A method and related device for checking power grid data quality
By constructing a linear regression function model and using the Dixon criterion in the grid data quality calibration, the problems of complexity and operation difficulty of traditional methods are solved, and the accuracy and efficiency of grid data quality calibration are improved.
Patent Information
- Application Number
- CN202411242633.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-05
AI Technical Summary
Traditional grid data quality verification methods are complex and difficult to operate, resulting in low grid troubleshooting and repair efficiency.
Based on the power grid topological relationship, a pre-constructed linear regression function model is constructed. Through real-time telemetry data and historical data verification, abnormal data is identified and filtered, and data outlier judgment is used to accurately identify and locate abnormal data.
It improves the accuracy and authenticity of grid data quality verification, simplifies the operation process, and significantly improves the efficiency of troubleshooting and repair.
Smart Images

Figure CN118761186B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power grid data evaluation, and in particular relates to a power grid data quality verification method and related devices. Background Art
[0002] The power grid refers to the whole composed of various voltage transformers and transmission and distribution lines in the power system, which covers various power equipment such as switches, transformers, feeders and circuit breakers. With the development of the power system, the scale of the power grid continues to expand. Therefore, the implementation of precise and sensitive abnormal monitoring of power grid data has become the key to maintaining the health of the power grid. At present, the traditional power grid data quality verification is mainly concentrated in a single professional field, and the importance of cross-professional data joint operations is often ignored. The quality verification process is complicated and the operation is difficult, which reduces the efficiency of power grid troubleshooting and repair. Summary of the invention
[0003] In view of the technical problems existing in the prior art, the present invention provides a power grid data quality verification method and related devices to solve the problems of traditional power grid data quality verification methods, such as complex quality verification process and difficult operation, which reduces the efficiency of power grid troubleshooting and repair.
[0004] In order to achieve the above object, the technical solution adopted by the present invention is:
[0005] The present invention provides a method for checking the quality of power grid data, comprising:
[0006] Based on the power grid topology relationship, determine the equipment to be analyzed; wherein the equipment to be analyzed includes an upstream key node device and a number of downstream devices, and there is a power grid topology relationship between the several downstream devices and the upstream key node device;
[0007] Obtain real-time telemetry data from the equipment to be analyzed;
[0008] Determine whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model; if so, output the power grid data quality verification result; if not, use the preset judgment criteria to perform data outlier judgment on the real-time telemetry data of the device to be analyzed, and output the power grid data quality verification result.
[0009] Furthermore, the construction process of the pre-built linear regression function model is as follows:
[0010] Obtain historical measurement data of the equipment to be analyzed; wherein the historical measurement data of the equipment to be analyzed includes historical telesignaling data of upstream key node equipment, historical telemetry data of upstream key node equipment, and historical telemetry data of several downstream equipment;
[0011] Based on the power grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed, a historical measurement data verification matrix with the time sequence of the acquisition points is generated;
[0012] Perform consistency check on the data in the to-be-checked matrix of historical measurement data with the time sequence of acquisition points to obtain a consistency check result matrix;
[0013] According to the consistency check result matrix, a check matrix of historical telemetry data with natural day time series is generated;
[0014] Perform normal distribution calibration on the data in the to-be-calibrated matrix of historical telemetry data with natural daily time series to obtain a normal distribution calibration result matrix;
[0015] A linear regression function model is constructed according to the normal distribution calibration result matrix to obtain a pre-constructed linear regression function model.
[0016] Further, the historical measurement data to be verified matrix with the time sequence of the acquisition points includes a plurality of historical measurement data columns, and the plurality of historical measurement data columns are arranged according to the time sequence of the acquisition points;
[0017] Each historical measurement data column includes a historical telesignaling data row of an upstream key node device, a historical telemetry data row of an upstream key node device, and a historical telemetry data row of several downstream devices.
[0018] Furthermore, the process of performing consistency check on the data in the historical measurement data to be checked matrix with the time sequence of the acquisition points to obtain the consistency check result matrix is as follows: performing a preset check criterion on each column of the historical measurement data to be checked matrix with the time sequence of the acquisition points to obtain the consistency check result matrix;
[0019] Among them, the preset verification criteria are as follows:
[0020] When the data in the historical telemetering data row of the upstream key node device is 0, determine whether more than 1 / 3 of the data in the historical telemetering data rows of several downstream devices are not 0; if so, modify the data that is 0 in the historical telemetering data row of the upstream key node device to 1; if not, modify the data that is not 0 in the historical telemetering data rows of several downstream devices to 0;
[0021] When the data in the historical telemetering data row of the upstream key node device is 1, determine whether the data in the historical telemetry data row of the upstream key node device is greater than the sum of all data in the historical telemetry data rows of several downstream devices; if not, all data in the historical measurement data column will be eliminated.
