for identifying cross-connection of user electricity meters based on big data of electricity, including: Based on the power user information in the power master station database, a set of users whose electricity meters are located in the same meter box is identified; Obtain the active power of each user in the user set at multiple different time points each day before and after the meter replacement n days, as well as the daily electricity consumption of each user before and after the meter replacement n days. Calculate the active power correlation coefficient between each user and each other based on the active power of each user before meter replacement and the active power of each user after meter replacement; Calculate the electricity consumption fluctuation coefficient for each user before and after the meter replacement based on the daily electricity consumption. The electricity meter pairs of the user set are identified based on the active power correlation coefficient and the electricity fluctuation coefficient.

[0006] According to the user meter cross-connection identification method based on power big data in this application, the user set whose meters are located in the same meter box is determined by power user information in the power master station database; the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days are obtained, as well as the daily electricity consumption of each user before and after the meter replacement n days; the active power correlation coefficient between each user and each other is calculated based on the active power of each user before the meter replacement and the active power of each user after the meter replacement; the electricity consumption fluctuation coefficient of each user before and after the meter replacement is calculated based on the daily electricity consumption; and the cross-connection pairs of the user set are identified based on the active power correlation coefficient and the electricity consumption fluctuation coefficient. This application's embodiments can accurately identify electricity meters of users located within the same meter box by comprehensively analyzing power big data. By calculating the active power correlation coefficient, it can capture the similarity or difference in the user's electricity consumption behavior characteristics before and after meter replacement. Furthermore, by calculating the power fluctuation coefficient, it verifies the rationality of the user's electricity consumption behavior from the perspective of power change. Combining the active power correlation coefficient and the power fluctuation coefficient for comprehensive identification and cross-verification from different dimensions, it can accurately identify electricity meter cross-connections. Compared with traditional methods, it does not require a large amount of manpower for on-site inspection, reducing operation and maintenance costs and improving the efficiency of electricity meter cross-connection identification.

[0007] According to one embodiment of this application, determining the set of users whose electricity meters are located in the same meter box based on electricity user information in the power master station database includes: Obtain power user information from the power master station database; Obtain the user's user code from the power user information; Based on the user code, the meter box code of the meter box where the user's electricity meter is located is found from the electricity user information; Based on the meter box code, the user code of the electricity meter located in the meter box is found from the electricity user information to obtain the user set in the same meter box.

[0008] In this embodiment, by extracting power user information from the power master station database and relying on the association between user information and meter box codes in the power master station database, users whose electricity meters are located in the same meter box can be accurately located.

[0009] According to one embodiment of this application, the active power at multiple different time points each day is 96 points of active power.

[0010] According to one embodiment of this application, the step of calculating the active power correlation coefficient between each user and other users based on the active power of each user before meter replacement and the active power of each user after meter replacement includes: Based on the active power of the 96 points, calculate the average active power of each user in the same time period n days before the meter replacement and the average active power of each user in the same time period n days after the meter replacement. The correlation coefficient of active power between each user and each other is calculated based on the average active power of each user before the meter replacement and the average active power of each user after the meter replacement.

[0011] In this embodiment, by using 96 active power data points, i.e., active power values ​​recorded every 15 minutes throughout the day, the electricity consumption behavior characteristics of users at different time periods can be reflected in detail. By calculating the average active power of each user in the same time period n days before and n days after the meter replacement, short-term data fluctuations are further smoothed out. On the one hand, the data scale is reduced, and on the other hand, the high-frequency fluctuations of active power are reduced, thus reducing the interference of random factors on the analysis results. Based on the average active power before and after the meter replacement, the correlation coefficient of active power between users can be calculated, which can quantify the similarity or difference between users' electricity consumption behavior characteristics, thereby accurately identifying cross-connection of electricity meters.

[0012] According to one embodiment of this application, the step of calculating the average active power of each user in the same time period n days before meter replacement and the average active power of each user in the same time period n days after meter replacement based on the 96 active power points includes: Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; The average active power of each user during the same time period in the n days prior to meter replacement is calculated based on the average active power of each user during the n days prior to meter replacement; and the average active power of each user during the same time period in the n days after meter replacement is calculated based on the average active power of each user during the n days after meter replacement.

[0013] In this embodiment, the hourly average value of 96 active power data points for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated. This preserves the user's electricity consumption characteristics at different time periods, effectively reduces the amount of data, lowers the computational complexity, and smooths out short-term fluctuations. Based on the hourly average active power, the average active power for the same time period for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated, further refining the user's electricity consumption characteristics and more accurately reflecting the overall trend of the user's electricity consumption behavior before and after the meter replacement.

[0014] According to one embodiment of this application, based on the formula: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1= (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; Among them, P A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-q-n-96,1 P A-q-n-96,2 P A-q-n-96,3 P A-q-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n before user A's meter was replaced. A-h-n-24,1 P represents the average active power of user A in the first hour of the nth day after meter replacement. A-h-n-96,1 P A-h-n-96,2 P A-h-n-96,3 P A-h-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n after user A replaced the meter.

[0015] According to one embodiment of this application, based on the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (P A-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n Calculate the average active power of each user in the same time period n days before meter replacement, and calculate the average active power of each user in the same time period n days after meter replacement. Among them, P A-q-24,1 P represents the average active power of user A in the first hour of the n days prior to meter replacement. A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-h-24,1 P represents the average active power of user A in the first hour, n days after the meter was replaced. A-h-n-24,1 This represents the average active power of user A in the first hour of the nth day after the meter was replaced.

[0016] According to one embodiment of this application, the step of calculating the active power correlation coefficient between each user and all other users based on the average active power of each user before meter replacement and the average active power of each user after meter replacement includes: Calculate the average active power of each user in the i-th hour of the n-th day before the meter replacement, and calculate the average active power of each user in the i-th hour of the n-th day after the meter replacement; Calculate the active power correlation coefficient between each user and all other users based on the average active power of each user in the i-th hour of the n-th day before meter replacement and the average active power of each user in the i-th hour of the n-th day after meter replacement.

[0017] According to one embodiment of this application, based on the formula:

[0018]

[0019] Calculate the active power correlation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set. This represents the active power correlation coefficient between user a and user b. This represents the average active power array for user a before the meter was replaced. This represents the average active power of user a in the i-th hour of the n-th day prior to meter replacement. This represents the average value of the average power in the average active power array before user A replaced the meter. This represents the average active power of user a in the i-th hour, n days after the meter was replaced. This represents the average active power array after user b replaced the meter. This represents the average active power of user b in the i-th hour, n days after the meter was replaced. This represents the average value of the average active power array after user b changed the table.

[0020] According to one embodiment of this application, the step of calculating the electricity consumption fluctuation coefficient for each user before and after the meter replacement based on the daily electricity consumption includes: Calculate the average daily electricity consumption for each user in the n days before the meter replacement, and calculate the average daily electricity consumption for each user in the n days after the meter replacement, based on the daily electricity consumption. The electricity consumption fluctuation coefficient for each user before and after the meter replacement is calculated based on the average daily electricity consumption for each user in the n days before the meter replacement and the average daily electricity consumption for each user in the n days after the meter replacement.

[0021] According to one embodiment of this application, based on the formula: Q a-q =( Q a-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h =( Q a-h-1 + Q a-h-2 +……+ Qa-h-n ) / n KQ a =min(Q a-q Q a-h ) / max(Q a-q Q a-h ) Calculate the electricity consumption fluctuation coefficient for each user before and after the meter replacement; Among them, Q a-q Q represents the average daily electricity consumption of user a over the n days prior to meter replacement. a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h Q represents the average daily electricity consumption over n days after the meter replacement. a-h-n KQ represents the electricity consumption of user a on day n after the meter was replaced. a This represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement.

[0022] According to one embodiment of this application, the step of identifying the electricity meter pairs of the user set based on the active power correlation coefficient and the power fluctuation coefficient includes: The cross-user evaluation coefficient between each user and other users is calculated based on the active power correlation coefficient and the power fluctuation coefficient. Two users whose evaluation coefficients meet preset conditions are identified as electricity meter pair.

[0023] In this embodiment, the active power correlation coefficient can reflect the similarity of users' electricity consumption behavior characteristics before and after meter replacement, while the power fluctuation coefficient further verifies the rationality of users' electricity consumption behavior from the perspective of power consumption changes. Combining these two indicators, the calculated cross-user evaluation coefficient can comprehensively reflect the possibility of cross-users. By setting preset conditions to screen user pairs that meet the conditions, the misjudgment that may be caused by a single indicator is reduced. It makes full use of various information in power big data and combines information from different dimensions for cross-validation, thereby improving the accuracy of cross-user identification.

[0024] According to one embodiment of this application, based on the formula: KE a-b =KC(P a-q-24 , P b-h-24 )×KQ b Calculate the cross-user evaluation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set, KE a-b KC(P) represents the cross-user rating coefficient between user a and user b. a-q-24 , P b-h-24 KQ represents the active power correlation coefficient between user a and user b.b This represents the fluctuation coefficient of user b's electricity consumption before and after the meter replacement.

[0025] According to one embodiment of this application, the preset condition is:

[0026] Where user i and user j are different users in the user set. This represents the cross-user evaluation coefficient between user i and user i'. This represents the cross-user rating coefficient between user j and user j. This represents the cross-user evaluation coefficient between user i and user j. This represents the user rating coefficient between user j and user i.

[0027] In this embodiment, by constructing the above-mentioned four-fold inequality constraints as preset conditions, theoretically, a user's own electricity consumption behavior characteristics should be highly consistent. Therefore, the cross-user evaluation coefficient between the same user should be close to 1. Due to various interference factors that may exist in the actual electricity consumption process, such as changes in temporary electrical equipment, the correlation coefficient may not reach 1 completely, but it should still remain at a high level, that is, greater than 0.6. When the cross-user evaluation coefficient between the same user is less than 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are significantly different. Under normal circumstances, the electricity consumption behavior characteristics of the same user before and after the meter replacement should be relatively similar. Therefore, it can be preliminarily confirmed that the metered user may have changed. Furthermore, if the cross-user evaluation coefficient between two different users is large, that is, greater than or equal to 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are highly consistent. Under normal circumstances, the user behavior characteristics between different users should be significantly different. By using the above-mentioned four-fold inequality constraints for joint judgment, normal users and users who may have cross-user problems can be accurately distinguished.

[0028] Secondly, this application provides a user electricity meter cross-connection identification device based on power big data, comprising: The determination module is used to determine the set of users whose electricity meters are located in the same meter box based on the electricity user information in the power master station database; The acquisition module is used to acquire the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days, as well as the daily electricity consumption of each user before and after the meter replacement n days. The first calculation module is used to calculate the active power correlation coefficient between each user and each other based on the active power of each user before the meter replacement and the active power of each user after the meter replacement. The second calculation module is used to calculate the electricity consumption fluctuation coefficient of each user before and after the meter replacement based on the daily electricity consumption. The identification module is used to identify the electricity meter pairs of the user set based on the active power correlation coefficient and the power fluctuation coefficient.

[0029] According to the user electricity meter cross-connection identification device based on power big data of this application, the user set whose electricity meters are located in the same meter box is determined by power user information in the power master station database; the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days are obtained, as well as the daily electricity consumption of each user before and after the meter replacement n days; the active power correlation coefficient between each user and each other is calculated based on the active power of each user before the meter replacement and the active power of each user after the meter replacement; the electricity consumption fluctuation coefficient of each user before and after the meter replacement is calculated based on the daily electricity consumption; and the cross-connection pairs of electricity meters in the user set are identified based on the active power correlation coefficient and the electricity consumption fluctuation coefficient. This application's embodiments can accurately identify electricity meters of users located within the same meter box by comprehensively analyzing power big data. By calculating the active power correlation coefficient, it can capture the similarity or difference in the user's electricity consumption behavior characteristics before and after meter replacement. Furthermore, by calculating the power fluctuation coefficient, it verifies the rationality of the user's electricity consumption behavior from the perspective of power change. Combining the active power correlation coefficient and the power fluctuation coefficient for comprehensive identification and cross-verification from different dimensions, it can accurately identify electricity meter cross-connections. Compared with traditional methods, it does not require a large amount of manpower for on-site inspection, reducing operation and maintenance costs and improving the efficiency of electricity meter cross-connection identification.

[0030] According to one embodiment of this application, the determining module is further configured to: Obtain power user information from the power master station database; Obtain the user's user code from the power user information; Based on the user code, the meter box code of the meter box where the user's electricity meter is located is found from the electricity user information; Based on the meter box code, the user code of the electricity meter located in the meter box is found from the electricity user information to obtain the user set in the same meter box.

[0031] According to one embodiment of this application, the active power at multiple different time points each day is 96 points of active power.

[0032] According to one embodiment of this application, the first calculation module is further configured to: Based on the active power of the 96 points, calculate the average active power of each user in the same time period n days before the meter replacement and the average active power of each user in the same time period n days after the meter replacement. The correlation coefficient of active power between each user and each other is calculated based on the average active power of each user before the meter replacement and the average active power of each user after the meter replacement.

[0033] According to one embodiment of this application, the first calculation module is further configured to: Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; The average active power of each user during the same time period in the n days prior to meter replacement is calculated based on the average active power of each user during the n days prior to meter replacement; and the average active power of each user during the same time period in the n days after meter replacement is calculated based on the average active power of each user during the n days after meter replacement.

