A data anomaly detection method, system, device and medium

Abstract

The application discloses a kind of data anomaly detection method, system, equipment and medium, by selecting the user data corresponding to each user in user service demand data from preset database.User service demand data is used to identify reference type, determine the reference type of judgment.If the reference type of judgment is first judgment reference, meter rotation data anomaly identification is carried out using user data, to generate first data anomaly detection result.If the reference type of judgment is second judgment reference, industry expansion suspension work order data anomaly identification is carried out using user data, to generate second data anomaly detection result.If the reference type of judgment is third judgment reference, basic file data anomaly identification is carried out using user data, to generate third data anomaly detection result.Based on the reference type of judgment, corresponding data anomaly identification is carried out using user data, quickly find out the error caused by manual operation process, improve the working accuracy of business, the working accuracy of business personnel.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to methods, systems, devices and media for detecting data anomalies. Background Technology

[0002] With the continuous development of my country's economy and the increasing number of electricity customers, changes in user records are generated during the processes of adding or modifying user information. In the marketing system, the data within each function menu of the marketing module is relatively independent and has low correlation. The marketing system requires manual verification to track and validate changes at every stage. Therefore, local power supply stations currently need to invest significant human resources in verifying business change data.

[0003] The marketing system only provides simple, passive queries for daily, individual, and single-household data statistics. Business personnel cannot intuitively and comprehensively judge the accuracy of business data or identify potential risks. Therefore, existing data anomaly detection methods, which rely on manual verification of user data, are prone to electricity billing errors, hindering comprehensive data analysis by business personnel and resulting in low accuracy of the detection results. Summary of the Invention

[0004] This invention provides a data anomaly detection method, system, device, and medium, which solves the technical problem that existing data anomaly detection methods, which rely on manual verification to detect anomalies in user data, are prone to electricity billing errors, are not conducive to business personnel's comprehensive data analysis, and result in low accuracy of the detection results.

[0005] The present invention provides a data anomaly detection method, comprising:

[0006] When user business requirement data is received, user data corresponding to each user in the user business requirement data is selected from the preset database;

[0007] The reference type is identified and determined using the user business requirement data.

[0008] If the determination reference type is the first determination reference, then the user data is used to identify abnormal electricity meter rotation data and generate a first data abnormality detection result.

[0009] If the determination reference type is the second determination reference, then the user data is used to identify abnormal data in the business expansion suspension work order, and a second data abnormality detection result is generated.

[0010] If the determination reference type is a third determination reference, then the user data is used to identify basic profile data anomalies and generate a third data anomaly detection result.

[0011] Optionally, the user data includes change data in the operation log of the electricity metering device; the step of using the user data to identify anomalies in meter rotation data and generating a first data anomaly detection result includes:

[0012] Determine whether the newly installed continuous data in the operation log change data of the power metering device is at a preset threshold;

[0013] If not, determine whether the change data in the operation log of the power metering device has a preset current type;

[0014] If so, the change data in the operation ledger of the power metering device will be screened for anomalies to generate a first data anomaly detection result;

[0015] If not, the abnormal result corresponding to the preset current type shall be taken as the first data abnormality detection result;

[0016] If so, the abnormal result corresponding to the preset threshold shall be taken as the first data anomaly detection result.

[0017] Optionally, the change data in the operation ledger of the electricity metering device includes forward and reverse active power meter code data, average voltage value, voltage value, and asset number; the step of performing anomaly screening on the change data in the operation ledger of the electricity metering device to generate a first data anomaly detection result includes:

[0018] According to the preset user screening rules, the users corresponding to the changes in the operation ledger of the power metering device are screened to generate multiple abnormal screening users;

[0019] Determine whether the positive and negative active power meter code data corresponding to the abnormal screening user are in the preset meter code state;

[0020] If so, the abnormal result of the table code corresponding to the preset table code status shall be used as the abnormal detection result of the user corresponding to the abnormal screening user.

[0021] If not, then determine whether the average voltage value corresponding to the abnormal screening user is less than the first preset voltage threshold.

[0022] If so, the voltage anomaly result corresponding to the first preset voltage threshold shall be taken as the user anomaly detection result corresponding to the anomaly screening user.

[0023] If not, determine whether the voltage value corresponding to the abnormal screening user is the second preset voltage threshold.

[0024] If so, the voltage anomaly result corresponding to the second preset voltage threshold shall be taken as the user anomaly detection result corresponding to the anomaly screening user;

[0025] If not, the asset number corresponding to the abnormal screening user is matched with the preset marketing system table replacement list to generate the user abnormality detection result corresponding to the abnormal screening user;

[0026] Using all the aforementioned user anomaly detection results, a first data anomaly detection result is constructed.

[0027] Optionally, the user data includes business expansion suspension work order data; the step of using the user data to identify anomalies in business expansion suspension work order data and generating a second data anomaly detection result includes:

[0028] The business expansion pause work order data is filtered according to the first preset filtering conditions to generate pause work order data and user pause count;

[0029] The remaining number of suspension days is calculated by comparing the expiration date corresponding to the number of user suspensions with the current date;

[0030] If the remaining number of suspension days is less than or equal to the preset number of days, the detection result corresponding to the preset number of days will be used as the initial data anomaly detection result.

[0031] The business expansion pause work order data is filtered for pause recovery data according to the second preset filtering conditions to generate associated work order numbers;

[0032] The associated work order number is matched with the pause work order data to generate the power supply time;

[0033] The initial data anomaly detection result is updated by comparing the power supply time with the corresponding expiration date, and the target data anomaly detection result and the number of days suspended are generated.

[0034] Determine whether the number of days that have been suspended meets the preset suspension threshold;

[0035] If so, the pause notification corresponding to the preset pause threshold will be taken as the second data anomaly detection result;

[0036] If not, the target data anomaly detection result shall be taken as the second data anomaly detection result.

[0037] Optionally, the step of updating the initial data anomaly detection result by using the comparison result between the power supply time and the corresponding due date, and generating the target data anomaly detection result and the number of days suspended, includes:

[0038] Determine whether the power supply time is less than or equal to the corresponding expiration date;

[0039] If so, determine whether the shutdown year corresponding to the shutdown time of the power supply time is consistent with the power supply year corresponding to the power supply time;

[0040] If so, calculate the difference between the power supply time and the shutdown time to generate the first number of days that have been suspended;

[0041] The first number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result;

[0042] If not, calculate the difference between the expiration date and the preset power supply time to generate a second number of days that have been suspended;

[0043] The second number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result;

[0044] If not, the initial data anomaly detection result is updated based on the power supply time, the shutdown time, and the expiration date to generate the target data anomaly detection result and the number of days suspended.

