Electricity utilization abnormity analysis system based on electricity utilization acquisition
Through the electricity usage anomaly analysis system, users are grouped using Pearson correlation coefficient and cyclic iteration method, and abnormal electricity usage is identified in combination with the standard deviation method, which solves the problem of low efficiency of existing anti-powered power prevention methods and realizes efficient power theft detection and management , maintain fair order in the power market.
Patent Information
- Application Number
- CN202410893940.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-04
- Publication Date
- 2024-11-26
AI Technical Summary
The existing anti-powered power theft is time-consuming and labor-intensive, low efficiency, and it is difficult to effectively investigate and deal with user power theft.
The electricity consumption abnormality analysis system based on electricity consumption acquisition includes a data acquisition unit, a data processing unit, a packet analysis unit, anomaly detection unit and a result processing unit. The user is grouped through Pearson's correlation coefficient, and the grouping is optimized using a cyclic iterative method, combining standard deviations. The method identifies abnormal electricity use behavior and performs on-site verification and results processing.
Effectively narrow the scope of power stolen detection, improve the efficiency of power stolen detection, realize comprehensive management of power consumption abnormalities, and maintain fair order in the power market.
Smart Images

Figure CN119026042A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power consumption anomaly analysis, and in particular to a power consumption anomaly analysis system based on power consumption collection. Background Art
[0002] With the development of economy and the increase of population, the demand for energy consumption is growing. This leads to a larger load on the power system and an increase in the possibility of abnormal power consumption data. Therefore, the importance of real-time monitoring and analysis of power consumption data is becoming increasingly prominent.
[0003] At present, in order to prevent users from stealing electricity, the main measures adopted are to strengthen line loss management, improve the electricity inspection system, and increase the number of meter readings. Due to the large number of electricity users in the area, manual comparison and summary are used to analyze users' electricity usage, which is time-consuming, labor-intensive, difficult to investigate and deal with, and inefficient. Summary of the invention
[0004] The purpose of the present invention is to provide an abnormal power consumption analysis system based on power consumption collection, which solves the problem that the existing anti-power theft means in the background technology are time-consuming and laborious.
[0005] To achieve the above object, the present invention provides the following technical solution: a power consumption anomaly analysis system based on power consumption collection, comprising:
[0006] Data collection unit, data processing unit, grouping analysis unit, anomaly detection unit, field verification unit and result processing unit;
[0007] The grouping analysis unit is used to calculate the correlation coefficient between the power consumption of each residential user and the line loss rate of the corresponding area, and the grouping analysis unit includes a pre-grouping module and an optimization grouping module;
[0008] Pre-grouping module: used to divide users into positive correlation group, negative correlation group, low power group and weak correlation group according to their power consumption data when power consumption abnormality occurs, and the grouping analysis module sets a positive correlation threshold, a negative correlation threshold and a low power threshold;
[0009] The pre-grouping module uses the Pearson correlation coefficient to calculate the correlation coefficient between the user's power consumption and the line loss rate of the corresponding area. The calculation formula is:
[0010]
[0011] Among them, x i It is a sample of the user's daily electricity consumption. is the average value of the user's daily electricity consumption sample; i is a sample of daily power loss in the area. It is the average value of daily power loss samples in the area;
[0012] According to the correlation coefficient and the threshold set by the grouping analysis module, users are pre-divided into corresponding groups. When the correlation coefficient r is greater than the positive correlation threshold, it is divided into a positive correlation group. When the correlation coefficient r is less than the negative correlation threshold, it is divided into a negative correlation group. When the correlation coefficient r is between the positive correlation threshold and the negative correlation threshold, it is divided into a weak correlation group. When the user's power consumption is less than 50% of the average power consumption in the area, the user is considered to belong to the low power group.