[0022] Furthermore, the historical telemetry data to be verified matrix with natural day time series includes several historical telemetry data columns, and the several historical telemetry data columns are arranged according to the natural day time series; wherein each historical telemetry data column includes historical telemetry data rows of upstream key node equipment and historical telemetry data rows of several downstream equipment.
[0023] Further, according to the normal distribution calibration result matrix, a linear regression function model is constructed to obtain a pre-constructed linear regression function model, as follows:
[0024] Complete the missing data in the normal distribution calibration result matrix to obtain the equipment effective data matrix;
[0025] Taking the historical telemetry data of upstream key node equipment in the equipment effective data matrix as the dependent variable, and the historical telemetry data of several downstream equipment in the equipment effective data matrix as the independent variable, a linear regression function is constructed for the same collection point on different natural days to obtain a pre-constructed linear regression function model.
[0026] Furthermore, the preset determination criterion is the Dixon criterion.
[0027] The present invention also provides a power grid data quality verification system, comprising:
[0028] The device determination module is used to determine the device to be analyzed based on the power grid topology relationship; wherein the device to be analyzed includes an upstream key node device and a plurality of downstream devices, and there is a power grid topology relationship between the plurality of downstream devices and the upstream key node device;
[0029] A real-time data acquisition module is used to obtain real-time telemetry data of the equipment to be analyzed;
[0030] The data quality verification module is used to determine whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model; if so, the power grid data quality verification result is output; if not, the real-time telemetry data of the device to be analyzed is judged as outlier using the preset judgment criteria, and the power grid data quality verification result is output.
[0031] The present invention also provides a power grid data quality checking device, comprising:
[0032] Memory for storing computer programs;
[0033] A processor is used to implement the steps of the power grid data quality verification method when executing the computer program.
[0034] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the power grid data quality verification method are implemented.
[0035] Compared with the prior art, the present invention has the following beneficial effects:
[0036] The power grid data quality verification method provided by the present invention can effectively identify and filter abnormal data that does not conform to linear planning by introducing a pre-built linear regression model to perform quality verification on the real-time telemetry data of the equipment to be analyzed; it uses a preset judgment criterion to perform data outlier judgment on the abnormal data that does not conform to linear planning, so as to accurately identify and locate the outliers in the real-time telemetry data of the equipment to be analyzed, thereby improving the accuracy of the power grid data quality verification; secondly, the power grid topology relationship is used to determine the equipment to be analyzed, so that the power grid data quality verification is closer to the actual operation of the power grid, and the influence of switch displacement or power grid current law can be fully considered, thereby ensuring the authenticity of the power grid data quality verification; the present invention can be used for quality verification of different professional data of the power system, the operation process is simple, and the difficulty of power grid data anomaly detection is reduced; at the same time, it can significantly improve the processing time efficiency of power grid data quality verification, accurately locate abnormal data, and greatly improve the efficiency of power grid troubleshooting and repair.
[0037] Furthermore, by obtaining the historical measurement data of the equipment to be analyzed and generating a historical measurement data verification matrix with the time series of the collection points based on the power grid topology relationship, the integrity and relevance of the historical measurement data are ensured; through consistency verification and normal distribution verification, abnormal data are effectively eliminated and corrected to ensure that the data used to construct the linear regression function model has a high quality, which lays the foundation for improving the accuracy and stability of the linear regression function model.
[0038] The power grid data quality verification system, power grid data quality verification device and computer-readable storage medium provided by the present invention have all the advantages of the above-mentioned power grid data quality verification method. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0040] Figure 1 A flow chart of the power grid data quality verification method provided in Example 1;
[0041] Figure 2A structural block diagram of a power grid data quality verification system provided in Example 2;
[0042] Figure 3 This is a structural block diagram of the power grid data quality verification device provided in Example 3. DETAILED DESCRIPTION
[0043] In order to make the technical problems, technical solutions and beneficial effects solved by this application clearer, the technical solutions in the embodiments of this application will be described clearly and completely in combination with the drawings in the embodiments of this application; obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0044] Example 1
[0045] As attached Figure 1 As shown, this embodiment 1 provides a method for checking the quality of power grid data, comprising the following steps:
[0046] Step 1: Based on the grid topology, determine the equipment to be analyzed; the equipment to be analyzed includes upstream key node equipment and several downstream equipment, and there is a grid topology relationship between the several downstream equipment and the upstream key node equipment. For example, when the upstream key node equipment is a distribution transformer, the several downstream equipment are several low-voltage users.