[0045] Optionally, the step of updating the initial data anomaly detection result based on the power supply time, the shutdown time, and the expiration date, and generating the target data anomaly detection result and the number of days suspended, includes:

[0046] Determine whether the year of the shutdown time corresponding to the power supply time is consistent with the year of power supply corresponding to the power supply time;

[0047] If so, calculate the difference between the expiration date and the corresponding suspension time to generate the third number of suspended days;

[0048] The initial data anomaly detection result is updated using the suspension notification corresponding to the third number of suspended days, a target data anomaly detection result is generated, and the third number of suspended days is used as the number of suspended days;

[0049] If not, calculate the difference between the expiration date and the corresponding preset power supply time to generate a fourth number of days that have been suspended;

[0050] The initial data anomaly detection result is updated using the suspension notification corresponding to the fourth number of suspended days, a target data anomaly detection result is generated, and the fourth number of suspended days is used as the number of suspended days.

[0051] Optionally, the user data includes market attributes and initial business work orders; the step of using the user data to perform basic profile data anomaly identification and generate a third data anomaly detection result includes:

[0052] The initial business work order is supplemented and updated according to the market attributes to generate the target business work order;

[0053] The user's electricity bill in the target business work order is matched and verified with the preset electricity bill, and abnormal electricity bill data is generated.

[0054] The user list in the target business work order is compared with the pre-designed fee policy to generate electricity bill processing data;

[0055] The demand meter data is generated by comparing the demand meter data in the target business work order with the preset meter data.

[0056] The abnormal electricity bill data, the electricity bill processing data, and the demand meter data are used to construct a third data anomaly detection result.

[0057] The present invention also provides a data anomaly detection system, comprising:

[0058] The user data generation module is used to acquire user business requirement data, select user data from a preset database according to the user business requirement data, and generate user data corresponding to the user business requirement.

[0059] The reference type determination module is used to identify the reference type using the user business requirement data and determine the reference type.

[0060] The first data anomaly detection result generation module is used to generate a first data anomaly detection result by using the user data to identify anomalies in the electricity meter rotation data if the determination reference type is the first determination reference.

[0061] The second data anomaly detection result generation module is used to identify data anomalies in business expansion suspension work orders by using the user data if the determination reference type is the second determination reference, and generate the second data anomaly detection result.

[0062] The third data anomaly detection result generation module is used to generate a third data anomaly detection result by using the user data to identify basic file data anomalies if the determination reference type is a third determination reference.

[0063] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of implementing any of the above-described data anomaly detection methods.

[0064] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements any of the above-described data anomaly detection methods.

[0065] As can be seen from the above technical solutions, the present invention has the following advantages:

[0066] This invention, upon receiving user business request data, selects user data corresponding to each user from a preset database. It then uses this user business request data for reference type identification to determine the judgment reference type. If the judgment reference type is the first judgment reference, it uses user data to identify anomalies in meter rotation data, generating a first data anomaly detection result. If the judgment reference type is the second judgment reference, it uses user data to identify anomalies in business expansion suspension work order data, generating a second data anomaly detection result. If the judgment reference type is the third judgment reference, it uses user data to identify anomalies in basic file data, generating a third data anomaly detection result. This solves the technical problem of existing data anomaly detection methods that rely on manual verification of user data, easily leading to electricity billing errors, hindering comprehensive data analysis by business personnel, and resulting in low accuracy of detection results. By analyzing user business request data, determining the judgment reference type, and using user data to identify corresponding data anomalies based on the judgment reference type, it quickly identifies errors caused by manual operation processes, improves the accuracy of business operations and the work efficiency of business personnel, promotes business process optimization, and effectively enhances customer service levels. Attached Figure Description

[0067] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] Figure 1 This is a flowchart of the steps of a data anomaly detection method provided in Embodiment 1 of the present invention;

[0069] Figure 2 This is a flowchart of the steps of a data anomaly detection method provided in Embodiment 2 of the present invention;

[0070] Figure 3 This is a structural block diagram of a data anomaly detection system provided in Embodiment 3 of the present invention. Detailed Implementation

[0071] This invention provides a data anomaly detection method, system, device, and medium to address the technical problem that existing data anomaly detection methods, which rely on manual verification to detect anomalies in user data, are prone to errors in electricity billing, hinder business personnel from comprehensively analyzing data, and result in low accuracy of the detection results.

[0072] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0073] Please see Figure 1 , Figure 1 The flowchart illustrates the steps of a data anomaly detection method provided in Embodiment 1 of the present invention.

[0074] Example 1 of this invention provides a data anomaly detection method, comprising:

[0075] Step 101: When user business requirement data is received, select the user data corresponding to each user from the preset database.

[0076] The pre-defined databases include a marketing system database and a metering system database. The marketing system database retrieves the latest data for each user daily, including changes to the operation logs of each user's electricity metering devices, business expansion suspension work orders, basic data such as electricity prices and market attributes, and related billing data. The metering system database retrieves meter reading data corresponding to various data types from the marketing system database. The data from both the marketing system and metering system databases serves as the data foundation for calculating business requirements.

[0077] In this embodiment of the invention, when user service demand data is received, the user corresponding to the user service demand data is determined, which is usually multiple users with the same reference type. User data corresponding to each user in the user service demand data is selected from a preset database, that is, the power metering device operation ledger change data, business expansion suspension work order data, basic data such as electricity price and market attributes, related billing data, and corresponding meter reading data are obtained for each user.

[0078] Step 102: Use user business requirement data to identify the reference type and determine the reference type.

[0079] In this embodiment of the invention, based on user business demand data, the required data is selected for analysis to determine whether the business demand data corresponds to meter rotation data, transformer downtime, or basic file data. If the business demand is meter rotation data, the reference type is determined to be the first reference. If the business demand is transformer downtime, the reference type is determined to be the second reference. If the business demand is basic file data, the reference type is determined to be the third reference.

[0080] Step 103: If the reference type is determined to be the first reference, then the user data is used to identify the abnormality of the meter rotation data and generate the first data abnormality detection result.

[0081] In this embodiment of the invention, when the judgment reference type corresponding to the user business demand data is the first judgment reference, the electricity metering device operation ledger change data in the user data is used to identify anomalies in the meter rotation data. It is determined whether the newly installed continuous data in the electricity metering device operation ledger change data is within a preset threshold. If not, it is determined whether the electricity metering device operation ledger change data contains a preset current type. If yes, the electricity metering device operation ledger change data is used for anomaly screening to generate a first data anomaly detection result. If not, the anomaly result corresponding to the preset current type is used as the first data anomaly detection result. If yes, the anomaly result corresponding to the preset threshold is used as the first data anomaly detection result.