[0034] According to one embodiment of this application, the first calculation module is further configured to: According to the formula: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1 = (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; Among them, P A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-q-n-96,1 P A-q-n-96,2 P A-q-n-96,3 P A-q-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n before user A's meter was replaced. A-h-n-24,1 P represents the average active power of user A in the first hour of the nth day after meter replacement. A-h-n-96,1 P A-h-n-96,2 P A-h-n-96,3 P A-h-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n after user A replaced the meter.

[0035] According to one embodiment of this application, the first calculation module is further configured to: According to the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (PA-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n Calculate the average active power of each user in the same time period n days before meter replacement, and calculate the average active power of each user in the same time period n days after meter replacement. Among them, P A-q-24,1 P represents the average active power of user A in the first hour of the n days prior to meter replacement. A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-h-24,1 P represents the average active power of user A in the first hour, n days after the meter was replaced. A-h-n-24,1 This represents the average active power of user A in the first hour of the nth day after the meter was replaced.

[0036] According to one embodiment of this application, the first calculation module is further configured to: Calculate the average active power of each user in the i-th hour of the n-th day before the meter replacement, and calculate the average active power of each user in the i-th hour of the n-th day after the meter replacement; Calculate the active power correlation coefficient between each user and all other users based on the average active power of each user in the i-th hour of the n-th day before meter replacement and the average active power of each user in the i-th hour of the n-th day after meter replacement.

[0037] According to one embodiment of this application, the first calculation module is further configured to: According to the formula:

[0038]

[0039] Calculate the active power correlation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set. This represents the active power correlation coefficient between user a and user b. This represents the average active power array for user a before the meter was replaced. This represents the average active power of user a in the i-th hour of the n-th day prior to meter replacement. This represents the average value of the average power in the average active power array before user A replaced the meter. This represents the average active power of user a in the i-th hour, n days after the meter was replaced. This represents the average active power array after user b replaced the meter. This represents the average active power of user b in the i-th hour, n days after the meter was replaced. This represents the average value of the average active power array after user b changed the table.

[0040] According to one embodiment of this application, the second calculation module is further configured to: Calculate the average daily electricity consumption for each user in the n days before the meter replacement, and calculate the average daily electricity consumption for each user in the n days after the meter replacement, based on the daily electricity consumption. The electricity consumption fluctuation coefficient for each user before and after the meter replacement is calculated based on the average daily electricity consumption for each user in the n days before the meter replacement and the average daily electricity consumption for each user in the n days after the meter replacement.

[0041] According to one embodiment of this application, the second calculation module is further configured to: According to the formula: Q a-q =( Q a-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h =( Q a-h-1 + Q a-h-2 +……+ Q a-h-n ) / n KQ a =min(Q a-q Q a-h ) / max(Q a-q Q a-h ) Calculate the electricity consumption fluctuation coefficient for each user before and after the meter replacement; Among them, Q a-q Q represents the average daily electricity consumption of user a over the n days prior to meter replacement. a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h Q represents the average daily electricity consumption over n days after the meter replacement. a-h-n KQ represents the electricity consumption of user a on day n after the meter was replaced. a This represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement.

[0042] According to one embodiment of this application, the identification module is further configured to: The cross-user evaluation coefficient between each user and other users is calculated based on the active power correlation coefficient and the power fluctuation coefficient. Two users whose evaluation coefficients meet preset conditions are identified as electricity meter pair.

[0043] According to one embodiment of this application, the identification module is further configured to: According to the formula: KE a-b =KC(P a-q-24 , P b-h-24 )×KQ b Calculate the cross-user evaluation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set, KE a-b KC(P) represents the cross-user rating coefficient between user a and user b. a-q-24 , P b-h-24 KQ represents the active power correlation coefficient between user a and user b. b This represents the fluctuation coefficient of user b's electricity consumption before and after the meter replacement.

[0044] According to one embodiment of this application, the preset condition is:

[0045] Where user i and user j are different users in the user set. This represents the cross-user evaluation coefficient between user i and user i'. This represents the cross-user rating coefficient between user j and user j. This represents the cross-user evaluation coefficient between user i and user j. This represents the user rating coefficient between user j and user i.

[0046] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the user electricity meter cross-connection identification method based on power big data as described in the first aspect above.

[0047] Fourthly, this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the user electricity meter cross-connection identification method based on power big data as described in the first aspect above.

[0048] Fifthly, this application provides a chip, which includes a processor and a communication interface, the communication interface being coupled to the processor, and the processor being used to run programs or instructions to implement the user electricity meter cross-connection identification method based on power big data as described in the first aspect above.

[0049] In a sixth aspect, this application provides a computer program product, including a computer program that, when executed by a processor, implements the user electricity meter serial identification method based on power big data as described in the first aspect above.

[0050] The above-described one or more technical solutions in the embodiments of this application have at least one of the following technical effects: According to the user meter cross-connection identification method based on power big data in this application, the user set whose meters are located in the same meter box is determined by power user information in the power master station database; the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days are obtained, as well as the daily electricity consumption of each user before and after the meter replacement n days; the active power correlation coefficient between each user and each other is calculated based on the active power of each user before the meter replacement and the active power of each user after the meter replacement; the electricity consumption fluctuation coefficient of each user before and after the meter replacement is calculated based on the daily electricity consumption; and the cross-connection pairs of the user set are identified based on the active power correlation coefficient and the electricity consumption fluctuation coefficient. This application's embodiments can accurately identify electricity meters of users located within the same meter box by comprehensively analyzing power big data. By calculating the active power correlation coefficient, it can capture the similarity or difference in the user's electricity consumption behavior characteristics before and after meter replacement. Furthermore, by calculating the power fluctuation coefficient, it verifies the rationality of the user's electricity consumption behavior from the perspective of power change. Combining the active power correlation coefficient and the power fluctuation coefficient for comprehensive identification and cross-verification from different dimensions, it can accurately identify electricity meter cross-connections. Compared with traditional methods, it does not require a large amount of manpower for on-site inspection, reducing operation and maintenance costs and improving the efficiency of electricity meter cross-connection identification.

[0051] Furthermore, in some embodiments, by extracting power user information from the power master station database and relying on the association between user information and meter box codes in the power master station database, users whose electricity meters are located in the same meter box can be accurately located.

[0052] Furthermore, in some embodiments, by employing 96 points of active power data, i.e., active power values ​​recorded every 15 minutes throughout the day, the electricity consumption behavior characteristics of users at different time periods can be reflected in detail. By calculating the average active power of each user in the same time period n days before and n days after the meter replacement, short-term data fluctuations are further smoothed out. On the one hand, the data scale is reduced, and on the other hand, the high-frequency fluctuations of active power are reduced, thus reducing the interference of accidental factors on the analysis results. Based on the average active power before and after the meter replacement, the correlation coefficient of active power between users can be calculated, which can quantify the similarity or difference between users' electricity consumption behavior characteristics, thereby accurately identifying cross-connection of electricity meters.

[0053] Furthermore, in some embodiments, the hourly average value of 96 active power data points for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated. This not only preserves the user's electricity consumption characteristics at different time periods but also effectively reduces the amount of data, lowers the computational complexity, and smooths out short-term fluctuations. Based on the hourly average active power, the average active power for the same time period for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated, further refining the user's electricity consumption characteristics and more accurately reflecting the overall trend of changes in the user's electricity consumption behavior before and after the meter replacement.

[0054] Furthermore, in some embodiments, the active power correlation coefficient can reflect the similarity of users' electricity consumption behavior characteristics before and after meter replacement, while the power fluctuation coefficient further verifies the rationality of users' electricity consumption behavior from the perspective of power consumption changes. Combining these two indicators, the calculated cross-user evaluation coefficient can comprehensively reflect the possibility of cross-users. By setting preset conditions to filter user pairs that meet the conditions, the misjudgment that may be caused by a single indicator is reduced. It makes full use of various information in power big data, and cross-validates information from different dimensions to improve the accuracy of cross-user identification.

[0055] Furthermore, in some embodiments, by constructing the above-mentioned four-fold inequality constraints as preset conditions, theoretically, a user's own electricity consumption behavior characteristics should be highly consistent. Therefore, the cross-user evaluation coefficient between the same user should be close to 1. Due to various interference factors that may exist in the actual electricity consumption process, such as changes in temporary electrical equipment, the correlation coefficient may not reach 1 completely, but it should still remain at a high level, that is, greater than 0.6. When the cross-user evaluation coefficient between the same user is less than 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are significantly different. Under normal circumstances, the electricity consumption behavior characteristics of the same user before and after the meter replacement should be relatively similar, so it can be preliminarily confirmed that the metered user may have changed. Furthermore, if the cross-user evaluation coefficient between two different users is large, that is, greater than or equal to 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are highly consistent. Under normal circumstances, the user behavior characteristics between different users should be significantly different. By using the above-mentioned four-fold inequality constraints for joint judgment, normal users and users who may have cross-user problems can be accurately distinguished.

[0056] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart illustrating the user meter cross-connection identification method based on power big data provided in this application embodiment; Figure 2 This is a schematic diagram of the energy meter inside the meter box K provided in an embodiment of this application; Figure 3 This application provides a schematic diagram of the active power curves of 96 points for user A 7 days before and 7 days after meter replacement. Figure 4 This application provides a schematic diagram of the 14-day average active power curves for user A 7 days before and 7 days after meter replacement. Figure 5 This is a schematic diagram of the 24-hour average active power curves of user A for 7 days before and 7 days after meter replacement, provided in an embodiment of this application. Figure 6 This is a schematic diagram from a two-dimensional perspective of the 24-hour average active power curves of user A for 7 days before and 7 days after meter replacement, provided in an embodiment of this application. Figure 7 This is a schematic diagram of the structure of the user electricity meter serial identification device based on power big data provided in the embodiments of this application; Figure 8 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0059] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0060] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0061] The following description, in conjunction with the accompanying drawings, details the user electricity meter cross-connection identification method, device, and medium based on power big data provided in this application, through specific embodiments and application scenarios.

[0062] Among them, the user electricity meter serial identification method based on power big data can be applied to the terminal, specifically executed by the hardware or software in the terminal.

[0063] The terminal includes, but is not limited to, portable communication devices such as mobile phones or tablets with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads). It should also be understood that, in some embodiments, the terminal may not be a portable communication device, but rather a desktop computer with touch-sensitive surfaces (e.g., touchscreen displays and / or touchpads).

[0064] The following embodiments describe a terminal including a display and a touch-sensitive surface. However, it should be understood that the terminal may include one or more other physical user interface devices such as a physical keyboard, mouse, and joystick.

[0065] The user electricity meter cross-connection identification method based on power big data provided in this application embodiment can be executed by an electronic device or a functional module or entity in an electronic device that can implement the user electricity meter cross-connection identification method based on power big data. The electronic devices mentioned in this application embodiment include, but are not limited to, mobile phones, tablets, computers, cameras and wearable devices. The following uses an electronic device as the execution subject to illustrate the user electricity meter cross-connection identification method based on power big data provided in this application embodiment.

[0066] like Figure 1 As shown, the user electricity meter cross-connection identification method based on power big data includes: steps 110, 120, 130, 140 and 150.

[0067] Step 110: Determine the set of users whose electricity meters are located in the same meter box based on the electricity user information in the power master station database.

[0068] The power master station database is the core information platform used by power grid companies to store and manage various types of data in the power system. It encompasses massive amounts of data from all stages, including power generation, transmission, substation, distribution, and consumption. The data in the power master station database is typically stored in structured or semi-structured form and is maintained and queried through a professional database management system. The power master station database provides powerful data support for power grid operation monitoring, fault diagnosis, data analysis, and business decision-making. For example, it can record key information such as the real-time operating status of power equipment and user electricity consumption.

[0069] In the power master station database, power user information is a collection of basic data related to user electricity consumption, covering multiple dimensions such as user identity, electrical equipment, and electricity usage relationships. This includes user codes, user types (e.g., residential, commercial, industrial users), user addresses, meter codes, and meter box codes. Specifically, user codes distinguish different users; meter codes distinguish different meters; and meter box codes distinguish different meter boxes.

[0070] In some embodiments, determining a set of users whose electricity meters are located in the same meter box based on electricity user information in the power master station database includes: Retrieve power user information from the power master station database; Obtain the user's user code from the electricity user information; The meter box code of the user's electricity meter is retrieved from the electricity user information based on the user code. Based on the meter box code, the user code of the electricity meter located in the meter box is found from the electricity user information, thus obtaining the set of users in the same meter box.

[0071] In this embodiment, after obtaining the power user information from the power master station database, the user's user code can be extracted from this information. For example, the user code N of a certain power user A can be found and obtained from the power user information. yh-A Then, based on the user code N... yh-A Find and obtain the code N of meter box K to which power user A belongs from the power user information. bx-K Next, the set of all user codes {N} belonging to meter box K is retrieved from the electricity user information. yh-all-bx-K} refers to the set of users in the same table box.