[0047] Specifically, the relationship set of the devices to be analyzed is:
[0048]
[0049] in, is the relationship set of the equipment to be analyzed; It is the upstream key node equipment; For the Downstream equipment, ; The total number of downstream devices.
[0050] It should be noted that the grid topology relationship refers to the connection relationship and configuration method between various components in the grid; specifically in this embodiment 1, the grid topology relationship refers to the connection relationship and configuration method between the upstream key node equipment and several downstream equipment.
[0051] Step 2: Obtain historical measurement data of the device to be analyzed. The process of obtaining historical measurement data of the device to be analyzed includes the following steps: randomly selecting a number of natural days; obtaining measurement data of the collection points of the device to be analyzed on each natural day, that is, obtaining historical measurement data of the device to be analyzed.
[0052] Specifically, the historical measurement data of the equipment to be analyzed includes historical telesignaling data of upstream key node equipment, historical telemetry data of upstream key node equipment, and historical telemetry data of several downstream equipment.
[0053] Preferably, the type of historical telemetry data of the upstream key node device is the current analog quantity of the upstream key node device, the active power analog quantity of the upstream key node device, or the reactive power analog quantity of the upstream key node device, and the type of historical telemetry data of the downstream device is the current analog quantity of the downstream device, the active power analog quantity of the downstream device, or the reactive power analog quantity of the downstream device.
[0054] It should be noted that the historical telemetering data of the upstream key node equipment includes but is not limited to the telemetering data of the upstream key node equipment or the historical telemetering data of the downstream equipment at the same time.
[0055] Step 3: Based on the power grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed, a historical measurement data verification matrix with the time sequence of the acquisition points is generated.
[0056] Among them, the measurement data to be verified matrix with the time sequence of the collection points includes several historical measurement data columns, and the several historical measurement data columns are arranged according to the time sequence of the collection points; each historical measurement data column includes historical telesignaling data rows of upstream key node equipment, historical telemetry data rows of upstream key node equipment and historical telemetry data rows of several downstream equipment.
[0057] It should be noted that, in each historical measurement data column, the data in the historical telesignaling data row of the upstream key node device corresponds to the historical telesignaling data of a certain collection point of the upstream key node device, the data in the historical telemetry data row of the upstream key node device corresponds to the historical telemetry data of a certain collection point of the upstream key node device, and the data in the historical telemetry data row of each downstream device corresponds to the historical telemetry data of a certain collection point of each downstream device.
[0058] Specifically, the matrix of historical measurement data to be verified with the time sequence of acquisition points is:
[0059]
[0060] in, For the The historical measurement data to be verified matrix with the time sequence of the collection points corresponding to each natural day; For upstream key node equipment Historical telesignaling data of each collection point; For upstream key node equipment Historical telemetry data of each collection point; For the The downstream device Historical telemetry data of each collection point; is the serial number of the collection point within a natural day, ; For the The total number of points collected within a natural day.
[0061] Step 4: Perform consistency check on the data in the historical measurement data to be checked matrix with the acquisition point time sequence to obtain a consistency check result matrix. Specifically, perform a preset check criterion on each column of historical measurement data in the historical measurement data to be checked matrix with the acquisition point time sequence to obtain a consistency check result matrix.
[0062] It should be noted that under the power grid topology, when the upper-level key node equipment is in a power outage state, that is, the telemetry data of the upper-level key node equipment is 0, the telemetry data of the upper-level key node equipment and the telemetry data of the downstream equipment should both be 0; when the upper-level key node equipment is in a normal working state, that is, the telemetry data of the upstream key node equipment is 1, then the telemetry data of the upper-level key node equipment must be greater than the sum of the telemetry data of several downstream devices.
[0063] Therefore, the default verification criteria are as follows:
[0064] Step 41, determine whether the data in the historical remote signaling data row of the upstream key node device is 0; if so, execute step 42; otherwise, execute step 45.
[0065] Step 42, determine whether more than 1 / 3 of the data in the historical telemetry data rows of several downstream devices are not zero; if so, execute step 43; if not, execute step 44.
[0066] Step 43: Change the data that is 0 in the historical remote signaling data row of the upstream key node device to 1.
[0067] Step 44: Change the non-zero data in the historical telemetry data rows of several downstream devices to zero.
[0068] Step 45: determine whether the data in the historical telemetry data row of the upstream key node device is greater than the sum of all the data in the historical telemetry data rows of several downstream devices; if not, remove all the data in the historical measurement data column.