[0082] Step 104: If the reference type is determined to be the second reference, then the user data is used to identify data anomalies in the business expansion suspension work order, and a second data anomaly detection result is generated.

[0083] In this embodiment of the invention, when the judgment reference type corresponding to the user business demand data is the second judgment reference, the business expansion suspension work order data in the user data is used to identify business expansion suspension work order data anomalies. First, the business expansion suspension work order data is filtered for suspension data according to the first preset filtering conditions to generate suspension work order data and user suspension counts. Then, the difference between the due date corresponding to the user suspension count and the current date is calculated to generate the remaining suspension days. If the remaining suspension days are less than or equal to the preset number of days, the detection result corresponding to the preset number of days is used as the initial data anomaly detection result. Next, the business expansion suspension work order data is filtered for suspension recovery data according to the second preset filtering conditions to generate associated work order numbers. By matching the associated work order numbers with the suspension work order data, a power supply time is generated. The initial data anomaly detection result is updated using the comparison result between the power supply time and the corresponding due date to generate the target data anomaly detection result and the number of suspension days. Finally, it is determined whether the number of suspension days meets the preset suspension threshold. If so, the suspension notification corresponding to the preset suspension threshold is used as the second data anomaly detection result. If not, the target data anomaly detection result will be used as the second data anomaly detection result.

[0084] Step 105: If the reference type is determined to be the third reference, then the user data is used to identify basic file data anomalies and generate the third data anomaly detection result.

[0085] In this embodiment of the invention, when the judgment reference type corresponding to the user business demand data is the third judgment reference, the market attributes in the user data and the initial business work order are used to identify anomalies in the basic file data. The initial business work order is supplemented and updated according to the market attributes to generate a target business work order. The user electricity bill list in the target business work order is matched and verified with the preset electricity bill list to generate abnormal electricity bill data. The user list in the target business work order is compared with the pre-designed fee policy to generate electricity bill processing data. The demand meter data in the target business work order is compared with the preset meter data to generate demand meter data. Using the abnormal electricity bill data, the electricity bill processing data, and the demand meter data, a third data anomaly detection result is constructed.

[0086] In this embodiment of the invention, when user business demand data is received, user data corresponding to each user is selected from a preset database. The user business demand data is used for reference type identification to determine the judgment reference type. If the judgment reference type is a first judgment reference, the user data is used to identify anomalies in meter rotation data, generating a first data anomaly detection result. If the judgment reference type is a second judgment reference, the user data is used to identify anomalies in business expansion suspension work order data, generating a second data anomaly detection result. If the judgment reference type is a third judgment reference, the user data is used to identify anomalies in basic file data, generating a third data anomaly detection result. This solves the technical problem that existing data anomaly detection methods, which rely on manual verification of user data, are prone to electricity billing errors, hindering comprehensive data analysis by business personnel and resulting in low accuracy of detection results. By analyzing user business demand data, determining the judgment reference type, and using user data to identify corresponding data anomalies based on the judgment reference type, errors caused by manual operation processes can be quickly identified, improving the accuracy of business operations and the work efficiency of business personnel, promoting business process optimization, and effectively enhancing customer service levels.

[0087] Please see Figure 2 , Figure 2 This is a flowchart of a data anomaly detection method provided in Embodiment 2 of the present invention.

[0088] Another data anomaly detection method provided in Example 2 of this invention includes:

[0089] Step 201: When user business requirement data is received, select the user data corresponding to each user from the preset database.

[0090] In this embodiment of the invention, the specific implementation process of step 201 is similar to that of step 101, and will not be repeated here.

[0091] Step 202: Use user business requirement data to identify the reference type and determine the reference type.

[0092] In this embodiment of the invention, the specific implementation process of step 202 is similar to that of step 102, and will not be repeated here.

[0093] Step 203: If the reference type is determined to be the first reference, then the user data is used to identify the abnormality of the meter rotation data and generate the first data abnormality detection result.

[0094] Furthermore, the user data includes changes to the operation log of the electricity metering device, and step 203 may include the following sub-steps S11-S15:

[0095] S11. Determine whether the newly installed continuous data in the power metering device operation log change data is at the preset threshold. If not, proceed to S12; if yes, proceed to S15.

[0096] S12. Determine whether there is a preset current type in the change data of the operation log of the power metering device. If yes, proceed to S13; otherwise, proceed to S14.

[0097] S13. Perform anomaly screening on the change data of the operation ledger of the power metering device and generate the first data anomaly detection result.

[0098] S14. Take the abnormal result corresponding to the preset current type as the first data abnormality detection result.

[0099] S15. The abnormal results corresponding to the preset threshold are taken as the first data anomaly detection results.

[0100] The data changes in the electricity metering device operation log include daily retrieval of the metering return database based on the meter replacement list from the marketing system until the next month's closing (around the 15th), ensuring that all meter replacement users have at least 10 days of data. Meter data is obtained based on the installation / removal date t+5, the occurrence of reverse current on one day after installation, and the abnormal average voltage value on one day after installation (voltage data is taken as a daily average; according to the rule that current and voltage concentrators for low-voltage users collect data every 3 hours, there are 8 data points per day; the average is taken, the sum of the 8 data points is divided by 8; if data is empty at a certain point in time, the voltage value at the point with available data is taken and divided by the number of points in time).

[0101] New installation continuous data refers to the current and voltage data of the user meter collected for 5 consecutive days after the new installation date.

[0102] The preset threshold refers to a situation where newly installed continuous data is "empty". The abnormal result corresponding to the preset threshold is "no data, collection abnormal".

[0103] The preset current type refers to the occurrence of reverse readings on one day after installation, or the occurrence of reverse current at eight different time points. The corresponding abnormal result for the preset current type is "No data abnormality".

[0104] In this embodiment of the invention, it is determined whether the newly installed continuous data in the power metering device operation ledger change data meets a preset threshold. If the user's meter readings for current and voltage are "empty" for five consecutive days after the new installation date, it is determined to be a data acquisition anomaly, and "No data, acquisition anomaly" is output. Otherwise, the result is returned in numerical format, and the next step is to determine whether the power metering device operation ledger change data contains a preset current type. If a reverse reading occurs on one day after the new installation or if reverse current occurs at eight different times, the power metering device operation ledger change data is used for anomaly screening to generate a first data anomaly detection result.

[0105] Furthermore, the data change in the operation log of the electricity metering device includes forward and reverse active power meter code data, average voltage value, voltage value, and asset number. Step S13 may include the following sub-steps S131-S139:

[0106] S131. According to the preset user screening rules, the users corresponding to the changes in the operation ledger of the power metering device are screened to generate multiple abnormal screening users.