[0013] Optimization grouping module: used to accurately classify users into corresponding positive correlation groups, negative correlation groups, low power groups or weak correlation groups, and use a cyclic iteration method to optimize the correlation between each group and the line loss curve of the area by calling in and out users;
[0014] The steps of the loop iteration method are:
[0015] S1: Transfer-in users: For users who are not in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer-in and the line loss curve of the area, and determine whether to keep the user in the current group based on the increase or decrease in the correlation after the transfer-out;
[0016] S2: Transfer out users: For users in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer out and the line loss curve of the area, and determine whether to keep the user in the current group according to the increase or decrease in the correlation after the transfer out;
[0017] S3: Iteration termination: Set an upper limit on the number of iterations and a correlation improvement threshold. When any of these conditions is met, the iteration is stopped.
[0018] Furthermore, the data collection unit collects the electricity usage data of all users in the area for the past year as the benchmark data when no electricity theft incident occurs.
[0019] Furthermore, the data acquisition unit is also used to export the required data from the metering automation system: regional user list, regional power data and user power data, and ensure that the data format is unified to facilitate subsequent processing. The data acquisition unit is also used to calculate the regional line loss based on the regional power data and user power data.
[0020] Furthermore, the data processing unit is responsible for data cleaning, preprocessing and normalization.
[0021] Furthermore, the step of transferring the positive correlation group is:
[0022] S1. Select users: Select users to be transferred from users who are not in the positive correlation group;
[0023] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new positive correlation group and the regional line loss curve after the user is transferred in;
[0024] S3. Determine and group: if the correlation coefficient increases after the user is transferred in, it is considered that the transfer of the user has increased the correlation of the positive correlation group, and the user is added to the positive correlation group;
[0025] The steps of calling out the positive correlation group are:
[0026] S1. Select user: Select the user to be transferred from the users in the current positive correlation group;
[0027] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining positively correlated group users and the regional line loss curve after the user is transferred out;
[0028] S3. Judge and group: If the correlation coefficient decreases after the user is called out, it is considered that the user's calling out reduces the correlation of the positive correlation group, and the user should be retained in the positive correlation group; if the correlation coefficient increases after the user is called out, it means that the user does not match other users in the positive correlation group and should be removed from the positive correlation group.
[0029] Furthermore, the anti-correlation group calling step is:
[0030] S1. Select user: select the user to be transferred from the users who are not in the anti-correlation group;
[0031] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new anti-correlation group and the regional line loss curve after the user is transferred in;
[0032] S3. Determine and group: If the correlation coefficient decreases after the user is transferred in (ie, closer to -1), it is considered that the user's transfer increases the correlation of the anti-correlation group, and the user is added to the anti-correlation group;
[0033] The anti-correlation group calling steps are:
[0034] S1. Select user: Select the user to be transferred from the users in the current anti-correlation group;
[0035] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the power load curve of the remaining anti-correlation group users and the regional line loss curve after the user is transferred out;
[0036] S3. Determine and group: If the correlation coefficient increases after the user is called out, it is considered that the user's calling out reduces the correlation of the anti-correlation group, and the user should be retained in the anti-correlation group; if the correlation coefficient decreases after the user is called out, it means that the user does not match the users in other anti-correlation groups and should be removed from the anti-correlation group.
[0037] Furthermore, the weakly correlated group importing step is:
[0038] S1. Select users: Select users to be transferred from users who are not in the weakly related group;
[0039] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the overall power load curve of the new weakly correlated group and the line loss curve of the substation area after the user is transferred in;
[0040] S3. Determine and group: If the correlation coefficient does not change much after the user is transferred in, and is still between the positive correlation threshold and the negative correlation threshold, it is considered that the transfer of the user does not change the characteristics of the weak correlation group, and the user is added to the weak correlation group;
[0041] The steps of calling out the weakly correlated group are:
[0042] S1. Select user: Select the user to be transferred from the users in the current weakly related group;
[0043] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining weakly correlated group users and the line loss curve of the substation area after the user is transferred out;
[0044] S3. Judgment and grouping: If the correlation coefficient does not change much after tuning out and is still between the positive correlation threshold and the negative correlation threshold, keep the user in the weak correlation group; if the correlation coefficient is close to the positive correlation threshold or the negative correlation threshold after tuning out, it means that the user may be more suitable for other groups and should be removed and reconsidered.