[0072] like Figure 2 As shown, assume that 6 electricity meters are installed in meter box K, corresponding to the electricity meters of users A, B, C, D, E, and F. The user codes corresponding to these electricity meters are N. yh-A N yh-B N yh-C N yh-D N yh-E and N yh-F Then the user set {N} in box K yh-all-bx-K}={N yh-A N yh-B N yh-C N yh-D N yh-E N yh-F}

[0073] In this embodiment, by extracting power user information from the power master station database and relying on the association between user information and meter box codes in the power master station database, users whose electricity meters are located in the same meter box can be accurately located.

[0074] In some embodiments, filtering can also be based on the meter box code. Since a meter box may house multiple users' electricity meters, a query can be used to find all user codes with the same meter box code. For example, suppose the meter box code for meter box K is N. bx-K All meter boxes with code N can be filtered out from the power station database. bx-K These users are the user set located in the same table box.

[0075] Step 120: Obtain the active power of each user at multiple different time points each day before and after the meter replacement for n days in the user set, as well as the daily electricity consumption of each user for n days before and after the meter replacement.

[0076] Active power is a physical quantity that measures the actual work capacity of a user's electrical equipment, reflecting the user's real electrical load at a specific point in time and corresponding to the actual operating state of the user's electrical equipment. For example, when a user turns on multiple electrical appliances in their home, the active power will increase accordingly; while when the user turns off some appliances, the active power will decrease. By monitoring active power, we can understand the user's electricity consumption behavior characteristics at different times. For example, when a user turns on lights, televisions, and other devices after waking up in the morning, the active power will show a significant increase; while at night when resting, the active power will decrease. In cross-meter detection scenarios, the time series data of active power can characterize the user's electricity consumption behavior characteristics (such as peak electricity consumption periods, equipment start-up and shutdown patterns, etc.). The differences or similarities in the user's electricity consumption behavior characteristics before and after meter replacement are important bases for determining whether the electricity meter is being used by multiple users.

[0077] Daily electricity consumption is the total electrical energy consumed by a user within a calendar day, reflecting the user's overall electricity usage throughout the day. For example, a user's electricity consumption may be relatively low on weekdays, while on weekends or holidays, due to spending more time at home, electricity consumption may increase significantly. By analyzing daily electricity consumption, we can understand a user's electricity usage habits. In cross-meter identification scenarios, daily electricity consumption not only reflects the user's overall electricity consumption scale, but also demonstrates the stability of electricity usage behavior through changes in values ​​before and after meter replacement. For example, under normal circumstances, a user's daily electricity consumption after meter replacement should maintain a relatively consistent fluctuation trend with that before the replacement. If a sudden change occurs, it may indicate an anomaly in the correspondence between the electricity meter and the user. Therefore, daily electricity consumption is an important data point for verifying the rationality of electricity usage behavior from a macro perspective.

[0078] In this embodiment, active power and daily electricity consumption can be obtained from the power station database. During data acquisition, the user's electricity meter replacement time is used as a reference point, tracing back n days to define the period before replacement and forward n days to define the period after replacement. The power station database typically collects active power at fixed intervals (usually 15 minutes, 30 minutes, or 1 hour) through the electricity meter's remote communication module. This data is stored in a historical database in an ordered manner according to timestamps. During retrieval, active power records from the n days before and n days after replacement can be filtered from the data stream corresponding to each user's electricity meter file within the user set.

[0079] Where n can be 7 days, 15 days, 30 days, etc. The value of n can be selected to cover the number of days of a complete weekly electricity consumption cycle, so as to fully capture the changes in the user's electricity consumption behavior before and after the meter replacement.

[0080] Step 130: Calculate the correlation coefficient of active power between each user and each other based on the active power of each user before the meter replacement and the active power of each user after the meter replacement.

[0081] In this embodiment, the active power correlation coefficient is a statistical indicator that measures the degree of linear correlation between two time series. In a power system, active power is presented in the form of a time series, reflecting the electricity consumption behavior of users at different points in time. The value of the active power correlation coefficient can range from -1 to 1, where 1 represents a perfect positive correlation, -1 represents a perfect negative correlation, and 0 represents no linear correlation. For example, if the trends of active power changes for two users are highly consistent before and after meter replacement, then the active power correlation coefficient of these two time series will be close to 1; if the trends are completely opposite, the active power correlation coefficient will be close to -1; when the active power correlation coefficient is close to 0, it indicates that the two time series have almost no significant correlation. By calculating the active power correlation coefficient, the similarity or difference in the electricity consumption behavior of users before and after meter replacement can be quantified. In cross-user identification scenarios, under normal circumstances, the active power sequences of the same user before and after meter replacement should have a high positive correlation coefficient (because the main electricity user remains unchanged and the behavior pattern is stable), with the active power correlation coefficient close to 1. However, if cross-user identification occurs, the active power sequence of the original user after meter replacement may show a higher correlation with the sequences of other users before meter replacement. Therefore, cross-user identification can be located by the difference in active power correlation coefficient.

[0082] In this embodiment, it is necessary to calculate the active power correlation coefficient between each user and all other users. Here, "each user" refers to any user in the user set, while "all other users" includes the user itself and all other users. For each user in the user set, it is necessary to calculate not only its own active power correlation coefficient before and after meter replacement, but also its active power correlation coefficient with other users before and after meter replacement. For example, if the user set includes users A, B, C, D, E, and F, for user A, the active power correlation coefficient between user A and user A needs to be calculated based on user A's active power before and after meter replacement. Similarly, the active power correlation coefficient between user A and user B needs to be calculated based on user A's active power before and after meter replacement, and so on. The same applies to users B, C, D, E, and F.

[0083] For the same user, calculating the correlation coefficient between the active power sequence before and after the meter replacement directly verifies the consistency of electricity consumption behavior before and after the replacement. If the active power correlation coefficient is significantly low, it may indicate that the user's electricity meter may have been replaced by another user. For different users, calculating the active power correlation coefficients of cross-combinations such as user A's active power sequence before the meter replacement and user B's active power sequence after the meter replacement, and user A's sequence after the meter replacement and user B's active power sequence before the meter replacement, can further identify abnormal correlations. For example, if user A's active power sequence after the meter replacement is highly correlated with user C's active power sequence before the meter replacement, but has a low correlation with its own sequence before the meter replacement, there may be cross-metering between user A and user C. Through the logic of the above full-volume correlation analysis, potential cross-metering relationships can be investigated from multiple dimensions.

[0084] In this embodiment, each user has an active power time series containing multiple time points. For example, suppose user A's active power series before meter replacement is as follows: , Indicates the time point before the timetable was changed. The active power, the active power sequence after user B's meter replacement is as follows: , Indicates the time point after the table was changed. If the active power is known, the correlation coefficient between the active power of user A and user B can be calculated using the following formula. :

[0085] in, express The average value of the sequence. express The average value of the sequence.

[0086] Step 140: Calculate the electricity consumption fluctuation coefficient for each user before and after the meter replacement based on daily electricity consumption.

[0087] The electricity consumption fluctuation coefficient is an indicator that measures the dispersion of a user's daily electricity consumption over a period of time. It reflects the stability of electricity consumption behavior. The smaller the electricity consumption fluctuation coefficient, the more balanced the daily electricity consumption and the more stable the electricity consumption pattern. The larger the electricity consumption fluctuation coefficient, the more significant the difference in daily electricity consumption and the greater the fluctuation in the electricity consumption pattern. In cross-user identification scenarios, under normal circumstances, the electricity consumption habits of the same user are continuous, and the electricity consumption fluctuation coefficient before and after the meter replacement should remain relatively consistent (e.g., both are at a low level). If cross-user identification occurs, due to the change of the electricity user, the daily electricity consumption after the meter replacement may show drastically different fluctuation characteristics compared to before the meter replacement. For example, the electricity consumption fluctuation coefficient may suddenly increase or deviate significantly from historical levels. Therefore, the electricity consumption coefficient can serve as an important basis for determining whether an electricity meter is being used by a different user.

[0088] By calculating the electricity consumption fluctuation coefficient, the rationality of users' electricity consumption behavior before and after meter replacement is verified from the perspective of the stability of macro-scale electricity consumption, providing another dimension of quantitative basis for cross-user identification.

[0089] In some embodiments, calculating the electricity consumption fluctuation coefficient for each user before and after meter replacement based on daily electricity consumption includes: Calculate the average daily electricity consumption for each user in the n days before the meter replacement, and calculate the average daily electricity consumption for each user in the n days after the meter replacement. The electricity consumption fluctuation coefficient for each user before and after the meter replacement is calculated based on the average daily electricity consumption for each user in the n days before the meter replacement and the average daily electricity consumption for each user in the n days after the meter replacement.

[0090] In some embodiments, the daily electricity consumption sequence for the user n days before the meter replacement and the daily electricity consumption sequence for the user n days after the meter replacement can be obtained. For example, for user a, the daily electricity consumption sequence for the user n days before the meter replacement is Q. a-q-1 Q a-q-2 Q a-q-n The daily electricity consumption sequence for n days after the meter replacement is Q. a-h-1 Q a-h-2 Q a-h-n , where Q a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h-nThis represents user A's electricity consumption on day n after the meter replacement. We can calculate user A's average daily electricity consumption for the n days before and after the meter replacement, and then calculate the electricity consumption fluctuation coefficient before and after the meter replacement based on the changes in user A's average daily electricity consumption for the n days before and after the meter replacement. The specific calculation process can be found in the formula: Q a-q =( Q a-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h =( Q a-h-1 + Q a-h-2 +……+ Q a-h-n ) / n KQ a =min(Q a-q Q a-h ) / max(Q a-q Q a-h ) Among them, Q a-q Q represents the average daily electricity consumption of user a over the n days prior to meter replacement. a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h Q represents the average daily electricity consumption over n days after the meter replacement. a-h-n KQ represents the electricity consumption of user a on day n after the meter was replaced. a This represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement.

[0091] In some embodiments, the electricity consumption fluctuation coefficient can also be calculated in other ways. For example, the variance of daily electricity consumption for user a in the n days before meter replacement can be calculated, the variance of daily electricity consumption for user a in the n days after meter replacement can be calculated, and then the variance of daily electricity consumption can be used to replace the average daily electricity consumption to calculate the electricity consumption fluctuation coefficient of user a before and after meter replacement using the above formula. Alternatively, the mode, median, or other statistical measures of daily electricity consumption can be used to replace the average daily electricity consumption to calculate the electricity consumption fluctuation coefficient of user a before and after meter replacement using the above formula. This application embodiment does not limit this approach.

[0092] In some embodiments, the electricity consumption fluctuation coefficient can also be defined as the ratio of the change in total electricity consumption before and after meter replacement to the total electricity consumption before meter replacement. This application does not limit this.

[0093] Step 150: Identify the electricity meter pairs of the user set based on the active power correlation coefficient and the power fluctuation coefficient.

[0094] In the preceding steps, the active power correlation coefficient between each user and other users, as well as the electricity consumption fluctuation coefficient before and after each user's meter replacement, have been calculated. The active power correlation coefficient and the electricity consumption fluctuation coefficient reflect the user's electricity consumption behavior characteristics from different perspectives. Under normal circumstances, if the correspondence between the electricity meter and the user remains unchanged, the active power sequence of the same user before and after the meter replacement should have a high active power correlation coefficient, typically close to 1, and the electricity consumption fluctuation coefficient before and after the replacement should be small, indicating consistent electricity consumption stability. However, when cross-user transactions occur, the actual user corresponding to the electricity meter changes, causing specific anomalies in both coefficients: the correlation between the active power sequence of the original user after the meter replacement and its active power sequence before the replacement decreases significantly, but it may show a high correlation with the sequence of another user before the replacement (because it actually corresponds to the electricity consumption behavior of another user); when cross-user transactions occur, the electricity consumption fluctuation coefficient before and after the replacement will also increase significantly, due to the abrupt change in electricity consumption stability caused by differences in the electricity consumption habits of different users. Based on this, cross-user pairs of electricity meters in a user set can be identified using the following methods. In some embodiments, thresholds can be set for the active power correlation coefficient and the power fluctuation coefficient, respectively. For example, the threshold for the active power correlation coefficient can be set to 0.7, where a value greater than or equal to 0.7 indicates similar electricity consumption behavior, and a value less than 0.7 indicates significant differences in electricity consumption; the threshold for the power fluctuation coefficient can be set to 0.3, where a value less than or equal to 0.3 indicates small changes in electricity consumption stability, and a value greater than 0.3 indicates a sudden change in electricity consumption stability.

[0095] For a user set within the same table box, we iterate through all possible user pairs. For example, if the user set includes user A, user B, user C, user D, user E, and user F, then we can iterate through pairs such as user i and user j. , , .

[0096] If the active power correlation coefficient between user i and user j is less than 0.7, and the active power correlation coefficient between user i and user j is greater than or equal to 0.7, it indicates that user i's electricity consumption behavior after the meter replacement is closer to that of user j. Furthermore, the electricity consumption fluctuation coefficient before and after the meter replacement for both user i and user j is greater than 0.3, suggesting that user i and user j may be a serial user pair, but further verification is needed.