[0107] S132. Determine whether the positive and negative active power meter code data corresponding to the abnormal screening user are in the preset meter code state. If yes, execute S133; otherwise, execute S134.

[0108] S133. Use the abnormal result of the table code corresponding to the preset table code status as the abnormal detection result of the user corresponding to the abnormal screening user.

[0109] S134. Determine whether the average voltage value corresponding to the abnormal screening user is less than the first preset voltage threshold. If yes, execute S135; otherwise, execute S136.

[0110] S135. The voltage anomaly result corresponding to the first preset voltage threshold is used as the user anomaly detection result corresponding to the anomaly screening user.

[0111] S136. Determine whether the voltage value corresponding to the abnormal screening user is the second preset voltage threshold. If yes, execute S137; otherwise, execute S138.

[0112] S137. The voltage anomaly result corresponding to the second preset voltage threshold is used as the user anomaly detection result corresponding to the anomaly screening user.

[0113] S138. Match the asset number corresponding to the abnormal screening user with the preset marketing system form replacement list to generate the user abnormality detection result corresponding to the abnormal screening user.

[0114] S139. Using all user anomaly detection results, construct the first data anomaly detection result.

[0115] The preset user filtering rules refer to excluding users who have normal reverse current, such as those in the photovoltaic, elevator, and metal industries. In other words, all users in the photovoltaic three-household model are excluded. The whitelist is formed by combining the marketing profile username and address with "elevator", the name with "property management", and the marketing profile username or industry category with keywords such as metal or hardware.

[0116] The preset meter reading states include: forward active power meter reading is 0, reverse active power meter reading is not 0; negative current occurs at 8 or more time points; forward active power meter reading is not 0, reverse active power meter reading is not 0, and the reverse active power meter reading is greater than or equal to "set lower limit coefficient" * forward active power meter reading and less than "set upper limit coefficient" * forward active power meter reading; forward active power meter reading is not 0, reverse active power meter reading is not 0, and the reverse active power meter reading is greater than "set upper limit coefficient" * forward active power meter reading. The corresponding abnormal meter reading results for the preset meter reading states are "suspected wiring error", "suspected one phase reversed", and "suspected two phases reversed".

[0117] The first preset voltage threshold refers to the average voltage value of any phase on one day after installation, which meets the requirement of "voltage value < 180". The voltage anomaly result corresponding to the first preset voltage threshold is "voltage anomaly of a certain phase".

[0118] The second preset voltage threshold refers to a point in time where any phase experiences a voltage value of 0. The voltage anomaly result corresponding to the second preset voltage threshold is "voltage loss in a certain phase".

[0119] The pre-set marketing system replacement list refers to a list that includes the asset number of each user before the replacement.

[0120] In this embodiment of the invention, users with inherently normal reverse current, such as those in the photovoltaic, elevator, and metal industries, are removed from the electricity metering device operation log change data, generating multiple anomaly screening users. The next step involves determining whether the forward and reverse active power meter data corresponding to each anomaly screening user are in a preset meter code state. When the forward and reverse active power meter data corresponding to an anomaly screening user are in a preset meter code state, the following situations apply:

[0121] If the forward active power meter reading is 0 and the reverse active power meter reading is not 0, output "Suspected wiring error". If negative current occurs at 8 or more time points, output "Suspected wiring error". If both the forward and reverse active power meter readings are not 0, and the reverse active power meter reading is greater than or equal to "set lower limit coefficient" * forward active power meter reading and less than "set upper limit coefficient" * forward active power meter reading, output "Suspected one phase reversed". If both the forward and reverse active power meter readings are not 0, and the reverse active power meter reading is greater than "set upper limit coefficient" * forward active power meter reading, output "Suspected two phases reversed".

[0122] When the forward and reverse active power meter readings for the user undergoing anomaly screening do not meet the preset meter readings, it is determined whether the average voltage value for that user is less than the first preset voltage threshold. Specifically, if the average voltage value of any phase on any day after installation meets the "voltage value < 180", then "phase voltage is abnormal" is output. Otherwise, it is determined whether the voltage value for the user undergoing anomaly screening meets the second preset voltage threshold. If any phase at a certain point in time shows "voltage value = 0", then "phase voltage is lost" is output. Otherwise, the asset number in the reverse forward list (attached separately) is matched with the asset number before replacement in the meter replacement list of the marketing system. The matched user is marked with a color, and "old meter reversed wiring, new meter wiring checked" is output. Finally, the first data anomaly detection result is constructed using the user anomaly detection results for each user undergoing anomaly screening, i.e., the output results for each user undergoing anomaly screening.

[0123] Step 204: If the reference type is determined to be the second reference, then the user data is used to identify data anomalies in the business expansion suspension work order, and a second data anomaly detection result is generated.

[0124] Furthermore, the user data includes business expansion suspension work order data, and step 204 may include the following sub-steps S21-S29:

[0125] S21. Filter the business expansion pause work order data according to the first preset filtering conditions, and generate pause work order data and user pause count.

[0126] S22. Calculate the difference between the expiration date corresponding to the number of times the user pauses the service and the current date to generate the remaining number of pause days.

[0127] S23. If the remaining number of suspension days is less than or equal to the preset number of days, the detection result corresponding to the preset number of days shall be used as the initial data anomaly detection result.

[0128] S24. According to the second preset filtering conditions, perform pause and resume data filtering on the business expansion pause work order data, and generate associated work order numbers.

[0129] S25. Match the associated work order number with the pause work order data to generate the power-on time.

[0130] S26. Update the initial data anomaly detection result by comparing the power supply time with the corresponding expiration date, and generate the target data anomaly detection result and the number of days suspended.

[0131] S27. Determine whether the number of days that have been paused meets the preset pause threshold. If yes, proceed to S28; otherwise, proceed to S29.

[0132] S28. The pause notification corresponding to the preset pause threshold is taken as the second data anomaly detection result.

[0133] S29. Use the target data anomaly detection result as the second data anomaly detection result.

[0134] On January 2nd of each year, a new table is created for business expansion work order data, named according to the current year. Work orders that did not have a "power supply time" in the previous year's table are copied to the current year's table, and the "number of days suspended" and "number of suspensions" are cleared.

[0135] The first preset filtering condition refers to filtering using "pause work order query" as the keyword.

[0136] The preset duration is 15 days. The corresponding detection result for the preset duration is "Suspension is about to expire".

[0137] The second preset filtering condition refers to filtering using "pause and resume work order query" as the keyword.

[0138] The preset suspension threshold refers to work orders that have been suspended for less than 15 days, or work orders with the same user ID whose total suspension days exceed 180, or work orders with more than 2 suspensions. The suspension notification corresponding to the preset suspension threshold is "Suspension Invalid".