[0045] Furthermore, the steps of transferring in and out the low power group are:
[0046] S1. Select users: Select users whose electricity consumption is less than 50% of the average electricity consumption in the area from users who have never been in the low-power group;
[0047] S2. Direct grouping: users who meet the above conditions are directly added to the low-power group;
[0048] S3. Monitoring power consumption: Regularly monitor the power consumption of users in the low power group;
[0049] S4. Regrouping: If a user's electricity consumption exceeds 50% of the average electricity consumption in the region for several consecutive months, the user will be removed from the low-power group and its grouping will be reconsidered.
[0050] Furthermore, the abnormality detection unit identifies abnormal power consumption behavior by a standard deviation method, and the formula of the standard deviation method is:
[0051]
[0052] Where S is the standard deviation, N is the number of hours of data points, and x i is the power consumption of the ith data point, is the average power consumption of all data points, which is calculated as:
[0053]
[0054] By setting a threshold, when S is greater than the threshold, it means that the user is suspected of stealing electricity.
[0055] Furthermore, the on-site verification unit is used to conduct surprise inspections on the suspected electricity theft users, observe their electricity equipment and electricity usage, and find evidence of electricity theft. The result processing unit handles the users confirmed to have committed electricity theft in accordance with relevant laws and regulations, including fines, power outages, and recovery of electricity charges. At the same time, it strengthens the monitoring and management of their subsequent electricity usage to prevent the recurrence of electricity theft.
[0056] Compared with the prior art, the present invention has the following beneficial effects:
[0057] The power consumption anomaly analysis system based on power consumption collection provided by the present invention is composed of multiple units. Through data collection, processing, grouping analysis, anomaly detection, on-site verification and result processing links, the identification scope is narrowed by grouping to achieve comprehensive management of power consumption anomalies. The data collection unit collects user power data and calculates regional line loss. The data processing unit ensures data quality. The grouping analysis unit groups users based on the Pearson correlation coefficient and optimizes the grouping through iteration. The anomaly detection unit identifies abnormal power consumption. The on-site verification unit conducts surprise inspections. The result processing unit handles power theft. The entire system effectively combats power theft, maintains fair order in the power market, narrows the scope of power theft detection, and effectively improves the efficiency of power theft detection. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 It is a schematic diagram of the modules of the present invention. DETAILED DESCRIPTION
[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0060] In order to solve the technical problems of the existing anti-electricity theft methods, such as Figure 1 As shown, the following preferred technical solutions are provided:
[0061] The power consumption anomaly analysis system based on power consumption collection includes:
[0062] Data collection unit, data processing unit, grouping analysis unit, anomaly detection unit, field verification unit and result processing unit;
[0063] The grouping analysis unit is used to calculate the correlation coefficient between the power consumption of each residential user and the line loss rate of the corresponding area, and the grouping analysis unit includes a pre-grouping module and an optimization grouping module;
[0064] Pre-grouping module: used to divide users into positive correlation group, negative correlation group, low power group and weak correlation group according to their power consumption data when abnormal power consumption occurs, and the grouping analysis module sets a positive correlation threshold, a negative correlation threshold and a low power threshold; the probability of abnormal power consumption in the four groups is from large to small: negative correlation group, low power group, weak correlation group and positive correlation group;
[0065] The pre-grouping module uses the Pearson correlation coefficient to calculate the correlation coefficient between the user's power consumption and the line loss rate of the corresponding area. The calculation formula is:
[0066]
[0067] Among them, x i It is a sample of the user's daily electricity consumption. is the average value of the user's daily electricity consumption sample; i is a sample of daily power loss in the area. It is the average value of daily power loss samples in the area;
[0068] According to the correlation coefficient and the threshold set by the grouping analysis module, users are pre-divided into corresponding groups. When the correlation coefficient r is greater than the positive correlation threshold, it is divided into a positive correlation group. When the correlation coefficient r is less than the negative correlation threshold, it is divided into a negative correlation group. When the correlation coefficient r is between the positive correlation threshold and the negative correlation threshold, it is divided into a weak correlation group. When the user's power consumption is less than 50% of the average power consumption in the area, the user is considered to belong to the low power group.