[0069] To the first The matrix of historical measurement data to be verified with the time sequence of acquisition points corresponding to natural days Middle Taking the process of executing the preset calibration criteria for the historical measurement data corresponding to each collection point as an example, the steps are as follows:
[0070] When the upstream key node equipment Historical remote signal data of each collection point When it is 0, it determines the Whether more than 1 / 3 of the historical telemetry data of the collection point is not 0; if so, the first Historical remote signal data of each collection point Replace with 1; if it does not exist, all downstream devices The non-zero data in the historical telemetry data of each collection point is modified to 0.
[0071] When the upstream key node equipment Historical telesignal data of the collection points When it is 1, it determines the upstream key node device Historical telemetry data of each collection point , whether it satisfies: , , If not, then The matrix of historical measurement data to be verified with the time sequence of acquisition points corresponding to natural days The The historical measurement data columns corresponding to the collection points are removed.
[0072] Step 5: Generate a matrix of historical telemetry data to be verified with a natural day time series based on the consistency verification result matrix. Specifically, convert the consistency verification result matrix into historical telemetry data based on a natural day time series; and generate a matrix of historical telemetry data to be verified with a natural day time series based on the historical telemetry data based on a natural day time series.
[0073] Among them, the historical telemetry data to be verified matrix with natural day time series includes several historical telemetry data columns, and the several historical telemetry data columns are arranged according to the natural day time series; each historical telemetry data column includes historical telemetry data rows of upstream key node equipment and historical telemetry data rows of several downstream equipment.
[0074] It should be noted that the data in the historical telemetry data row of the upstream key node device in each historical telemetry data column corresponds to the data in the upstream key node device. The first day of the natural The historical telemetry data of each downstream device is collected at the same time as the historical telemetry data of each downstream device at the The first day of the natural The data corresponds to the historical telemetry data of each collection point.
[0075] Specifically, the historical telemetry data to be verified matrix with natural day time series is:
[0076]
[0077] in, For the The historical telemetry data matrix to be verified with natural day time series corresponding to each collection point; For upstream key node equipment The first day of the natural Historical telemetry data of each collection point; For the The downstream device is The first day of the natural The historical telemetry data of each collection point.
[0078] It should be noted that, assuming that the data collection cycle of upstream key node equipment and downstream equipment is 15 minutes, the number of collection points per natural day is 96, that is, .
[0079] Step 6: Perform normal distribution verification on the data in the matrix of historical telemetry data to be verified with natural daily time series to obtain a normal distribution verification result matrix. Specifically, the process of normal distribution verification is as follows:
[0080] Step 61, determine the effective range of normal distribution, and obtain a predetermined effective range of normal distribution; preferably, the predetermined effective range of normal distribution is ;in, It is the mathematical expectation of the historical telemetry data of the upstream key node equipment or the historical telemetry data of the downstream equipment in the historical telemetry data to be verified matrix with natural day time series; It is the variance of the historical telemetry data of the upstream key node equipment or the historical telemetry data of the downstream equipment in the historical telemetry data to be verified matrix with natural day time series.
[0081] Step 62, compare the data in the matrix of historical telemetry data to be verified with the predetermined valid range of normal distribution; if the data in the matrix of historical telemetry data to be verified with the natural day time series is within the predetermined valid range of normal distribution, it is marked as reasonable validity data; otherwise, it is marked as doubtful validity data, and the doubtful validity data is eliminated to output the normal distribution verification result matrix.
[0082] It should be noted that if a data in the historical telemetry data to be verified matrix with natural daily time series is eliminated, a null value is used to fill the position of the eliminated data, that is, a null value is used to represent the missing data in the normal distribution verification result matrix.
[0083] Step 7: construct a linear regression function model according to the normal distribution check result matrix to obtain a pre-constructed linear regression function model. Specifically, the process of constructing a linear regression function model according to the normal distribution check result matrix is as follows:
[0084] Step 71, complete the missing data in the normal distribution verification result matrix to obtain the equipment effective data matrix; specifically, use the 4-point average method to calculate the missing data in the normal distribution verification result matrix to obtain the missing data calculation value; use the missing data calculation value to replace the empty values in the normal distribution verification result matrix to obtain the equipment effective data matrix; wherein, the calculation process of the missing data calculation value is as follows:
[0085]
[0086] or
[0087] in, Calculate values for missing data; The historical telemetry data is the second natural day before the missing data; The historical telemetry data is the first natural day before the missing data; The historical telemetry data is the first natural day after the missing data; The historical telemetry data is the second natural day after the missing data; The historical telemetry data of the second collection point before the missing data; The historical telemetry data of the first collection point before the missing data; It is the historical telemetry data of the first collection point after missing data; This is the historical telemetry data of the second collection point after the missing data.