[0139] In this embodiment of the invention, the field of "pause work order query" is obtained, which is the pause work order data, and the number of pauses n (n is the number of times the account number appears) is recorded, which is the number of pauses by the user. The difference between the expiration date corresponding to the number of pauses and the current date is calculated to generate the remaining number of pause days. If the remaining number of pause days is ≤15 days, a reminder that the pause is about to expire is sent to the customer service manager every day, and the detection result corresponding to the preset number of days is used as the initial data anomaly detection result.

[0140] The fields obtained by querying the "Suspension and Resumption Work Order" criteria generate the associated work order number. The associated work order number is then matched with the work order number in the suspended work order. The power restoration time is entered, and the notification that the suspension is about to expire is stopped. Then, the comparison between the power restoration time and the corresponding expiration date is used to update the initial data anomaly detection result, and the target data anomaly detection result and the number of days suspended are generated.

[0141] Determine if the number of days the work order has been suspended meets the preset suspension threshold. If so, send a suspension invalid notification to the customer service manager for work orders with a suspension period of less than 15 days, or work orders with a total suspension period of more than 180 days for the same user ID, or work orders with more than 2 suspensions. If not, the target data anomaly detection result is used as the second data anomaly detection result.

[0142] Further, step S26 may include the following sub-steps S261-S267:

[0143] S261. Determine whether the power supply time is less than or equal to the corresponding expiration date. If yes, proceed to S262; otherwise, proceed to S267.

[0144] S262. Determine whether the shutdown year corresponding to the power supply time is consistent with the power supply year corresponding to the power supply time. If yes, execute S263; otherwise, execute S265.

[0145] S263. Calculate the difference between the power supply time and the shutdown time to generate the first number of days that have been suspended.

[0146] S264. Take the first number of days that have been suspended as the number of days that have been suspended, and take the initial data anomaly detection result as the target data anomaly detection result.

[0147] S265. Calculate the difference between the expiration date and the suspension time to generate the second number of suspended days.

[0148] S266. Take the second number of days that have been suspended as the number of days that have been suspended, and take the initial data anomaly detection result as the target data anomaly detection result.

[0149] S267. Update the initial data anomaly detection results based on the power supply time, shutdown time, and expiration date, and generate the target data anomaly detection results and the number of days suspended.

[0150] The preset power supply time refers to January 1st of the year in which power will be supplied.

[0151] In this embodiment of the invention, when the power supply time is less than or equal to the expiration date, and the year of the shutdown time is the same as the year of the power supply time, then the first number of suspended days = power supply time - shutdown time. The first number of suspended days is taken as the total number of suspended days, and the initial data anomaly detection result is taken as the target data anomaly detection result. When the power supply time is less than or equal to the expiration date, and the year of the shutdown time is different from the year of the power supply time, then the second number of suspended days = power supply time - preset power supply time. The second number of suspended days is taken as the total number of suspended days, and the initial data anomaly detection result is taken as the target data anomaly detection result. When the power supply time is greater than the expiration date, the initial data anomaly detection result is updated using the power supply time, shutdown time, and expiration date to generate the target data anomaly detection result and the number of suspended days.

[0152] Further, step S267 may include the following sub-steps S2671-S2675:

[0153] S2671. Determine whether the year of the shutdown time corresponding to the power supply time is consistent with the year of the power supply time corresponding to the power supply time. If yes, execute S2672; otherwise, execute S2674.

[0154] S2672. Calculate the difference between the expiration date and the corresponding suspension time to generate the third number of suspended days.

[0155] S2673. Update the initial data anomaly detection result using the suspension notification corresponding to the third number of suspended days, generate the target data anomaly detection result, and use the third number of suspended days as the number of suspended days.

[0156] S2674. Calculate the difference between the expiration date and the year of power transmission to generate the fourth number of days that have been suspended.

[0157] S2675. Update the initial data anomaly detection result using the suspension notification corresponding to the fourth suspended day, generate the target data anomaly detection result, and use the fourth suspended day as the suspended day.

[0158] The third suspension period corresponds to the number of days already suspended, indicating that the suspension recovery notice has expired. The fourth suspension period corresponds to the number of days already suspended, indicating that the suspension recovery notice has expired.

[0159] In this embodiment of the invention, when the power supply time is greater than the expiration date, it is determined whether the year of the shutdown time corresponding to the power supply time is consistent with the year of power supply. When the power supply time is greater than the expiration date and the year of shutdown time is consistent with the year of power supply time, the third number of suspended days is recorded as expiration date - shutdown time, a suspension recovery overdue notification is sent, that is, the initial data anomaly detection result is updated using the suspension notification corresponding to the third number of suspended days, the target data anomaly detection result is generated, and the third number of suspended days is taken as the number of suspended days.

[0160] When the power supply time is greater than the expiration date, and the year of the shutdown time is inconsistent with the year of the power supply time, then the fourth number of suspended days = expiration date - preset power supply time. A suspension recovery overdue notification is sent to the customer service manager. That is, the suspension notification corresponding to the fourth number of suspended days is used to update the initial data anomaly detection result, generate the target data anomaly detection result, and use the fourth number of suspended days as the number of suspended days.

[0161] Step 205: If the reference type is determined to be the third reference, the initial business work order is supplemented and updated according to the market attributes to generate the target business work order.

[0162] In this embodiment of the invention, when the reference type is determined to be the third reference, it checks whether there are transmission and distribution prices for market-oriented trading users and grid-purchased users with market attributes. If not, it registers the missing transmission and distribution price files. It also checks whether there are transmission and distribution prices for ordinary electricity users with market attributes. If so, it registers the missing transmission and distribution price settings and deletes them, thereby obtaining the target business work order.

[0163] Step 206: Match and verify the user electricity bill list in the target business work order with the preset electricity bill list to generate abnormal electricity bill data.

[0164] In this embodiment of the invention, electricity bill lists that meet preset criteria from the target business work orders are selected to generate user electricity bill lists. The preset criteria refer to users with a 1.5x mark. All electricity bill lists corresponding to users with the 1.5x mark in the target business work orders are used to construct the user electricity bill list. Then, the user electricity bill list is matched and verified against the preset electricity bill list to generate abnormal electricity bill data. By matching the user electricity bill list with the list of electricity bills payable at 1.5x the preset electricity bill list, errors and omissions in electricity billing are checked, and abnormal electricity bill data is generated.

[0165] Step 207: Compare the user list in the target business work order with the pre-designed fee policy to generate electricity fee processing data.