[0069] Optimization grouping module: used to accurately classify users into corresponding positive correlation groups, negative correlation groups, low power groups or weak correlation groups, and use a cyclic iteration method to optimize the correlation between each group and the line loss curve of the area by calling in and out users;
[0070] The steps of the loop iteration method are:
[0071] S1: Transfer-in users: For users who are not in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer-in and the line loss curve of the area, and determine whether to keep the user in the current group based on the increase or decrease in the correlation after the transfer-out;
[0072] S2: Transfer out users: For users in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer out and the line loss curve of the area, and determine whether to keep the user in the current group according to the increase or decrease in the correlation after the transfer out;
[0073] S3: Iteration termination: Set an upper limit on the number of iterations and a correlation improvement threshold. When any of these conditions is met, the iteration is stopped.
[0074] The data collection unit collects the electricity usage data of all users in the area for the past year as the baseline data when no electricity theft occurs.
[0075] The data acquisition unit is also used to export the required data from the metering automation system: regional user list, regional power data and user power data, and ensure that the data format is unified for subsequent processing. The data acquisition unit is also used to calculate the regional line loss based on the regional power data and user power data.
[0076] The data processing unit is responsible for data cleaning, preprocessing and normalization.
[0077] The steps for adding the positive correlation group are:
[0078] S1. Select users: Select users to be transferred from users who are not in the positive correlation group;
[0079] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new positive correlation group and the regional line loss curve after the user is transferred in;
[0080] S3. Determine and group: if the correlation coefficient increases after the user is transferred in, it is considered that the transfer of the user has increased the correlation of the positive correlation group, and the user is added to the positive correlation group;
[0081] The steps to call out the positive correlation group are:
[0082] S1. Select user: Select the user to be transferred from the users in the current positive correlation group;
[0083] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining positively correlated group users and the regional line loss curve after the user is transferred out;
[0084] S3. Judge and group: If the correlation coefficient decreases after the user is called out, it is considered that the user's calling out reduces the correlation of the positive correlation group, and the user should be retained in the positive correlation group; if the correlation coefficient increases after the user is called out, it means that the user does not match other users in the positive correlation group and should be removed from the positive correlation group.
[0085] The steps for adding the anti-correlation group are:
[0086] S1. Select user: select the user to be transferred from the users who are not in the anti-correlation group;
[0087] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new anti-correlation group and the regional line loss curve after the user is transferred in;
[0088] S3. Determine and group: If the correlation coefficient decreases after the user is transferred in (ie, closer to -1), it is considered that the user's transfer increases the correlation of the anti-correlation group, and the user is added to the anti-correlation group;
[0089] The steps to call out the anti-correlation group are:
[0090] S1. Select user: Select the user to be transferred from the users in the current anti-correlation group;
[0091] S2. Calculate the correlation: calculate the Pearson correlation coefficient between the power load curve of the remaining anti-correlation group users and the regional line loss curve after the user is transferred out;
[0092] S3. Determine and group: If the correlation coefficient increases after the user is called out, it is considered that the user's calling out reduces the correlation of the anti-correlation group, and the user should be retained in the anti-correlation group; if the correlation coefficient decreases after the user is called out, it means that the user does not match the users in other anti-correlation groups and should be removed from the anti-correlation group.