[0097] Specifically, for the selected parallel user pairs, the judgment can be strengthened through reverse verification: check whether the correlation coefficient between the active power sequence of user j before the meter replacement and the active power sequence of user i after the meter replacement is also greater than or equal to 0.7, and whether the correlation coefficient between user j's active power before the meter replacement and its own active power after the meter replacement is less than 0.7. If the correlation coefficient between user j's active power sequence before the meter replacement and user i's active power sequence after the meter replacement is also greater than or equal to 0.7, and the correlation coefficient between user j's active power before the meter replacement and its own active power after the meter replacement is less than 0.7, then it can be finally determined that user i and user j are parallel user pairs.

[0098] According to the user meter cross-connection identification method based on power big data in this application, the user set whose meters are located in the same meter box is determined by power user information in the power master station database; the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days are obtained, as well as the daily electricity consumption of each user before and after the meter replacement n days; the active power correlation coefficient between each user and other users is calculated based on the active power of each user before the meter replacement and the active power of each user after the meter replacement; the electricity consumption fluctuation coefficient of each user before and after the meter replacement is calculated based on the daily electricity consumption; and the cross-connection pairs of meters in the user set are identified based on the active power correlation coefficient and the electricity consumption fluctuation coefficient. This application's embodiments can accurately identify electricity meters of users located within the same meter box by comprehensively analyzing power big data. By calculating the active power correlation coefficient, it can capture the similarity or difference in the user's electricity consumption behavior characteristics before and after meter replacement. Furthermore, by calculating the power fluctuation coefficient, it verifies the rationality of the user's electricity consumption behavior from the perspective of power change. Combining the active power correlation coefficient and the power fluctuation coefficient for comprehensive identification and cross-verification from different dimensions, it can accurately identify electricity meter cross-connections. Compared with traditional methods, it does not require a large amount of manpower for on-site inspection, reducing operation and maintenance costs and improving the efficiency of electricity meter cross-connection identification.

[0099] In some embodiments, the active power at multiple different time points each day is the active power at 96 points. Calculate the active power correlation coefficient between each user and all other users based on the active power before and after the meter replacement, including: Calculate the average active power of each user in the same time period n days before meter replacement and the average active power of each user in the same time period n days after meter replacement based on the active power of 96 points. The correlation coefficient of active power between each user and each other is calculated based on the average active power of each user before the meter replacement and the average active power of each user after the meter replacement.

[0100] In this embodiment, the electricity meter typically records the user's active power at 15-minute intervals, resulting in 96 data points per day, or 24 hours × 4 time points / hour = 96 points.

[0101] For example, taking n days as 7 days, and the user set including user A, user B, user C, user D, user E, and user F, the active power P of 96 points for the 7 days before user A replaced their electricity meter can be queried and exported from the power master station database. A-q-1~7-96 The active power P at 96 points 7 days after the electricity meter was replaced A-h-1~7-96 Similarly, export the data for users B, C, D, E, and F 7 days before the electricity meter replacement. B-q-1~7-96 P C-q-1~7-96 P D-q-1~7-96 P E-q-1~7-96 P F-q-1~7-96 The active power P at 96 points 7 days after the electricity meter was replaced. B-h-1~7-96 P C-h-1~7-96 P D-h-1~7-96 P E-h-1~7-96 P F-h-1~7-96 This forms the 14-day × 96-point active power set {P} for users A to F. A~F-2×7~96}: {P A~F-14-96}={ P A-q-1~7-96 , P A-h-1~7-96 ,P B-q-1~7-96 , P B-h-1~7-96 ,… P C-q-1~7-96 , P C-h-1~7-96 ,P D-q-1~7-96 , P D-h-1~7-96 ,… P E-q-1~7-96 , P E-h-1~7-96 , P F-q-1~7-96 , P F-h-1~7-96} in: {P A-q-1~7-96}={ P A-q-1-96 , P A-q-2-96 , P A-q-3-96 , P A-q-4-96 , P A-q-5-96 , P A-q-6-96 ,P A-q-7-96 ,} {P A-h-1~7-96}={ P A-h-1-96 , P A-h-2-96 , P A-h-3-96 , P A-h-4-96 , P A-h-5-96 , P A-h-6-96 ,PA-h-7-96 ,} ... {P F-h-1~7-96}={ P F-h-1-96 , P F-h-2-96 , P F-h-3-96 , P F-h-4-96 , P F-h-5-96 , P F-h-6-96 ,P F-h-7-96 ,} in: P A-q-1-96 =[ P A-q-1-96,1 , P A-q-1-96,2 , P A-q-1-96,3 ,……, P A-q-1-1,96 ] P A-h-1-96 =[ P A-h-1-96,1 , P A-h-1-96,2 , P A-h-1-96,3 ,……, P A-h-1-1,96 ] ... P F-h-7-96 =[ P F-h-7-96,1 , P F-h-7-96,2 , P F-h-7-96,3 ,……, P F-h-7-1,96 ] Among them, P A-q-1-96 This represents the active power (P) at 96 points on the first day before user A replaced the electricity meter. A-q-1-96,1 This represents the first data point out of 96 power data points on the first day before user A replaced their electricity meter. A-h-1-96 This represents the active power (P) at 96 points on the first day after user A replaced their electricity meter. A-h-1-96,1 This represents the first data point out of 96 power data points on the first day after user A replaced their meter, and so on. For households A to F in meter box K, the total power data for users A to F retrieved and exported from the power station is: 6 households × 14 days × 96 points = 8064 data points.

[0102] To reduce data volatility and extract more stable electricity consumption behavior characteristics, the average active power of each user during the same time period in the n days before meter replacement and the average active power of each user during the same time period in the n days after meter replacement can be calculated. For example, the 96 active power data points for each day can be grouped by time period, such as one group per hour, one group per two hours, etc., and the average active power of each user during the same time period in the n days before meter replacement and the average active power of each user during the same time period in the n days after meter replacement can be calculated. Taking one group per hour as an example, the average active power of the 96 active power data points in the 00:00-01:00 time period for each day in the n days before meter replacement can be calculated to obtain the average active power of the 00:00-01:00 time period in the n days before meter replacement. Then, based on the average active power of different time periods in the n days before meter replacement, an average active power sequence is formed. Referring to the description in step 130, the active power correlation coefficient between each user and other users can be calculated based on the average active power sequence of the n days before meter replacement and the average active power sequence of the n days after meter replacement.

[0103] In this embodiment, by using 96 active power data points, i.e., active power values ​​recorded every 15 minutes throughout the day, the electricity consumption behavior characteristics of users at different time periods can be reflected in detail. By calculating the average active power of each user in the same time period n days before and n days after the meter replacement, short-term data fluctuations are further smoothed out. On the one hand, the data scale is reduced, and on the other hand, the high-frequency fluctuations of active power are reduced, thus reducing the interference of random factors on the analysis results. Based on the average active power before and after the meter replacement, the correlation coefficient of active power between users can be calculated, which can quantify the similarity or difference between users' electricity consumption behavior characteristics, thereby accurately identifying cross-connection of electricity meters.

[0104] In some embodiments, calculating the average active power of each user in the same time period n days before meter replacement and the average active power of each user in the same time period n days after meter replacement based on 96 active power points includes: Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; The average active power of each user during the same time period in the n days prior to meter replacement is calculated based on the average active power of each user during the n days prior to meter replacement; and the average active power of each user during the same time period in the n days after meter replacement is calculated based on the average active power of each user during the n days after meter replacement.

[0105] In this embodiment, each hour can be considered as a time period, and the average active power of each user in each hour for n days before the meter replacement and the average active power of each user in each hour for n days after the meter replacement can be calculated.

[0106] In some embodiments, the formula can be used: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1 = (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; Among them, P A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-q-n-96,1 P A-q-n-96,2 P A-q-n-96,3 P A-q-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n before user A's meter was replaced. A-h-n-24,1 P represents the average active power of user A in the first hour of the nth day after meter replacement. A-h-n-96,1 P A-h-n-96,2 P A-h-n-96,3 P A-h-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n after user A replaced the meter.

[0107] According to the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (P A-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n Calculate the average active power of each user in the same time period n days before meter replacement, and calculate the average active power of each user in the same time period n days after meter replacement. Among them, P A-q-24,1 P represents the average active power of user A in the first hour of the n days prior to meter replacement. A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-h-24,1 P represents the average active power of user A in the first hour, n days after the meter was replaced. A-h-n-24,1 This represents the average active power of user A in the first hour of the nth day after the meter was replaced.

[0108] For example, the average of the power data points 1-4 on the first day before the meter replacement for user A can be calculated to obtain the first data point; the average of the data points 5-8 can be calculated to obtain the second data point; and so on, until the average of the data points 93-96 can be calculated to obtain the 24th data point, forming the 24-point average active power array P for user A on the first day before the meter replacement. A-q-1-24 As shown in the following formula: P A-q-1-24,1 = (P A-q-1-96,1 +P A-q-1-96,2 +P A-q-1-96,3 +P A-q-1-96,4 ) / 4 P A-q-1-24,2 =(P A-q-1-96,5 +P A-q-1-96,6 +P A-q-1-96,7 +P A-q-1-96,8 ) / 4 ... P A-q-1-24,24 = (P A-q-1-96,93 +P A-q-1-96,94 +P A-q-1-96,95 +P A-q-1-96,96 ) / 4 The array is formed as follows: P A-q-1-24 =[ P A-q-1-24,1 , P A-q-1-24,2 , P A-q-1-24,3 , ……P A-q-1-24,24 ] Among them, P A-q-1-24,1 P represents the average active power of household A in the first hour of the first day before the meter was replaced. A-q-1-96,1 This represents the active power of the first data point in the 96-point power data of the first day before user A replaced the meter, and so on, P A-q-1-24,24 This represents the average active power, P, of household A in the 24th hour of the first day before meter replacement. A-q-1-96,93 This represents the active power at the 93rd data point of the 96th power data point on the first day before user A replaced the meter.

[0109] In the same way, the 24-hour average active power array for household A from day 2 to day 7 before the meter replacement can be obtained: P A-h-2-24 P A-h-3-24 ...P A-h-7-24 Together, they form the 24-point average active power set {P} for household A during the 1st to 7th days before the meter replacement. A-q-1~7-24}: {P A-q-1~7-24}={ P A-q-1-24 , P A-q-2-24 , P A-q-3-24 , P A-q-4-24 , P A-q-5-24 , P A-q-6-24 ,PA-q-7-24} Using the same method, we can obtain the average active power set {P} for household A from day 1 to day 7 after the meter replacement. A-h-1~7-24}: {P A-h-1~7-24}={ P A-h-1-24 , P A-h-2-24 , P A-h-3-24 , P A-h-4-24 , P A-h-5-24 , P A-h-6-24 ,P A-h-7-24} Among them, P A-h-1-24 This represents the average active power data at 24:00 on the first day after the meter was replaced for Household A.

[0110] Using the same method, we can obtain the average power data sets for households B to F before and after meter replacement, covering days 1 to 7, and then construct the 14-day × 24-point average active power data set {P} for households A to F. A~F-2×7~24}: {P A~F-14-24}={ P A-q-1~7-24 , P A-h-1~7-24 , P B-q-1~7-24 , P B-h-1~7-24 , P C-q-1~7-24 ,P C-h-1~7-24 ,… P D-q-1~7-24 , P D-h-1~7-24 , P E-q-1~7-24 , P E-h-1~7-24 , P F-q-1~7-24 , P F-h-1~7-24} {P A~F-14-24 The dataset contains 6 households × 14 days × 24 = 2016 data points. The dataset is first processed by averaging the 96 daily power data points for each user across 4 points. This reduces the data size and, by averaging, combines the power consumption recorded every 15 minutes into an hourly average power consumption, reducing high-frequency fluctuations in power consumption and enabling preliminary extraction of user power consumption curve features.

[0111] Take the average active power of user A for the first hour from day 1 to day 7 before meter replacement: P A-q-1-24,1 P A-q-2-24,1 P A-q-3-24,1 ... P A-q-7-24,1 Find the average P of these 7 numbers. A-q-24,1 : P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1+……+P A-q-7-24,1 ) / 7 P A-q-24,1 This represents the average active power of user A during the first hour of each day from day 1 to day 7 before the meter was replaced.

[0112] Using the same method, the average power of user A in the 2nd, 3rd, ..., 24th hours before the meter replacement can be obtained, and the average active power array P of user A before the meter replacement can be constructed. A-q-24 : P A-q-24 =[ P A-q-24,1 , P A-q-24,2 , P A-q-24,3 ,……, P A-q-24,24 ] The same method can be used to obtain the average active power array P for user A from the first to the second 24 hours after the meter was replaced. A-h-24 : P A-h-24 =[ P A-h-24,1 , P A-h-24,2 , P A-h-24,3 ,……, P A-h-24,24 ] P A-h-24,1 This represents the average active power during the first hour of each day from day 1 to day 7 after user A replaced the meter.

[0113] Using the same method, we can obtain the average active power arrays for users B~F before and after meter replacement for the first 1 to 24 hours, and form the 24-hour average active power set {P} for users A~F before and after meter replacement. A~F-24}: {P A~F-24}={ P A-q-24 , P A-h-24 , P B-q-24 , P B-h-24 , P C-q-24 , P C-h-24 ,… P D-q-24 , P D-h-24 , P E-q-24 , P E-h-24 , P F-q-24 , P F-h-24 ,} {P A~F-24The dataset contains 6 households × 2 × 24 hours = 288 data points. The average active power of each user over the first 7 days is averaged into 24 values, and the average active power over the next 7 days is averaged into 24 values. This second processing of the dataset further reduces the data size. Calculating the hourly power average over 7 days reduces power fluctuations and allows for further extraction of the characteristics of the user's electricity consumption curve.