[0166] In this embodiment of the invention, a user list is generated by selecting a list of users in the target business work order who meet the preset acceptance conditions. The preset acceptance conditions refer to a household with multiple residents. A user list is generated by selecting a list of users in the target business work order who have registered a household with multiple residents. The user list is compared with a pre-designed billing policy to generate electricity billing data. The pre-designed billing policy refers to the billing policy corresponding to a household with multiple residents. The user list is then checked for expired users subject to the pre-designed billing policy for households with multiple residents, and it is determined whether the electricity price has been cancelled as required, thereby obtaining the electricity billing data.

[0167] Step 208: Compare the demand meters in the target work order with the preset meters to generate demand meter data.

[0168] In this embodiment of the invention, meter users whose basic electricity charge calculation method is a preset method are selected from the target business work orders. The preset method is to calculate the basic electricity charge based on actual maximum demand or based on maximum demand. Entering the work order, clicking the accounting center review stage, and selecting users whose basic electricity charge calculation method has been changed to be based on actual maximum demand or based on maximum demand as meter users. The demand meter corresponding to the meter user is compared with the preset meter, generating demand meter data. The preset meter refers to the maximum demand meter set based on actual needs. It is determined whether the meter user has a maximum demand meter; if not, it is registered, the missing maximum demand meter is added to the file, and demand meter data is generated.

[0169] Step 209: Construct a third data anomaly detection result using abnormal electricity bill data, electricity bill processing data, and demand meter data.

[0170] In this embodiment of the invention, after calculating the abnormal electricity bill data, electricity bill processing data and demand meter data corresponding to each user, the third data anomaly detection result is obtained by using all the abnormal electricity bill data, electricity bill processing data and demand meter data.

[0171] In this embodiment of the invention, when user business demand data is received, user data corresponding to each user in the user business demand data is selected from a preset database. The user business demand data is used for reference type identification to determine the judgment reference type. If the judgment reference type is the first judgment reference, the user data is used to identify meter rotation data anomalies, generating a first data anomaly detection result. If the judgment reference type is the second judgment reference, the user data is used to identify business expansion suspension work order data anomalies, generating a second data anomaly detection result. If the judgment reference type is the third judgment reference, the initial business work order is supplemented and updated according to market attributes to generate a target business work order. Electricity bill lists that meet preset criteria in the target business work order are selected to generate user electricity bill lists. The user electricity bill lists are matched and verified with preset electricity bill lists to generate electricity bill anomaly data. A user list is selected from the target business work order to meet preset acceptance conditions. The user list is compared with a preset fee policy to generate electricity bill processing data. Meter users in the target business work order whose basic electricity bill calculation method is a preset method are selected. The system compares the demand meters corresponding to each user with preset meters to generate demand meter data. It then constructs a third data anomaly detection system using abnormal electricity bill data, electricity bill processing data, and demand meter data. By identifying data anomalies through user business demand data and user data, it achieves a comprehensive display centered on meter rotation data, transformer downtime, and basic marketing records. This supports business management personnel in decision-making, improves customer service quality, and ensures a high level of efficient power supply service capabilities.

[0172] Please see Figure 3 , Figure 3 This is a structural block diagram of a data anomaly detection system provided in Embodiment 3 of the present invention.

[0173] Example 3 of this invention provides a data anomaly detection system, comprising:

[0174] User data generation module 301 is used to acquire user business requirement data, select user data from a preset database according to the user business requirement data, and generate user data corresponding to the user business requirements.

[0175] The reference type determination module 302 is used to identify the reference type using user business requirement data and determine the reference type.

[0176] The first data anomaly detection result generation module 303 is used to generate a first data anomaly detection result by using user data to identify anomalies in meter rotation data if the determination reference type is the first determination reference.

[0177] The second data anomaly detection result generation module 304 is used to generate a second data anomaly detection result by using user data to identify data anomalies in business expansion suspension work orders if the determination reference type is the second determination reference.

[0178] The third data anomaly detection result generation module 305 is used to identify basic file data anomalies and generate a third data anomaly detection result if the determination reference type is a third determination reference.

[0179] Optionally, user data includes change data in the operation log of electricity metering devices. The first data anomaly detection result generation module 303 includes:

[0180] The newly installed continuous data judgment module is used to determine whether the newly installed continuous data in the change data of the power metering device operation ledger is at a preset threshold.

[0181] The power metering device operation log change data judgment module is used to determine whether the power metering device operation log change data has a preset current type if no.

[0182] The first data anomaly detection result generation module is used to perform anomaly screening on the change data of the power metering device operation ledger if the anomaly is found, and generate the first data anomaly detection result.

[0183] The first data anomaly detection result first generation submodule is used to take the anomaly result corresponding to the preset current type as the first data anomaly detection result if no.

[0184] The first data anomaly detection result second generation submodule is used to take the anomaly result corresponding to the preset threshold as the first data anomaly detection result if the anomaly is detected.

[0185] Optionally, the change data in the operation ledger of the electricity metering device includes forward and reverse active power meter code data, average voltage value, voltage value, and asset number. The first data anomaly detection result generation module can perform the following steps:

[0186] According to the preset user screening rules, the users corresponding to the changes in the operation ledger of the power metering device are screened, and multiple abnormal screening users are generated.

[0187] Determine whether the positive and negative active power meter code data corresponding to the abnormal screening user are in the preset meter code state.

[0188] If so, the abnormal result of the table code corresponding to the preset table code status will be used as the abnormal detection result of the user corresponding to the abnormal screening user.

[0189] If not, determine whether the average voltage value corresponding to the abnormal screening user is less than the first preset voltage threshold.

[0190] If so, the voltage anomaly result corresponding to the first preset voltage threshold will be used as the user anomaly detection result corresponding to the anomaly screening user.

[0191] If not, determine whether the voltage value corresponding to the abnormal screening user is the second preset voltage threshold.

[0192] If so, the voltage anomaly result corresponding to the second preset voltage threshold will be used as the user anomaly detection result corresponding to the anomaly screening user.

[0193] If not, the asset number corresponding to the user in the anomaly screening will be matched with the preset marketing system's table replacement list to generate the user anomaly detection result corresponding to the user in the anomaly screening.

[0194] The first set of data anomaly detection results is constructed using all user anomaly detection results.

[0195] Optionally, the user data includes business expansion suspension work order data, and the second data anomaly detection result generation module 304 includes:

[0196] The pause work order data and user pause count generation module is used to filter the pause work order data according to the first preset filtering conditions, and generate pause work order data and user pause count.

[0197] The remaining pause days generation module is used to calculate the difference between the expiration date corresponding to the number of pauses by the user and the current date to generate the remaining pause days.