[0093] The steps for transferring weakly correlated groups are:
[0094] S1. Select users: Select users to be transferred from users who are not in the weakly related group;
[0095] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the overall power load curve of the new weakly correlated group and the line loss curve of the substation area after the user is transferred in;
[0096] S3. Determine and group: If the correlation coefficient does not change much after the user is transferred in, and is still between the positive correlation threshold and the negative correlation threshold, it is considered that the transfer of the user does not change the characteristics of the weak correlation group, and the user is added to the weak correlation group;
[0097] The steps to call out the weakly correlated group are:
[0098] S1. Select user: Select the user to be transferred from the users in the current weakly related group;
[0099] S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining weakly correlated group users and the line loss curve of the substation area after the user is transferred out;
[0100] S3. Judgment and grouping: If the correlation coefficient does not change much after tuning out and is still between the positive correlation threshold and the negative correlation threshold, keep the user in the weak correlation group; if the correlation coefficient is close to the positive correlation threshold or the negative correlation threshold after tuning out, it means that the user may be more suitable for other groups and should be removed and reconsidered.
[0101] The steps to transfer in and out the low battery group are as follows:
[0102] S1. Select users: Select users whose electricity consumption is less than 50% of the average electricity consumption in the area from users who have never been in the low-power group;
[0103] S2. Direct grouping: users who meet the above conditions are directly added to the low-power group;
[0104] S3. Monitoring power consumption: Regularly monitor the power consumption of users in the low power group;
[0105] S4. Regrouping: If a user's electricity consumption exceeds 50% of the average electricity consumption in the region for several consecutive months, the user will be removed from the low-power group and its grouping will be reconsidered.
[0106] The abnormal detection unit identifies abnormal power consumption behavior through the standard deviation method, and the formula of the standard deviation method is:
[0107]
[0108] Where S is the standard deviation, N is the number of hours of data points, and x i is the power consumption of the ith data point, is the average power consumption of all data points, which is calculated as:
[0109]
[0110] By setting a threshold, when S is greater than the threshold, it means that the user is suspected of stealing electricity.
[0111] The on-site verification unit is used to conduct surprise inspections on users suspected of electricity theft, observe their electrical equipment and electricity usage, and look for evidence of electricity theft. The result processing unit handles users confirmed to have committed electricity theft in accordance with relevant laws and regulations, including fines, power outages, and recovery of electricity bills. At the same time, it strengthens the monitoring and management of their subsequent electricity usage to prevent electricity theft from happening again.
[0112] Specifically, the system consists of a data acquisition unit, a data processing unit, a group analysis unit, an anomaly detection unit, a field verification unit and a result processing unit. The data acquisition unit is responsible for collecting the electricity usage data of all users in the area for the past year as benchmark data, exporting the regional user list, regional electricity data and user electricity data from the metering automation system, and ensuring that the data format is unified. Based on these data, the system can calculate the regional line loss. The data processing unit cleans, preprocesses and normalizes the collected data to ensure data quality and provide a reliable basis for subsequent analysis. The Pearson correlation coefficient is used to calculate the correlation coefficient between the user's power consumption and the line loss rate of the corresponding area. According to the set positive correlation threshold, negative correlation threshold and low power threshold, the users are initially divided into positive correlation group, negative correlation group, low power group and weak correlation group. The cyclic iteration method is used to optimize the correlation between each group and the regional line loss curve by transferring in and out users. The anomaly detection unit uses the standard deviation method to identify abnormal power consumption behavior. The on-site verification unit conducts surprise inspections on the screened suspected power theft users, observes their power equipment and power consumption, and searches for evidence of power theft. The result processing unit processes the users who are confirmed to have power theft. Through multiple links such as data collection, processing, group analysis, anomaly detection, on-site verification and result processing, a comprehensive analysis and management of power consumption anomalies is achieved, power theft is effectively identified and combated, and the fairness and order of the power market are maintained.
[0113] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0114] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.