[0114] like Figures 3-6 As shown, Figure 3 A schematic diagram of the active power curves at 96 points for user A 7 days before and 7 days after meter replacement; Figure 4 To process the data set by averaging 4 points of power data from 96 points per day for user A, the following diagram shows the 14-day average active power curves for user A before and after the meter replacement. Figure 5 To obtain a second dataset by averaging the 24-hour average active power of user A for the first 7 days and the 24-hour average active power for the last 7 days, a three-dimensional schematic diagram of the 24-hour average active power curves for user A before and after the meter replacement is generated. Figure 6 This is a two-dimensional schematic diagram of the 24-hour average active power curves for user A 7 days before and 7 days after meter replacement. Figures 3-6 It can be seen that by processing the data twice, the amount of data was reduced, and short-term data fluctuations were further smoothed out.

[0115] In this embodiment, the hourly average value of 96 active power data points for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated. This preserves the user's electricity consumption characteristics at different time periods, effectively reduces the amount of data, lowers the computational complexity, and smooths out short-term fluctuations. Based on the hourly average active power, the average active power for the same time period for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated, further refining the user's electricity consumption characteristics and more accurately reflecting the overall trend of the user's electricity consumption behavior before and after the meter replacement.

[0116] In some embodiments, the active power correlation coefficient between each user and all other users is calculated based on the average active power of each user before meter replacement and the average active power of each user after meter replacement, including: Calculate the average active power of each user in the i-th hour of the n-th day before the meter replacement, and calculate the average active power of each user in the i-th hour of the n-th day after the meter replacement; Calculate the active power correlation coefficient between each user and all other users based on the average active power of each user in the i-th hour of the n-th day before meter replacement and the average active power of each user in the i-th hour of the n-th day after meter replacement.

[0117] In some embodiments, the formula can be used:

[0118]

[0119] Calculate the active power correlation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set. This represents the active power correlation coefficient between user a and user b. This represents the average active power array for user a before the meter was replaced. This represents the average active power of user a in the i-th hour of the n-th day prior to meter replacement. This represents the average value of the average power in the average active power array before user A replaced the meter. This represents the average active power of user a in the i-th hour, n days after the meter was replaced. This represents the average active power array after user b replaced the meter. This represents the average active power of user b in the i-th hour, n days after the meter was replaced. This represents the average value of the average active power array after user b changed the table.

[0120] Taking a user set including user A, user B, user C, user D, user E, and user F as an example, , , It can be equal to b, or it can be not equal to b. For user A, the active power correlation coefficient is calculated as follows:

[0121]

[0122] ...

[0123] in, This represents the active power correlation coefficient between user A and user A's. This represents the average active power array for user A before the meter was replaced. This represents the average active power of user A in the i-th hour of the n-th day prior to meter replacement. This represents the average value of the average power in the average active power array before user A replaced the meter. This represents the average active power array after user B replaced the meter. This represents the average active power of user b in the i-th hour, n days after the meter was replaced. This represents the average value of the average active power array after user B changed the table, and so on.

[0124] In this embodiment, the active power correlation coefficients between user A and A~F can be used to form an active power correlation coefficient array KC for user A. A : KC A =[KC(P A-q-24 , P A-h-24 ), KC (P A-q-24 , P B-h-24 ),... KC (P A-q-24 , P C-h-24 ), KC (P A-q-24 , P D-h-24 ),... KC (P A-q-24 , P E-h-24 ), KC (P A-q-24 , P F-h-24 )] Similarly, the active power correlation coefficient array KC between the active power sequences of users B~F before meter replacement and the active power sequences of users A~F after meter replacement can be obtained respectively. B KC C KC D KC E and KC F Combined into a correlation coefficient matrix KC A~F The expression is as follows: KC A~F ={KC A KC B KC C KC D KC E KC F} It should be noted that the active power correlation coefficient between user A and user B is different from the active power correlation coefficient between user B and user A. The active power correlation coefficient between user A and user B is expressed as follows: The active power correlation coefficient between user B and user A is expressed as: .

[0125] In some embodiments, identifying the electricity meter pairs of a user set based on the active power correlation coefficient and the power fluctuation coefficient includes: The cross-user evaluation coefficient between each user and other users is calculated based on the active power correlation coefficient and the power fluctuation coefficient. Two users whose cross-meter evaluation coefficient meets the preset conditions are identified as cross-meter pairs.

[0126] In this embodiment, the household evaluation coefficient is a quantitative indicator that integrates the active power correlation coefficient and the power fluctuation coefficient. Its function is to integrate the features of the two dimensions into a single dimension, thereby simplifying the complexity of multi-indicator judgment, while retaining the complementarity of key information.

[0127] In some embodiments, the inter-household evaluation coefficient can be obtained by weighted summing of the active power correlation coefficient and the power fluctuation coefficient, such as: KE a-b =α[1-KC(P a-q-24 , P b-h-24 )]+βKQ b Wherein, user a and user b are either the same user or different users in the user set, KE a-b KC(P) represents the cross-user rating coefficient between user a and user b. a-q-24 , P b-h-24 KQ represents the active power correlation coefficient between user a and user b. b This represents the fluctuation coefficient of user b's electricity consumption before and after the meter replacement, with α and β representing weighting parameters.

[0128] In the following embodiments, it can also be based on the formula: KE a-b =KC(P a-q-24 , P b-h-24 )×KQ b Calculate the cross-user evaluation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set, KE a-b KC(P) represents the cross-user rating coefficient between user a and user b. a-q-24 , P b-h-24 KQ represents the active power correlation coefficient between user a and user b. b This represents the fluctuation coefficient of user b's electricity consumption before and after the meter replacement.

[0129] For example, for user A, the cross-user evaluation coefficient between user A and users A~F is calculated as follows: KE A =[KE A-A , KE A-B , KE A-C , KE A-D , KE A-E , KE A-F ] =[ KC(P A-q-24 , P A-h-24 )×KQ A , KC(PA-q-24 , P B-h-24 )×KQ B ,… KC(P A-q-24 , P C-h-24 )×KQ C , KC(P A-q-24 , P D-h-24 )×KQ D ,… KC(P A-q-24 , P E-h-24 )×KQ E , KC(P A-q-24 , P F-h-24 )×KQ F ] Among them, KE A KE represents the array of user rating coefficients between user A and users A~F. A-A KE represents the cross-user rating coefficient between user A and user A. A-B KE represents the cross-user rating coefficient between user A and user B, ... A-F KC(P) represents the cross-user rating coefficient between user A and user F. A-q-24 , P A-h-24 KC(P) represents the active power correlation coefficient between user A and user A's. A-q-24 , P B-h-24 KC(P) represents the active power correlation coefficient between user A and user B, ..., KC(P) A-q-24 ,P F-h-24 KQ represents the active power correlation coefficient between user A and user F. A KQ represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement. F This represents the fluctuation coefficient of user F's electricity consumption before and after the meter replacement.

[0130] Similarly, the cross-user evaluation coefficient array KE between users B~F and users A~F can be calculated separately. B KE C KE D KE E KE F Form the user evaluation coefficient matrix KE for users A~F before and after the form change. A~F .in: KE A~F ={KE A ; KE B ; KE C ; KE D ; KE E ; KE F} =[ KE A-A , KE A-B , KE A-C , KE A-D , KE A-E , KE A-F ; KE B-A , KE B-B , KE B-C , KE B-D , KE B-E , KE B-F ; KE C-A , KE C-B , KE C-C , KE C-D , KE C-E , KE C-F ; KE D-A , KE D-B , KE D-C , KE D-D , KE D-E , KE D-F ; KE E-A , KE E-B , KE E-C , KE E-D , KE E-E , KE E-F ; KE F-A , KE F-B , KE F-C , KE F-D , KE F-E , KE F-F ] Among them, KE B-A KE represents the cross-user rating coefficient between user B and user A, ... B-F KE represents the cross-user evaluation coefficient between user B and user F. C-A KE represents the cross-user rating coefficient between user C and user A, ... C-F KE represents the cross-user evaluation coefficient between user C and user F. D-A KE represents the cross-user rating coefficient between user D and user A. D-F KE represents the cross-user evaluation coefficient between user D and user F. E-A , represents the cross-user evaluation coefficient between user E and user A, ..., KE E-F KE represents the cross-user evaluation coefficient between user E and user F. F-AKE represents the cross-user evaluation coefficient between user F and user A, ... F-F This represents the cross-user evaluation coefficient between user F and user F.

[0131] In some embodiments, the preset conditions are:

[0132] Where user i and user j are different users in the user set. This represents the cross-user evaluation coefficient between user i and user i'. This represents the cross-user rating coefficient between user j and user j. This represents the cross-user evaluation coefficient between user i and user j. This represents the user rating coefficient between user j and user i.

[0133] In this embodiment, it is assumed that the user set includes users A to F, and among users A to F there are two distinct users i and j:

[0134] If user i and user j both meet the above preset conditions, then user i and user j are marked as a pair of users, and the user identification is completed.

[0135] In this embodiment, by constructing the above-mentioned four-fold inequality constraints as preset conditions, theoretically, a user's own electricity consumption behavior characteristics should be highly consistent. Therefore, the cross-user evaluation coefficient between the same user should be close to 1. Due to various interference factors that may exist in the actual electricity consumption process, such as changes in temporary electrical equipment, the correlation coefficient may not reach 1 completely, but it should still remain at a high level, that is, greater than 0.6. When the cross-user evaluation coefficient between the same user is less than 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are significantly different. Under normal circumstances, the electricity consumption behavior characteristics of the same user before and after the meter replacement should be relatively similar. Therefore, it can be preliminarily confirmed that the metered user may have changed. Furthermore, if the cross-user evaluation coefficient between two different users is large, that is, greater than or equal to 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are highly consistent. Under normal circumstances, the user behavior characteristics between different users should be significantly different. By using the above-mentioned four-fold inequality constraints for joint judgment, normal users and users who may have cross-user problems can be accurately distinguished.

[0136] In this embodiment, the active power correlation coefficient can reflect the similarity of users' electricity consumption behavior characteristics before and after meter replacement, while the power fluctuation coefficient further verifies the rationality of users' electricity consumption behavior from the perspective of power consumption changes. Combining these two indicators, the calculated cross-user evaluation coefficient can comprehensively reflect the possibility of cross-users. By setting preset conditions to screen user pairs that meet the conditions, the misjudgment that may be caused by a single indicator is reduced. It makes full use of various information in power big data and combines information from different dimensions for cross-validation, thereby improving the accuracy of cross-user identification.

[0137] The user electricity meter cross-connection identification method based on power big data provided in this application embodiment can be executed by a user electricity meter cross-connection identification device based on power big data. This application embodiment uses the user electricity meter cross-connection identification device based on power big data executing the user electricity meter cross-connection identification method based on power big data as an example to illustrate the user electricity meter cross-connection identification device based on power big data provided in this application embodiment.

[0138] This application also provides a user electricity meter cross-connection identification device based on power big data.

[0139] like Figure 7 As shown, the user electricity meter cross-connection identification device based on big data of electricity includes: The determination module 710 is used to determine the set of users whose electricity meters are located in the same meter box based on the electricity user information in the power master station database. The acquisition module 720 is used to acquire the active power of each user at multiple different time points every day for n days before and n days after the meter replacement, as well as the daily electricity consumption of each user for n days before and n days after the meter replacement. The first calculation module 730 is used to calculate the active power correlation coefficient between each user and each other based on the active power of each user before the meter replacement and the active power of each user after the meter replacement. The second calculation module 740 is used to calculate the electricity consumption fluctuation coefficient of each user before and after the meter replacement based on the daily electricity consumption. The identification module 750 is used to identify the electricity meter pairs of the user set based on the active power correlation coefficient and the power fluctuation coefficient.

[0140] The power master station database is the core information platform used by power grid companies to store and manage various types of data in the power system. It encompasses massive amounts of data from all stages, including power generation, transmission, substation, distribution, and consumption. The data in the power master station database is typically stored in structured or semi-structured form and is maintained and queried through a professional database management system. The power master station database provides powerful data support for power grid operation monitoring, fault diagnosis, data analysis, and business decision-making. For example, it can record key information such as the real-time operating status of power equipment and user electricity consumption.

[0141] In the power master station database, power user information is a collection of basic data related to user electricity consumption, covering multiple dimensions such as user identity, electrical equipment, and electricity usage relationships. This includes user codes, user types (e.g., residential, commercial, industrial users), user addresses, meter codes, and meter box codes. Specifically, user codes distinguish different users; meter codes distinguish different meters; and meter box codes distinguish different meter boxes.

[0142] Active power is a physical quantity that measures the actual work capacity of a user's electrical equipment, reflecting the user's real electrical load at a specific point in time and corresponding to the actual operating state of the user's electrical equipment. For example, when a user turns on multiple electrical appliances in their home, the active power will increase accordingly; while when the user turns off some appliances, the active power will decrease. By monitoring active power, we can understand the user's electricity consumption behavior characteristics at different times. For example, when a user turns on lights, televisions, and other devices after waking up in the morning, the active power will show a significant increase; while at night when resting, the active power will decrease. In cross-meter detection scenarios, the time series data of active power can characterize the user's electricity consumption behavior characteristics (such as peak electricity consumption periods, equipment start-up and shutdown patterns, etc.). The differences or similarities in the user's electricity consumption behavior characteristics before and after meter replacement are important bases for determining whether the electricity meter is being used by multiple users.