[0198] The initial data anomaly detection result generation module is used to take the detection result corresponding to the preset number of days as the initial data anomaly detection result if the remaining suspension days are less than or equal to the preset number of days.

[0199] The associated work order number generation module is used to filter the pause and resume data of business expansion pause work order data according to the second preset filtering conditions and generate associated work order numbers.

[0200] The power supply time generation module is used to match the associated work order number with the pause work order data to generate the power supply time.

[0201] The target data anomaly detection result and suspended days generation module is used to update the initial data anomaly detection result by comparing the power supply time with the corresponding expiration date, and to generate the target data anomaly detection result and suspended days.

[0202] The module for determining the number of days that have been paused is used to determine whether the number of days that have been paused meets the preset pause threshold.

[0203] The second data anomaly detection result first generation submodule is used to take the pause notification corresponding to the preset pause threshold as the second data anomaly detection result if the condition is met.

[0204] The second data anomaly detection result generation submodule is used to take the target data anomaly detection result as the second data anomaly detection result if the result is not found.

[0205] Optionally, the target data anomaly detection result and suspended days generation module can perform the following steps:

[0206] Determine whether the power supply time is less than or equal to the corresponding expiration date;

[0207] If so, then determine whether the shutdown year corresponding to the power supply time is consistent with the power supply year corresponding to the power supply time;

[0208] If so, calculate the difference between the power supply time and the shutdown time to generate the first number of days that have been suspended;

[0209] The first number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result.

[0210] If not, calculate the difference between the due date and the preset power supply time to generate a second number of days that have been suspended;

[0211] The second number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result.

[0212] If not, the initial data anomaly detection results are updated based on the power supply time, shutdown time, and expiration date, and the target data anomaly detection results and the number of days suspended are generated.

[0213] Optionally, the step of updating the initial data anomaly detection result based on the power supply time, shutdown time, and expiration date, and generating the target data anomaly detection result and the number of days suspended, includes the following steps:

[0214] Determine whether the year of the shutdown corresponding to the power supply time is consistent with the year of power supply corresponding to the power supply time.

[0215] If so, calculate the difference between the expiration date and the corresponding suspension time to generate the third number of suspended days;

[0216] The initial data anomaly detection result is updated using the suspension notification corresponding to the third number of suspended days, the target data anomaly detection result is generated, and the third number of suspended days is used as the number of suspended days;

[0217] If not, calculate the difference between the due date and the corresponding preset power supply time to generate the fourth number of days that have been suspended;

[0218] The initial data anomaly detection result is updated using the suspension notification corresponding to the fourth suspended day, the target data anomaly detection result is generated, and the fourth suspended day is taken as the suspended day.

[0219] Optionally, user data includes market attributes and initial business work orders. The third data anomaly detection result generation module 305 includes:

[0220] The target business work order generation module is used to supplement and update the initial business work order according to market attributes, and generate the target business work order.

[0221] The electricity bill anomaly data generation module is used to match and verify the user's electricity bill list in the target business work order with the preset electricity bill list to generate electricity bill anomaly data.

[0222] The electricity billing data generation module is used to compare the user list in the target business work order with the pre-designed fee policy to generate electricity billing data.

[0223] The demand meter data generation module is used to compare the demand meters in the target business work order with the preset meters to generate demand meter data.

[0224] The third data anomaly detection result generation submodule is used to construct the third data anomaly detection result using abnormal electricity bill data, electricity bill processing data, and demand meter data.

[0225] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor performs a data anomaly detection method as described in any of the above embodiments.

[0226] The memory can be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. The memory has storage space for program code used to perform any of the method steps described above. For example, the storage space for program code may include individual program codes for implementing the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the data anomaly detection method described above.

[0227] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the data anomaly detection method as described in any of the above embodiments.

[0228] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0229] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0230] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0231] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0232] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0233] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting data anomalies, characterized in that, include: When user business requirement data is received, user data corresponding to each user in the user business requirement data is selected from the preset database; The reference type is identified and determined using the user business requirement data. If the determination reference type is the first determination reference, then the user data is used to identify abnormal electricity meter rotation data and generate a first data abnormality detection result. If the determination reference type is the second determination reference, then the user data is used to identify abnormal data in the business expansion suspension work order, and a second data abnormality detection result is generated. If the determination reference type is a third determination reference, then the user data is used to identify basic profile data anomalies and generate a third data anomaly detection result. The user data includes business expansion suspension work order data; The step of using the user data to identify data anomalies in business expansion suspension work orders and generating a second data anomaly detection result includes: The business expansion pause work order data is filtered according to the first preset filtering conditions to generate pause work order data and user pause count; The remaining number of suspension days is calculated by comparing the expiration date corresponding to the number of user suspensions with the current date; If the remaining number of suspension days is less than or equal to the preset number of days, the detection result corresponding to the preset number of days will be used as the initial data anomaly detection result. The business expansion pause work order data is filtered for pause recovery data according to the second preset filtering conditions to generate associated work order numbers; The associated work order number is matched with the pause work order data to generate the power supply time; The initial data anomaly detection result is updated by comparing the power supply time with the corresponding expiration date, and the target data anomaly detection result and the number of days suspended are generated. Determine whether the number of days that have been suspended meets the preset suspension threshold; If so, the pause notification corresponding to the preset pause threshold will be taken as the second data anomaly detection result; If not, the target data anomaly detection result shall be taken as the second data anomaly detection result; The step of updating the initial data anomaly detection result by comparing the power supply time with the corresponding expiration date, and generating the target data anomaly detection result and the number of days suspended, includes: Determine whether the power supply time is less than or equal to the corresponding expiration date; If so, determine whether the shutdown year corresponding to the shutdown time of the power supply time is consistent with the power supply year corresponding to the power supply time; If so, calculate the difference between the power supply time and the shutdown time to generate the first number of days that have been suspended; The first number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result; If not, calculate the difference between the expiration date and the preset power supply time to generate a second number of days that have been suspended; The second number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result; If not, the initial data anomaly detection result is updated based on the power supply time, the shutdown time, and the expiration date to generate the target data anomaly detection result and the number of days suspended; The step of updating the initial data anomaly detection result based on the power supply time, the shutdown time, and the expiration date, and generating the target data anomaly detection result and the number of days suspended, includes: Determine whether the year of the shutdown time corresponding to the power supply time is consistent with the year of power supply corresponding to the power supply time; If so, calculate the difference between the expiration date and the corresponding suspension time to generate the third number of suspended days; The initial data anomaly detection result is updated using the suspension notification corresponding to the third number of suspended days, a target data anomaly detection result is generated, and the third number of suspended days is used as the number of suspended days; If not, calculate the difference between the expiration date and the corresponding preset power supply time to generate a fourth number of days that have been suspended; The initial data anomaly detection result is updated using the suspension notification corresponding to the fourth number of suspended days, a target data anomaly detection result is generated, and the fourth number of suspended days is used as the number of suspended days.