Claims
1. The power consumption anomaly analysis system based on power consumption collection is characterized by: include: Data collection unit, data processing unit, grouping analysis unit, anomaly detection unit, field verification unit and result processing unit; The grouping analysis unit is used to calculate the correlation coefficient between the power consumption of each residential user and the line loss rate of the corresponding area, and the grouping analysis unit includes a pre-grouping module and an optimization grouping module; Pre-grouping module: used to divide users into positive correlation group, negative correlation group, low power group and weak correlation group according to their power consumption data when power consumption abnormality occurs, and the grouping analysis module sets a positive correlation threshold, a negative correlation threshold and a low power threshold; The pre-grouping module uses the Pearson correlation coefficient to calculate the correlation coefficient between the user's power consumption and the line loss rate of the corresponding area. The calculation formula is: Among them, x i It is a sample of the user's daily electricity consumption. is the average value of the user's daily electricity consumption sample; i is a sample of daily power loss in the area. It is the average value of daily power loss samples in the area; According to the correlation coefficient and the threshold set by the grouping analysis module, users are pre-divided into corresponding groups. When the correlation coefficient r is greater than the positive correlation threshold, it is divided into a positive correlation group. When the correlation coefficient r is less than the negative correlation threshold, it is divided into a negative correlation group. When the correlation coefficient r is between the positive correlation threshold and the negative correlation threshold, it is divided into a weak correlation group. When the user's power consumption is less than 50% of the average power consumption in the area, the user is considered to belong to the low power group. Optimization grouping module: used to accurately classify users into corresponding positive correlation groups, negative correlation groups, low power groups or weak correlation groups, and use a cyclic iteration method to optimize the correlation between each group and the line loss curve of the area by calling in and out users; The steps of the loop iteration method are: S1: Transfer-in users: For users who are not in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer-in and the line loss curve of the area, and determine whether to keep the user in the current group based on the increase or decrease in the correlation after the transfer-out; S2: Transfer out users: For users in the current group, calculate the Pearson correlation coefficient between the overall power load curve of the group after the transfer out and the line loss curve of the area, and determine whether to keep the user in the current group according to the increase or decrease in the correlation after the transfer out; S3: Iteration termination: Set an upper limit on the number of iterations and a correlation improvement threshold. When any of these conditions is met, the iteration is stopped.
2. The power consumption anomaly analysis system based on power consumption collection according to claim 1, characterized in that: The data collection unit collects the electricity usage data of all users in the area for the past year as the benchmark data when no electricity theft incident occurs.
3. The power consumption anomaly analysis system based on power consumption collection according to claim 1 is characterized in that: The data acquisition unit is also used to export the required data from the metering automation system: regional user list, regional power data and user power data, and ensure that the data format is unified for subsequent processing. The data acquisition unit is also used to calculate the regional line loss based on the regional power data and user power data.
4. The power consumption anomaly analysis system based on power consumption collection as claimed in claim 3 is characterized in that: The data processing unit is responsible for data cleaning, preprocessing and normalization.
5. The power consumption anomaly analysis system based on power consumption collection according to claim 1, characterized in that: The steps of transferring the positive correlation group are as follows: S1. Select users: Select users to be transferred from users who are not in the positive correlation group; S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new positive correlation group and the regional line loss curve after the user is transferred in; S3. Determine and group: if the correlation coefficient increases after the user is transferred in, it is considered that the transfer of the user has increased the correlation of the positive correlation group, and the user is added to the positive correlation group; The steps of calling out the positive correlation group are: S1. Select user: Select the user to be transferred from the users in the current positive correlation group; S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining positively correlated group users and the regional line loss curve after the user is transferred out; S3. Judge and group: If the correlation coefficient decreases after the user is called out, it is considered that the user's calling out reduces the correlation of the positive correlation group, and the user should be retained in the positive correlation group; if the correlation coefficient increases after the user is called out, it means that the user does not match other users in the positive correlation group and should be removed from the positive correlation group.