[0143] Daily electricity consumption is the total electrical energy consumed by a user within a calendar day, reflecting the user's overall electricity usage throughout the day. For example, a user's electricity consumption may be relatively low on weekdays, while on weekends or holidays, due to spending more time at home, electricity consumption may increase significantly. By analyzing daily electricity consumption, we can understand a user's electricity usage habits. In cross-meter identification scenarios, daily electricity consumption not only reflects the user's overall electricity consumption scale, but also demonstrates the stability of electricity usage behavior through changes in values ​​before and after meter replacement. For example, under normal circumstances, a user's daily electricity consumption after meter replacement should maintain a relatively consistent fluctuation trend with that before the replacement. If a sudden change occurs, it may indicate an anomaly in the correspondence between the electricity meter and the user. Therefore, daily electricity consumption is an important data point for verifying the rationality of electricity usage behavior from a macro perspective.

[0144] In this embodiment, active power and daily electricity consumption can be obtained from the power station database. During data acquisition, the user's electricity meter replacement time is used as a reference point, tracing back n days to define the period before replacement and forward n days to define the period after replacement. The power station database typically collects active power at fixed intervals (usually 15 minutes, 30 minutes, or 1 hour) through the electricity meter's remote communication module. This data is stored in a historical database in an ordered manner according to timestamps. During retrieval, active power records from the n days before and n days after replacement can be filtered from the data stream corresponding to each user's electricity meter file within the user set.

[0145] Where n can be 7 days, 15 days, 30 days, etc. The value of n can be selected to cover the number of days of a complete weekly electricity consumption cycle, so as to fully capture the changes in the user's electricity consumption behavior before and after the meter replacement.

[0146] In this embodiment, the active power correlation coefficient is a statistical indicator that measures the degree of linear correlation between two time series. In a power system, active power is presented in the form of a time series, reflecting the electricity consumption behavior of users at different points in time. The value of the active power correlation coefficient can range from -1 to 1, where 1 represents a perfect positive correlation, -1 represents a perfect negative correlation, and 0 represents no linear correlation. For example, if the trends of active power changes for two users are highly consistent before and after meter replacement, then the active power correlation coefficient of these two time series will be close to 1; if the trends are completely opposite, the active power correlation coefficient will be close to -1; when the active power correlation coefficient is close to 0, it indicates that the two time series have almost no significant correlation. By calculating the active power correlation coefficient, the similarity or difference in the electricity consumption behavior of users before and after meter replacement can be quantified. In cross-user identification scenarios, under normal circumstances, the active power sequences of the same user before and after meter replacement should have a high positive correlation coefficient (because the main electricity user remains unchanged and the behavior pattern is stable), with the active power correlation coefficient close to 1. However, if cross-user identification occurs, the active power sequence of the original user after meter replacement may show a higher correlation with the sequences of other users before meter replacement. Therefore, cross-user identification can be located by the difference in active power correlation coefficient.

[0147] In this embodiment, it is necessary to calculate the active power correlation coefficient between each user and all other users. Here, "each user" refers to any user in the user set, while "all other users" includes the user itself and all other users. For each user in the user set, it is necessary to calculate not only its own active power correlation coefficient before and after meter replacement, but also its active power correlation coefficient with other users before and after meter replacement. For example, if the user set includes users A, B, C, D, E, and F, for user A, the active power correlation coefficient between user A and user A needs to be calculated based on user A's active power before and after meter replacement. Similarly, the active power correlation coefficient between user A and user B needs to be calculated based on user A's active power before and after meter replacement, and so on. The same applies to users B, C, D, E, and F.

[0148] For the same user, calculating the correlation coefficient between the active power sequence before and after the meter replacement directly verifies the consistency of electricity consumption behavior before and after the replacement. If the active power correlation coefficient is significantly low, it may indicate that the user's electricity meter may have been replaced by another user. For different users, calculating the active power correlation coefficients of cross-combinations such as user A's active power sequence before the meter replacement and user B's active power sequence after the meter replacement, and user A's sequence after the meter replacement and user B's active power sequence before the meter replacement, can further identify abnormal correlations. For example, if user A's active power sequence after the meter replacement is highly correlated with user C's active power sequence before the meter replacement, but has a low correlation with its own sequence before the meter replacement, there may be cross-metering between user A and user C. Through the logic of the above full-volume correlation analysis, potential cross-metering relationships can be investigated from multiple dimensions.

[0149] In this embodiment, each user has an active power time series containing multiple time points. For example, suppose user A's active power series before meter replacement is as follows: , Indicates the time point before the timetable was changed. The active power, the active power sequence after user B's meter replacement is as follows: , Indicates the time point after the table was changed. If the active power is known, the correlation coefficient between the active power of user A and user B can be calculated using the following formula. :

[0150] in, express The average value of the sequence. express The average value of the sequence.

[0151] The electricity consumption fluctuation coefficient is an indicator that measures the dispersion of a user's daily electricity consumption over a period of time. It reflects the stability of electricity consumption behavior. The smaller the electricity consumption fluctuation coefficient, the more balanced the daily electricity consumption and the more stable the electricity consumption pattern. The larger the electricity consumption fluctuation coefficient, the more significant the difference in daily electricity consumption and the greater the fluctuation in the electricity consumption pattern. In cross-user identification scenarios, under normal circumstances, the electricity consumption habits of the same user are continuous, and the electricity consumption fluctuation coefficient before and after the meter replacement should remain relatively consistent (e.g., both are at a low level). If cross-user identification occurs, due to the change of the electricity user, the daily electricity consumption after the meter replacement may show drastically different fluctuation characteristics compared to before the meter replacement. For example, the electricity consumption fluctuation coefficient may suddenly increase or deviate significantly from historical levels. Therefore, the electricity consumption coefficient can serve as an important basis for determining whether an electricity meter is being used by a different user.

[0152] By calculating the electricity consumption fluctuation coefficient, the rationality of users' electricity consumption behavior before and after meter replacement is verified from the perspective of the stability of macro-scale electricity consumption, providing another dimension of quantitative basis for cross-user identification.

[0153] In the preceding steps, the active power correlation coefficient between each user and other users, as well as the electricity consumption fluctuation coefficient before and after each user's meter replacement, have been calculated. The active power correlation coefficient and the electricity consumption fluctuation coefficient reflect the user's electricity consumption behavior characteristics from different perspectives. Under normal circumstances, if the correspondence between the electricity meter and the user remains unchanged, the active power sequence of the same user before and after the meter replacement should have a high active power correlation coefficient, typically close to 1, and the electricity consumption fluctuation coefficient before and after the replacement should be small, indicating consistent electricity consumption stability. However, when cross-user transactions occur, the actual user corresponding to the electricity meter changes, causing specific anomalies in both coefficients: the correlation between the active power sequence of the original user after the meter replacement and its active power sequence before the replacement decreases significantly, but it may show a high correlation with the sequence of another user before the replacement (because it actually corresponds to the electricity consumption behavior of another user); when cross-user transactions occur, the electricity consumption fluctuation coefficient before and after the replacement will also increase significantly, due to the abrupt change in electricity consumption stability caused by differences in the electricity consumption habits of different users. Based on this, cross-user pairs of electricity meters in a user set can be identified using the following methods. In some embodiments, thresholds can be set for the active power correlation coefficient and the power fluctuation coefficient, respectively. For example, the threshold for the active power correlation coefficient can be set to 0.7, where a value greater than or equal to 0.7 indicates similar electricity consumption behavior, and a value less than 0.7 indicates significant differences in electricity consumption; the threshold for the power fluctuation coefficient can be set to 0.3, where a value less than or equal to 0.3 indicates small changes in electricity consumption stability, and a value greater than 0.3 indicates a sudden change in electricity consumption stability.

[0154] For a user set within the same table box, we iterate through all possible user pairs. For example, if the user set includes user A, user B, user C, user D, user E, and user F, then we can iterate through pairs such as user i and user j. , , .

[0155] If the active power correlation coefficient between user i and user j is less than 0.7, and the active power correlation coefficient between user i and user j is greater than or equal to 0.7, it indicates that user i's electricity consumption behavior after the meter replacement is closer to that of user j. Furthermore, the electricity consumption fluctuation coefficient before and after the meter replacement for both user i and user j is greater than 0.3, suggesting that user i and user j may be a serial user pair, but further verification is needed.

[0156] Specifically, for the selected parallel user pairs, the judgment can be strengthened through reverse verification: check whether the correlation coefficient between the active power sequence of user j before the meter replacement and the active power sequence of user i after the meter replacement is also greater than or equal to 0.7, and whether the correlation coefficient between user j's active power before the meter replacement and its own active power after the meter replacement is less than 0.7. If the correlation coefficient between user j's active power sequence before the meter replacement and user i's active power sequence after the meter replacement is also greater than or equal to 0.7, and the correlation coefficient between user j's active power before the meter replacement and its own active power after the meter replacement is less than 0.7, then it can be finally determined that user i and user j are parallel user pairs.

[0157] According to the user electricity meter cross-connection identification device based on power big data in this application, the user set whose electricity meters are located in the same meter box is determined based on the power user information in the power master station database; the active power of each user in the user set at multiple different time points every day before and after the meter replacement n days are obtained, as well as the daily electricity consumption of each user before and after the meter replacement n days; the active power correlation coefficient between each user and each other is calculated based on the active power of each user before the meter replacement and the active power of each user after the meter replacement; the electricity consumption fluctuation coefficient of each user before and after the meter replacement is calculated based on the daily electricity consumption; and the cross-connection pairs of electricity meters in the user set are identified based on the active power correlation coefficient and the electricity consumption fluctuation coefficient. This application's embodiments can accurately identify electricity meters of users located within the same meter box by comprehensively analyzing power big data. By calculating the active power correlation coefficient, it can capture the similarity or difference in the user's electricity consumption behavior characteristics before and after meter replacement. Furthermore, by calculating the power fluctuation coefficient, it verifies the rationality of the user's electricity consumption behavior from the perspective of power change. Combining the active power correlation coefficient and the power fluctuation coefficient for comprehensive identification and cross-verification from different dimensions, it can accurately identify electricity meter cross-connections. Compared with traditional methods, it does not require a large amount of manpower for on-site inspection, reducing operation and maintenance costs and improving the efficiency of electricity meter cross-connection identification.

[0158] In some embodiments, the determining module 710 is further configured to: Retrieve power user information from the power master station database; Obtain the user's user code from the electricity user information; The meter box code of the user's electricity meter is retrieved from the electricity user information based on the user code. Based on the meter box code, the user code of the electricity meter located in the meter box is found from the electricity user information, thus obtaining the set of users in the same meter box.

[0159] In this embodiment, after obtaining the power user information from the power master station database, the user's user code can be extracted from this information. For example, the user code N of a certain power user A can be found and obtained from the power user information. yh-A Then, based on the user code N... yh-AFind and obtain the code N of meter box K to which power user A belongs from the power user information. bx-K Next, the set of all user codes {N} belonging to meter box K is retrieved from the electricity user information. yh-all-bx-K} refers to the set of users in the same table box.

[0160] like Figure 2 As shown, assume that 6 electricity meters are installed in meter box K, corresponding to the electricity meters of users A, B, C, D, E, and F. The user codes corresponding to these electricity meters are N. yh-A N yh-B N yh-C N yh-D N yh-E and N yh-F Then the user set {N} in box K yh-all-bx-K}={N yh-A N yh-B N yh-C N yh-D N yh-E N yh-F}

[0161] In this embodiment, by extracting power user information from the power master station database and relying on the association between user information and meter box codes in the power master station database, users whose electricity meters are located in the same meter box can be accurately located.

[0162] In some embodiments, the active power at multiple different time points each day is 96 points of active power.

[0163] In some embodiments, the first computing module 730 is further configured to: Calculate the active power correlation coefficient between each user and all other users based on the active power before and after the meter replacement, including: Calculate the average active power of each user in the same time period n days before meter replacement and the average active power of each user in the same time period n days after meter replacement based on the active power of 96 points. The correlation coefficient of active power between each user and each other is calculated based on the average active power of each user before the meter replacement and the average active power of each user after the meter replacement.

[0164] In this embodiment, by using 96 active power data points, i.e., active power values ​​recorded every 15 minutes throughout the day, the electricity consumption behavior characteristics of users at different time periods can be reflected in detail. By calculating the average active power of each user in the same time period n days before and n days after the meter replacement, short-term data fluctuations are further smoothed out. On the one hand, the data scale is reduced, and on the other hand, the high-frequency fluctuations of active power are reduced, thus reducing the interference of random factors on the analysis results. Based on the average active power before and after the meter replacement, the correlation coefficient of active power between users can be calculated, which can quantify the similarity or difference between users' electricity consumption behavior characteristics, thereby accurately identifying cross-connection of electricity meters.