2. The data anomaly detection method according to claim 1, characterized in that, The user data includes change data in the operation log of the electricity metering device; the step of using the user data to identify anomalies in meter rotation data and generating a first data anomaly detection result includes: Determine whether the newly installed continuous data in the operation log change data of the power metering device is at a preset threshold; If not, determine whether the change data in the operation log of the power metering device has a preset current type; If so, the change data in the operation ledger of the power metering device will be screened for anomalies to generate a first data anomaly detection result; If not, the abnormal result corresponding to the preset current type shall be taken as the first data abnormality detection result; If so, the abnormal result corresponding to the preset threshold shall be taken as the first data anomaly detection result.

3. The data anomaly detection method according to claim 2, characterized in that, The data change in the operation ledger of the electricity metering device includes forward and reverse active power meter code data, average voltage value, voltage value, and asset number; the step of performing anomaly screening on the data change in the operation ledger of the electricity metering device to generate a first data anomaly detection result includes: According to the preset user screening rules, the users corresponding to the changes in the operation ledger of the power metering device are screened to generate multiple abnormal screening users; Determine whether the positive and negative active power meter code data corresponding to the abnormal screening user are in the preset meter code state; If so, the abnormal result of the table code corresponding to the preset table code status shall be used as the abnormal detection result of the user corresponding to the abnormal screening user. If not, then determine whether the average voltage value corresponding to the abnormal screening user is less than the first preset voltage threshold. If so, the voltage anomaly result corresponding to the first preset voltage threshold shall be taken as the user anomaly detection result corresponding to the anomaly screening user. If not, determine whether the voltage value corresponding to the abnormal screening user is the second preset voltage threshold. If so, the voltage anomaly result corresponding to the second preset voltage threshold shall be taken as the user anomaly detection result corresponding to the anomaly screening user; If not, the asset number corresponding to the abnormal screening user is matched with the preset marketing system table replacement list to generate the user abnormality detection result corresponding to the abnormal screening user; Using all the aforementioned user anomaly detection results, a first data anomaly detection result is constructed.

4. The data anomaly detection method according to claim 1, characterized in that, The user data includes market attributes and initial business work orders; the step of using the user data to identify basic profile data anomalies and generate a third data anomaly detection result includes: The initial business work order is supplemented and updated according to the market attributes to generate the target business work order; The user's electricity bill in the target business work order is matched and verified with the preset electricity bill, and abnormal electricity bill data is generated. The user list in the target business work order is compared with the pre-designed fee policy to generate electricity bill processing data; The demand meter data is generated by comparing the demand meter data in the target business work order with the preset meter data. The abnormal electricity bill data, the electricity bill processing data, and the demand meter data are used to construct a third data anomaly detection result.

5. A data anomaly detection system, characterized in that, include: The user data generation module is used to acquire user business requirement data, select user data from a preset database according to the user business requirement data, and generate user data corresponding to the user business requirement. The reference type determination module is used to identify the reference type using the user business requirement data and determine the reference type. The first data anomaly detection result generation module is used to generate a first data anomaly detection result by using the user data to identify anomalies in the electricity meter rotation data if the determination reference type is the first determination reference. The second data anomaly detection result generation module is used to identify data anomalies in business expansion suspension work orders by using the user data if the determination reference type is the second determination reference, and generate the second data anomaly detection result. The third data anomaly detection result generation module is used to generate a third data anomaly detection result by using the user data to identify basic file data anomalies if the determination reference type is a third determination reference. The user data includes business expansion suspension work order data, and the second data anomaly detection result generation module includes: The pause work order data and user pause count generation module is used to filter the pause work order data according to the first preset filtering conditions and generate pause work order data and user pause count; the remaining pause days generation module is used to calculate the difference between the due date corresponding to the user pause count and the current date to generate the remaining pause days; The initial data anomaly detection result generation module is used to take the detection result corresponding to the preset number of days as the initial data anomaly detection result if the remaining suspension days are less than or equal to the preset number of days. The associated work order number generation module is used to filter the business expansion paused work order data according to the second preset filtering conditions and generate associated work order numbers. The power supply time generation module is used to match the associated work order number with the pause work order data to generate the power supply time. The target data anomaly detection result and suspended days generation module is used to update the initial data anomaly detection result by comparing the power supply time with the corresponding expiration date, and to generate the target data anomaly detection result and the suspended days. The module for determining the number of days that have been paused is used to determine whether the number of days that have been paused meets the preset pause threshold. The first generation submodule for the second data anomaly detection result is used to take the pause notification corresponding to the preset pause threshold as the second data anomaly detection result if the result is true. The second data anomaly detection result second generation submodule is used to take the target data anomaly detection result as the second data anomaly detection result if not. The target data anomaly detection result and the suspended days generation module perform the following steps: Determine whether the power supply time is less than or equal to the corresponding expiration date; If so, determine whether the shutdown year corresponding to the shutdown time of the power supply time is consistent with the power supply year corresponding to the power supply time; If so, calculate the difference between the power supply time and the shutdown time to generate the first number of days that have been suspended; The first number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result; If not, calculate the difference between the expiration date and the preset power supply time to generate a second number of days that have been suspended; The second number of days that have been suspended is taken as the number of days that have been suspended, and the initial data anomaly detection result is taken as the target data anomaly detection result; If not, the initial data anomaly detection result is updated based on the power supply time, the shutdown time, and the expiration date to generate the target data anomaly detection result and the number of days suspended; The step of updating the initial data anomaly detection result based on the power supply time, the shutdown time, and the expiration date, and generating the target data anomaly detection result and the number of days suspended, includes: Determine whether the year of the shutdown time corresponding to the power supply time is consistent with the year of power supply corresponding to the power supply time; If so, calculate the difference between the expiration date and the corresponding suspension time to generate the third number of suspended days; The initial data anomaly detection result is updated using the suspension notification corresponding to the third number of suspended days, a target data anomaly detection result is generated, and the third number of suspended days is used as the number of suspended days; If not, calculate the difference between the expiration date and the corresponding preset power supply time to generate a fourth number of days that have been suspended; The initial data anomaly detection result is updated using the suspension notification corresponding to the fourth number of suspended days, a target data anomaly detection result is generated, and the fourth number of suspended days is used as the number of suspended days.

6. An electronic device, characterized in that, The device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the data anomaly detection method as described in any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the data anomaly detection method as described in any one of claims 1 to 4.

Images