6. The power consumption anomaly analysis system based on power consumption collection according to claim 5, characterized in that: The steps of calling the anti-correlation group are as follows: S1. Select user: select the user to be transferred from the users who are not in the anti-correlation group; S2. Calculate the correlation: calculate the Pearson correlation coefficient between the overall power load curve of the new anti-correlation group and the regional line loss curve after the user is transferred in; S3. Determine and group: If the correlation coefficient decreases after the user is transferred in (ie, closer to -1), it is considered that the user's transfer increases the correlation of the anti-correlation group, and the user is added to the anti-correlation group; The anti-correlation group calling step is: S1. Select user: Select the user to be transferred from the users in the current anti-correlation group; S2. Calculate the correlation: calculate the Pearson correlation coefficient between the power load curve of the remaining anti-correlation group users and the regional line loss curve after the user is transferred out; S3. Determine and group: If the correlation coefficient increases after the user is called out, it is considered that the user's calling out reduces the correlation of the anti-correlation group, and the user should be retained in the anti-correlation group; if the correlation coefficient decreases after the user is called out, it means that the user does not match the users in other anti-correlation groups and should be removed from the anti-correlation group.
7. The power consumption anomaly analysis system based on power consumption collection according to claim 6, characterized in that: The steps of transferring the weakly correlated group are as follows: S1. Select users: Select users to be transferred from users who are not in the weakly related group; S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the overall power load curve of the new weakly correlated group and the line loss curve of the substation area after the user is transferred in; S3. Determine and group: If the correlation coefficient does not change much after the user is transferred in, and is still between the positive correlation threshold and the negative correlation threshold, it is considered that the transfer of the user does not change the characteristics of the weak correlation group, and the user is added to the weak correlation group; The steps of calling out the weakly correlated group are: S1. Select user: Select the user to be transferred from the users in the current weakly related group; S2. Calculate the correlation: Calculate the Pearson correlation coefficient between the power load curve of the remaining weakly correlated group users and the line loss curve of the substation area after the user is transferred out; S3. Judgment and grouping: If the correlation coefficient does not change much after tuning out and is still between the positive correlation threshold and the negative correlation threshold, keep the user in the weak correlation group; if the correlation coefficient is close to the positive correlation threshold or the negative correlation threshold after tuning out, it means that the user may be more suitable for other groups and should be removed and reconsidered.
8. The power consumption anomaly analysis system based on power consumption collection according to claim 7 is characterized in that: The steps of transferring in and out the low power group are as follows: S1. Select users: Select users whose electricity consumption is less than 50% of the average electricity consumption in the area from users who have never been in the low-power group; S2. Direct grouping: users who meet the above conditions are directly added to the low-power group; S3. Monitoring power consumption: Regularly monitor the power consumption of users in the low power group; S4. Regrouping: If a user's electricity consumption exceeds 50% of the average electricity consumption in the region for several consecutive months, the user will be removed from the low-power group and its grouping will be reconsidered.
9. The power consumption anomaly analysis system based on power consumption collection according to claim 8, characterized in that: The abnormality detection unit identifies abnormal power consumption behavior by using a standard deviation method, and the formula of the standard deviation method is: Where S is the standard deviation, N is the number of hours of data points, and x i is the power consumption of the ith data point, is the average power consumption of all data points, which is calculated as: By setting a threshold, when S is greater than the threshold, it means that the user is suspected of stealing electricity.
10. The power consumption anomaly analysis system based on power consumption collection according to claim 9, characterized in that: The on-site verification unit is used to conduct surprise inspections on the screened suspected electricity theft users, observe their electrical equipment and electricity usage, and find evidence of electricity theft. The result processing unit handles the users confirmed to have committed electricity theft in accordance with relevant laws and regulations, including fines, power outages, and recovery of electricity charges. At the same time, it strengthens the monitoring and management of their subsequent electricity usage to prevent electricity theft from happening again.