[0165] In some embodiments, the first computing module 730 is further configured to: Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; The average active power of each user during the same time period in the n days prior to meter replacement is calculated based on the average active power of each user during the n days prior to meter replacement; and the average active power of each user during the same time period in the n days after meter replacement is calculated based on the average active power of each user during the n days after meter replacement.

[0166] In this embodiment, the hourly average value of 96 active power data points for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated. This preserves the user's electricity consumption characteristics at different time periods, effectively reduces the amount of data, lowers the computational complexity, and smooths out short-term fluctuations. Based on the hourly average active power, the average active power for the same time period for each user before and after the meter replacement (n days prior and n days after the replacement) is calculated, further refining the user's electricity consumption characteristics and more accurately reflecting the overall trend of the user's electricity consumption behavior before and after the meter replacement.

[0167] In some embodiments, the first computing module 730 is further configured to: According to the formula: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1 = (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 Calculate the average active power per hour for each user in the n days before meter replacement and the average active power per hour for each user in the n days after meter replacement; Among them, P A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced.A-q-n-96,1 P A-q-n-96,2 P A-q-n-96,3 P A-q-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n before user A's meter was replaced. A-h-n-24,1 P represents the average active power of user A in the first hour of the nth day after meter replacement. A-h-n-96,1 P A-h-n-96,2 P A-h-n-96,3 P A-h-n-96,4 These represent the 1st, 2nd, 3rd, and 4th power data points out of 96 data points on day n after user A replaced the meter.

[0168] In some embodiments, the first computing module 730 is further configured to: According to the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (P A-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n Calculate the average active power of each user in the same time period n days before meter replacement, and calculate the average active power of each user in the same time period n days after meter replacement. Among them, P A-q-24,1 P represents the average active power of user A in the first hour of the n days prior to meter replacement. A-q-n-24,1 P represents the average active power of user A in the first hour of the nth day before the meter was replaced. A-h-24,1 P represents the average active power of user A in the first hour, n days after the meter was replaced. A-h-n-24,1 This represents the average active power of user A in the first hour of the nth day after the meter was replaced.

[0169] In some embodiments, the first computing module 730 is further configured to: Calculate the average active power of each user in the i-th hour of the n-th day before the meter replacement, and calculate the average active power of each user in the i-th hour of the n-th day after the meter replacement; Calculate the active power correlation coefficient between each user and all other users based on the average active power of each user in the i-th hour of the n-th day before meter replacement and the average active power of each user in the i-th hour of the n-th day after meter replacement.

[0170] In some embodiments, the first computing module 730 is further configured to: According to the formula:

[0171]

[0172] Calculate the active power correlation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set. This represents the active power correlation coefficient between user a and user b. This represents the average active power array for user a before the meter was replaced. This represents the average active power of user a in the i-th hour of the n-th day prior to meter replacement. This represents the average value of the average power in the average active power array before user A replaced the meter. This represents the average active power array after user b replaced the meter. This represents the average active power of user b in the i-th hour, n days after the meter was replaced. This represents the average value of the average active power array after user b changed the table.

[0173] In some embodiments, the second computing module 740 is further configured to: Calculate the average daily electricity consumption for each user in the n days before the meter replacement, and calculate the average daily electricity consumption for each user in the n days after the meter replacement. The electricity consumption fluctuation coefficient for each user before and after the meter replacement is calculated based on the average daily electricity consumption for each user in the n days before the meter replacement and the average daily electricity consumption for each user in the n days after the meter replacement.

[0174] In this embodiment, the daily electricity consumption sequence for the user n days before the meter replacement and the daily electricity consumption sequence for the user n days after the meter replacement can be obtained. For example, for user a, the daily electricity consumption sequence for the user n days before the meter replacement is Q. a-q-1 Q a-q-2 Q a-q-n The daily electricity consumption sequence for n days after the meter replacement is Q. a-h-1 Q a-h-2 Q a-h-n , where Q a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h-n This represents user A's electricity consumption on day n after the meter replacement. We can calculate user A's average daily electricity consumption for the n days before and after the meter replacement, and then calculate the electricity consumption fluctuation coefficient before and after the meter replacement based on the changes in user A's average daily electricity consumption for the n days before and after the meter replacement. In some embodiments, the second computing module 740 is further configured to: According to the formula: Q a-q =( Qa-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h =( Q a-h-1 + Q a-h-2 +……+ Q a-h-n ) / n KQ a =min(Q a-q Q a-h ) / max(Q a-q Q a-h ) Calculate the electricity consumption fluctuation coefficient for each user before and after the meter replacement; Among them, Q a-q Q represents the average daily electricity consumption of user a over the n days prior to meter replacement. a-q-n Q represents the electricity consumption of user a on day n before the meter was replaced. a-h Q represents the average daily electricity consumption over n days after the meter replacement. a-h-n KQ represents the electricity consumption of user a on day n after the meter was replaced. a This represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement.

[0175] In some embodiments, the identification module 750 is further configured to: The cross-user evaluation coefficient between each user and other users is calculated based on the active power correlation coefficient and the power fluctuation coefficient. Two users whose cross-meter evaluation coefficient meets the preset conditions are identified as cross-meter pairs.

[0176] In this embodiment, the active power correlation coefficient can reflect the similarity of users' electricity consumption behavior characteristics before and after meter replacement, while the power fluctuation coefficient further verifies the rationality of users' electricity consumption behavior from the perspective of power consumption changes. Combining these two indicators, the calculated cross-user evaluation coefficient can comprehensively reflect the possibility of cross-users. By setting preset conditions to screen user pairs that meet the conditions, the misjudgment that may be caused by a single indicator is reduced. It makes full use of various information in power big data and combines information from different dimensions for cross-validation, thereby improving the accuracy of cross-user identification.

[0177] In some embodiments, the identification module 750 is further configured to: According to the formula: KE a-b =KC(P a-q-24 , P b-h-24 )×KQ b Calculate the cross-user evaluation coefficient between each user and all other users; Wherein, user a and user b are either the same user or different users in the user set, KE a-b KC(P) represents the cross-user rating coefficient between user a and user b. a-q-24 , P b-h-24 KQ represents the active power correlation coefficient between user a and user b. b This represents the fluctuation coefficient of user b's electricity consumption before and after the meter replacement.

[0178] For example, for user A, the cross-user evaluation coefficient between user A and users A~F is calculated as follows: KE A =[KE A-A , KE A-B , KE A-C , KE A-D , KE A-E , KE A-F ] =[ KC(P A-q-24 , P A-h-24 )×KQ A , KC(P A-q-24 , P B-h-24 )×KQ B ,… KC(P A-q-24 , P C-h-24 )×KQ C , KC(P A-q-24 , P D-h-24 )×KQ D ,… KC(P A-q-24 , P E-h-24 )×KQ E , KC(P A-q-24 , P F-h-24 )×KQ F ] Among them, KE A KE represents the array of user rating coefficients between user A and users A~F. A-A KE represents the cross-user rating coefficient between user A and user A. A-B KE represents the cross-user rating coefficient between user A and user B, ... A-F KC(P) represents the cross-user rating coefficient between user A and user F. A-q-24 , P A-h-24 KC(P) represents the active power correlation coefficient between user A and user A's. A-q-24 , P B-h-24 KC(P) represents the active power correlation coefficient between user A and user B, ..., KC(P) A-q-24 ,P F-h-24KQ represents the active power correlation coefficient between user A and user F. A KQ represents the fluctuation coefficient of user A's electricity consumption before and after the meter replacement. F This represents the fluctuation coefficient of user F's electricity consumption before and after the meter replacement.

[0179] Similarly, the cross-user evaluation coefficient array KE between users B~F and users A~F can be calculated separately. B KE C KE D KE E KE F Form the user evaluation coefficient matrix KE for users A~F before and after the form change. A~F .in: KE A~F ={KE A ; KE B ; KE C ; KE D ; KE E ; KE F} =[ KE A-A , KE A-B , KE A-C , KE A-D , KE A-E , KE A-F ; KE B-A , KE B-B , KE B-C , KE B-D , KE B-E , KE B-F ; KE C-A , KE C-B , KE C-C , KE C-D , KE C-E , KE C-F ; KE D-A , KE D-B , KE D-C , KE D-D , KE D-E , KE D-F ; KE E-A , KE E-B , KE E-C , KE E-D , KE E-E , KE E-F ; KEF-A , KE F-B , KE F-C , KE F-D , KE F-E , KE F-F ] Among them, KE B-A KE represents the cross-user rating coefficient between user B and user A, ... B-F KE represents the cross-user evaluation coefficient between user B and user F. C-A KE represents the cross-user rating coefficient between user C and user A, ... C-F KE represents the cross-user evaluation coefficient between user C and user F. D-A KE represents the cross-user rating coefficient between user D and user A. D-F KE represents the cross-user evaluation coefficient between user D and user F. E-A , represents the cross-user evaluation coefficient between user E and user A, ..., KE E-F KE represents the cross-user evaluation coefficient between user E and user F. F-A KE represents the cross-user evaluation coefficient between user F and user A, ... F-F This represents the cross-user evaluation coefficient between user F and user F.

[0180] In some embodiments, the preset conditions are:

[0181] Where user i and user j are different users in the user set. This represents the cross-user evaluation coefficient between user i and user i'. This represents the cross-user rating coefficient between user j and user j. This represents the cross-user evaluation coefficient between user i and user j. This represents the user rating coefficient between user j and user i.

[0182] In this embodiment, by constructing the above-mentioned four-fold inequality constraints as preset conditions, theoretically, a user's own electricity consumption behavior characteristics should be highly consistent. Therefore, the cross-user evaluation coefficient between the same user should be close to 1. Due to various interference factors that may exist in the actual electricity consumption process, such as changes in temporary electrical equipment, the correlation coefficient may not reach 1 completely, but it should still remain at a high level, that is, greater than 0.6. When the cross-user evaluation coefficient between the same user is less than 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are significantly different. Under normal circumstances, the electricity consumption behavior characteristics of the same user before and after the meter replacement should be relatively similar. Therefore, it can be preliminarily confirmed that the metered user may have changed. Furthermore, if the cross-user evaluation coefficient between two different users is large, that is, greater than or equal to 0.6, it indicates that the user's behavior characteristics before and after the meter replacement are highly consistent. Under normal circumstances, the user behavior characteristics between different users should be significantly different. By using the above-mentioned four-fold inequality constraints for joint judgment, normal users and users who may have cross-user problems can be accurately distinguished.

[0183] The user electricity meter serial identification device based on power big data in this application embodiment can be an electronic device or a component of an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other devices besides a terminal. For example, the electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. It can also be a server, network attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc. This application embodiment does not specifically limit the specific device.

[0184] The user electricity meter serial identification device based on power big data in this application embodiment can be a device with an operating system. This operating system can be Microsoft (Windows), Android, iOS, or other possible operating systems; this application embodiment does not specifically limit it.

[0185] In some embodiments, such as Figure 8 As shown, this application embodiment also provides an electronic device 800, including a processor 801, a memory 802, and a computer program stored in the memory 802 and executable on the processor 801. When the program is executed by the processor 801, it implements the various processes of the above-described user electricity meter serial identification method embodiment based on power big data and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0186] It should be noted that the electronic devices in the embodiments of this application include the aforementioned mobile electronic devices and non-mobile electronic devices.

[0187] This application also provides a non-transitory computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described user electricity meter serial identification method embodiment based on power big data and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0188] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0189] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for identifying user electricity meters based on big data of electricity.

[0190] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0191] This application also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described user electricity meter serial identification method based on power big data, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0192] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0193] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0194] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0195] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

[0196] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0197] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A user electric energy meter string household identification method based on electric power big data, characterized in that, The method comprises the following steps: determining a user set in which the electric energy meters are in the same meter box based on the electric power user information in the electric power master station database; obtaining the active power of each user in the user set at different time points in each day for n days before and after the replacement of the electric energy meter, and the daily power consumption of each user for n days before and after the replacement of the electric energy meter; calculating the active power correlation coefficient between each user and each user based on the active power of each user before the replacement of the electric energy meter and the active power of each user after the replacement of the electric energy meter; calculating the power consumption fluctuation coefficient of each user before and after the replacement of the electric energy meter based on the daily power consumption; identifying the electric energy meter stringing household pair of the user set based on the active power correlation coefficient and the power consumption fluctuation coefficient.

2. The method of claim 1, wherein, The method comprises the following steps: obtaining the electric power user information from the electric power master station database; obtaining the user code of the user from the electric power user information; finding the meter box code of the meter box in which the electric energy meter of the user is located from the electric power user information based on the user code; finding the user code of the user in which the electric energy meter is located in the meter box from the electric power user information based on the meter box code, and obtaining the user set in which the electric energy meters are in the same meter box.

3. The method of claim 1, wherein, The active power at different time points in each day is 96-point active power.

4. The method of claim 3, wherein, The method comprises the following steps: calculating the average active power of each user in the same time period for n days before the replacement of the electric energy meter and the average active power of each user in the same time period for n days after the replacement of the electric energy meter based on the 96-point active power; calculating the active power correlation coefficient between each user and each user based on the average active power of each user before the replacement of the electric energy meter and the average active power of each user after the replacement of the electric energy meter.

5. The method of claim 4, wherein, The method comprises the following steps: calculating the average value of the active power of each user for each hour for n days before the replacement of the electric energy meter and the average value of the active power of each user for each hour for n days after the replacement of the electric energy meter; calculating the average active power of each user in the same time period for n days before the replacement of the electric energy meter based on the average value of the active power of each user for each hour for n days before the replacement of the electric energy meter, and calculating the average active power of each user in the same time period for n days after the replacement of the electric energy meter based on the average value of the active power of each user for each hour for n days after the replacement of the electric energy meter.

6. The method of claim 4, wherein, The average value of the active power of each user for each hour for n days before the replacement of the electric energy meter and the average value of the active power of each user for each hour for n days after the replacement of the electric energy meter are calculated according to the formula: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1 = (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 The average active power of each user in the same time period for n days before the replacement of the electric energy meter and the average active power of each user in the same time period for n days after the replacement of the electric energy meter are calculated according to the formula: where P A-q-n-24,1 represents the average of the active power in the first hour of the nth day before the meter replacement of user A, P A-q-n-96,1 , P A-q-n-96,2 , P A-q-n-96,3 , P A-q-n-96,4 respectively represent the first, second, third and fourth power data in the 96-point power data of the nth day before the meter replacement of user A. P A-h-n-24,1 represents the average of the active power in the first hour of the nth day after the meter replacement of user A, P A-h-n-96,1 , P A-h-n-96,2 , P A-h-n-96,3 , P A-h-n-96,4 respectively represent the first, second, third and fourth power data in the 96-point power data of the nth day after the meter replacement of user A.

7. The method of claim 4, wherein, The average active power of each user in the same time period for n days before the replacement of the electric energy meter and the average active power of each user in the same time period for n days after the replacement of the electric energy meter are calculated according to the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (P A-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n The method comprises the following steps: where P A-q-24,1 represents the average active power of the first hour of the nth day before the meter change of user A, P A-q-n-24,1 represents the average value of the active power of the first hour of the nth day before the meter change of user A, P A-h-24,1 represents the average active power of the first hour of the nth day after the meter change of user A, P A-h-n-24,1 represents the average value of the active power of the first hour of the nth day after the meter change of user A.

8. The method of claim 1, wherein, calculating the average active power of each user in the i-th hour for n days before the replacement of the electric energy meter, and calculating the average active power of each user in the i-th hour for n days after the replacement of the electric energy meter; ​ The active power correlation coefficient between each user and each user is calculated according to the average active power of each user in the ith hour of n days before the meter replacement and the average active power of each user in the ith hour of n days after the meter replacement.

9. The method of claim 8, wherein, According to the formula: The active power correlation coefficient between each user and each user is calculated; Wherein, the user a and the user b are same users or different users in the user set, denotes the active power correlation coefficient between the user a and the user b, denotes the average active power array of the user a before changing the meter, denotes the average active power of the i-th hour of the n-th day before the user a changes the meter, denotes the average value of the average power in the average active power array of the user a before changing the meter, denotes the average active power of the i-th hour of the n-th day after the user a changes the meter, denotes the average active power array of the user b after changing the meter, denotes the average active power of the i-th hour of the n-th day after the user b changes the meter, denotes the average value of the average power in the average active power array of the user b after changing the meter.

10. The method of claim 1, wherein, The power fluctuation coefficient of each user before and after the meter replacement is calculated according to the daily power consumption of each user. The daily average power consumption of each user in n days before the meter replacement and the daily average power consumption of each user in n days after the meter replacement are calculated according to the daily power consumption of each user. The power fluctuation coefficient of each user before and after the meter replacement is calculated according to the daily average power consumption of each user in n days before the meter replacement and the daily average power consumption of each user in n days after the meter replacement.

11. The method of claim 10, wherein, According to the formula: Q a-q = ( Q a-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h = ( Q a-h-1 + Q a-h-2 +……+ Q a-h-n ) / n KQ a =min(Q a-q , Q a-h ) / max(Q a-q , Q a-h ) The power fluctuation coefficient of each user before and after the meter replacement is calculated; Wherein, Q a-q represents the daily average electricity consumption of user a n days before the meter replacement, Q a-q-n represents the electricity consumption of user a on the nth day before the meter replacement, Q a-h represents the daily average electricity consumption n days after the meter replacement, Q a-h-n represents the electricity consumption of user a on the nth day after the meter replacement, KQ a represents the electricity consumption fluctuation coefficient of user a before and after the meter replacement.

12. The method of claim 1, wherein, The electric energy meter string household pair of the user set is identified according to the active power correlation coefficient and the power fluctuation coefficient. The string household evaluation coefficient between each user and each user is calculated according to the active power correlation coefficient and the power fluctuation coefficient. Two users whose string household evaluation coefficient meets a preset condition are determined as an electric energy meter string household pair.

13. The method of claim 12, wherein, According to the formula: KE a-b = KC(P a-q-24 , P b-h-24 ) x KQ b The string household evaluation coefficient between each user and each user is calculated; wherein the user a and the user b are same users or different users in the user set, KE a-b represents a cross-user evaluation coefficient between the user a and the user b, KC(P a-q-24 , P b-h-24 ) represents an active power correlation coefficient between the user a and the user b, KQ b represents a power fluctuation coefficient of the user b before and after the meter replacement.

14. The method of claim 13, wherein, The preset condition is: wherein user i and user j are different users in the set of users, denotes the cross-user evaluation coefficient between user i and user i, denotes the cross-user evaluation coefficient between user j and user j, denotes the cross-user evaluation coefficient between user i and user j, denotes the cross-user evaluation coefficient between user j and user i.

15. A user electric energy meter string household identification device based on electric power big data, characterized in that, It includes: A determination module is configured to determine a user set in which electric energy meters are in the same meter box based on power user information in a power master station database; An acquisition module is configured to acquire active power of each user in multiple different time points of each day in n days before and after the meter replacement of each user in the user set, and daily power consumption of each user in n days before and after the meter replacement; A first calculation module is configured to calculate an active power correlation coefficient between each user and each user according to active power of each user before the meter replacement and active power of each user after the meter replacement; A second calculation module is configured to calculate a power fluctuation coefficient of each user before and after the meter replacement according to the daily power consumption; An identification module is configured to identify an electric energy meter string household pair of the user set according to the active power correlation coefficient and the power fluctuation coefficient.

16. The apparatus of claim 15, wherein, The determination module is further configured to: Acquire power user information from the power master station database; Acquire user codes of users from the power user information; Find meter box codes of meter boxes in which electric energy meters of the users are located from the power user information according to the user codes; Find user codes of users in which electric energy meters are located from the power user information according to the meter box codes, to obtain a user set in which electric energy meters are in the same meter box.

17. The apparatus of claim 15, wherein, The active power of the multiple different time points of each day is 96-point active power.

18. The apparatus of claim 17, wherein, The first calculation module is further configured to: Calculate average active power of each user in the same time period of n days before the meter replacement and average active power of each user in the same time period of n days after the meter replacement according to the 96-point active power; Calculate an active power correlation coefficient between each user and each user according to the average active power of each user before the meter replacement and the average active power of each user after the meter replacement.

19. The apparatus of claim 18, wherein, The first calculation module is further configured to: calculate the average active power of each user in the same time period before the meter replacement of each user according to the average active power of each user in each hour before the meter replacement of each user for n days; calculate the average active power of each user in the same time period before the meter replacement of each user according to the average active power of each user in each hour before the meter replacement of each user for n days; and calculate the average active power of each user in the same time period after the meter replacement of each user according to the average active power of each user in each hour after the meter replacement of each user for n days.

20. The apparatus of claim 18, wherein, The first calculation module is further configured to: according to the formula: P A-q-n-24,1 = (P A-q-n-96,1 +P A-q-n-96,2 +P A-q-n-96,3 +P A-q-n-96,4 ) / 4 P A-h-n-24,1 = (P A-h-1-96,1 +P A-h-n-96,2 +P A-h-n-96,3 +P A-h-n-96,4 ) / 4 calculate the average active power of each user in each hour before the meter replacement of each user for n days and the average active power of each user in each hour after the meter replacement of each user for n days; wherein P A-q-n-24,1 represents the average of the active power in the first hour of the nth day before the meter change of user A, P A-q-n-96,1 , P A-q-n-96,2 , P A-q-n-96,3 , P A-q-n-96,4 respectively represent the first, second, third, and fourth power data in the 96-point power data of the nth day before the meter change of user A, P A-h-n-24,1 represents the average of the active power in the first hour of the nth day after the meter change of user A, P A-h-n-96,1 , P A-h-n-96,2 , P A-h-n-96,3 , P A-h-n-96,4 respectively represent the first, second, third, and fourth power data in the 96-point power data of the nth day after the meter change of user A.

21. The apparatus of claim 18, wherein, The first calculation module is further configured to: according to the formula: P A-q-24,1 = (P A-q-1-24,1 +P A-q-2-24,1 +P A-q-3-24,1 +……+P A-q-n-24,1 ) / n P A-h-24,1 = (P A-h-1-24,1 +P A-h-2-24,1 +P A-h-3-24,1 +……+P A-h-n-24,1 ) / n calculate the average active power of each user in the same time period before the meter replacement of each user, and calculate the average active power of each user in the same time period after the meter replacement of each user; where P A-q-24,1 represents the average active power of the first hour of the nth day before the meter change of user A, P A-q-n-24,1 represents the average value of the active power of the first hour of the nth day before the meter change of user A, P A-h-24,1 represents the average active power of the first hour of the nth day after the meter change of user A, P A-h-n-24,1 represents the average value of the active power of the first hour of the nth day after the meter change of user A.

22. The apparatus of claim 18, wherein, The first calculation module is further configured to: calculate the average active power of each user in the i-th hour before the meter replacement of each user, and calculate the average active power of each user in the i-th hour after the meter replacement of each user; calculate the active power correlation coefficient between each user and each user according to the average active power of each user in the i-th hour before the meter replacement of each user and the average active power of each user in the i-th hour after the meter replacement of each user.

23. The apparatus of claim 18, wherein, The first calculation module is further configured to: according to the formula: calculate the active power correlation coefficient between each user and each user; Wherein, the user a and the user b are same users or different users in the user set, denotes the active power correlation coefficient between the user a and the user b, denotes the average active power array of the user a before changing the meter, denotes the average active power of the i-th hour of the n-th day before the user a changes the meter, denotes the average value of the average power in the average active power array of the user a before changing the meter, denotes the average active power of the i-th hour of the n-th day after the user a changes the meter, denotes the average active power array of the user b after changing the meter, denotes the average active power of the i-th hour of the n-th day after the user b changes the meter, denotes the average value of the average power in the average active power array of the user b after changing the meter.

24. The apparatus of claim 18, wherein, The second calculation module is further configured to: calculate the daily average power consumption of each user before the meter replacement of each user for n days, and calculate the daily average power consumption of each user after the meter replacement of each user for n days according to the daily power consumption; calculate the power consumption fluctuation coefficient of each user before and after the meter replacement according to the daily average power consumption of each user before the meter replacement for n days and the daily average power consumption of each user after the meter replacement for n days.

25. The apparatus of claim 24, wherein, The second calculation module is further configured to: according to the formula: Q a-q = ( Q a-q-1 + Q a-q-2 +……+ Q a-q-n ) / n Q a-h = ( Q a-h-1 + Q a-h-2 +……+ Q a-h-n ) / n KQ a =min(Q a-q , Q a-h ) / max(Q a-q , Q a-h ) calculate the power consumption fluctuation coefficient of each user before and after the meter replacement; wherein Q a-q represents the daily average power consumption of user a n days before the meter replacement, Q a-q-n represents the power consumption of user a on the nth day before the meter replacement, Q a-h represents the daily average power consumption n days after the meter replacement, Q a-h-n represents the power consumption of user a on the nth day after the meter replacement, KQ a represents the power consumption fluctuation coefficient of user a before and after the meter replacement.

26. The apparatus of claim 15, wherein, The identification module is further configured to: calculate the cross-user evaluation coefficient between each user and each user according to the active power correlation coefficient and the power consumption fluctuation coefficient; determine two users whose cross-user evaluation coefficient meets a preset condition as a meter cross-user pair.

27. The apparatus of claim 26, wherein, The identification module is further configured to: according to the formula: KE a-b = KC(P a-q-24 , P b-h-24 ) x KQ b calculate the cross-user evaluation coefficient between each user and each user; wherein the user a and the user b are same users or different users in the user set, KE a-b represents a cross-user evaluation coefficient between the user a and the user b, KC(P a-q-24 , P b-h-24 ) represents an active power correlation coefficient between the user a and the user b, KQ b represents a power fluctuation coefficient of the user b before and after the meter replacement.

28. The apparatus of claim 27, wherein, The preset condition is: wherein user i and user j are different users in the set of users, denotes the cross-user evaluation coefficient between user i and user i, denotes the cross-user evaluation coefficient between user j and user j, denotes the cross-user evaluation coefficient between user i and user j, denotes the cross-user evaluation coefficient between user j and user i.

29. An electronic device, comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the user electric energy meter cross-user identification method based on power big data according to any one of claims 1-14 when executing the computer program. 30.A non-transitory computer-readable storage medium having stored thereon a computer program. The computer program is executed by the processor to implement the user electric energy meter cross-user identification method based on power big data according to any one of claims 1-14.

Images