[0088] Among them, the equipment effective data matrix is as follows:
[0089]
[0090] in, For the The equipment valid data matrix corresponding to each natural day.
[0091] Step 72: Taking the historical telemetry data of the upstream key node equipment in the equipment effective data matrix as the dependent variable and the historical telemetry data of several downstream equipment in the equipment effective data matrix as the independent variable, a linear regression function is constructed for the same collection point on different natural days to obtain a pre-constructed linear regression function model.
[0092] Among them, the pre-built linear regression function model is specifically:
[0093]
[0094] in, For the The downstream device Characteristic coefficients of historical telemetry data of each collection point; is the error term.
[0095] The constraints of the pre-built linear regression function model are:
[0096]
[0097] in, is the minimum error amount; is the maximum error.
[0098] Step 8: Acquire real-time telemetry data of the device to be analyzed. Specifically, the real-time telemetry data of the device to be analyzed includes real-time telemetry data of upstream key node devices and real-time telemetry data of several downstream devices.
[0099] Step 9: Determine whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model. If yes, execute step 10; if not, execute step 11. Specifically, the process of determining whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model is as follows:
[0100] Substitute the real-time telemetry data of the device to be analyzed into the pre-built linear regression function model to obtain the error term The calculated value of The calculated value of is compared with the constraints of the pre-built linear regression function model; if the error term The calculated value of satisfies the constraints of the pre-built linear regression function model, then execute step 10; if the error term If the calculated value of does not satisfy the constraints of the pre-built linear regression function model, execute step 11.
[0101] It should be noted that if the error term If the calculated value of satisfies the constraints of the pre-built linear regression function model, it means that the real-time telemetry data of the device to be analyzed are all valid data; if the error term If the calculated value of does not meet the constraints of the pre-built linear regression function model, it means that the real-time telemetry data of the device to be analyzed contains data with poor rationality.
[0102] Step 10: Output the power grid data quality verification result. At this time, the power grid data quality verification result is that the real-time telemetry data of the equipment to be analyzed are all valid data.
[0103] Step 11: Use the preset judgment criteria to perform data outlier judgment on the real-time telemetry data of the device to be analyzed, and output the power grid data quality verification result. Specifically, use the Dixon criterion to perform data outlier judgment on the real-time telemetry data of the device to be analyzed, and output the power grid data quality verification result.
[0104] Among them, the process of using the Dixon criterion to determine the data outliers of the real-time telemetry data of the analyzed equipment is as follows:
[0105] Step 111, based on the real-time telemetry data of the upstream key node equipment in the real-time telemetry data of the equipment to be analyzed, obtain the historical telemetry data of the upstream key node equipment within 9 natural days before the real-time telemetry data of the upstream key node equipment from the equipment effective data matrix, and obtain the latest 10 effective telemetry data of the upstream key node equipment.
[0106] It should be noted that the latest 10 valid telemetry data of the upstream key node equipment include the real-time telemetry data of the upstream key node equipment and the historical telemetry data of the upstream key node equipment within 9 natural days before the real-time telemetry data of the upstream key node equipment.
[0107] Step 112: Based on the real-time telemetry data of several downstream devices in the real-time telemetry data of the device to be analyzed, the historical telemetry data of the downstream devices within 9 natural days before the real-time telemetry data of each downstream device is obtained from the device valid data matrix to obtain the latest 10 valid telemetry data of each downstream device.
[0108] It should be noted that the latest 10 valid telemetry data of each downstream device include the real-time telemetry data of each downstream device and the historical telemetry data of the downstream device within 9 natural days before the real-time telemetry data of each downstream device.
[0109] Step 113: Sort the latest 10 valid telemetry data of the upstream key node equipment and the latest 10 valid telemetry data of each downstream device according to the data size ascending rule to obtain the sorting result of the valid telemetry data of the device to be analyzed; and construct a data outlier judgment matrix based on the sorting result of the valid telemetry data of the device to be analyzed.
[0110] Among them, the data outlier judgment matrix is specifically:
[0111]
[0112] in, is the data outlier determination matrix; The latest 10 valid telemetry data of upstream key node devices are sorted in ascending order according to the data size. Telemetry data, ; Calculate statistical values for outliers in real-time telemetry data of upstream key node devices; It is the trustworthy value of the upstream key node equipment; For the The latest 10 valid telemetry data of the downstream devices are sorted in ascending order according to the data size. Telemetry data; For the Calculate outlier statistics for real-time telemetry data from downstream devices; For the The trusted value of a downstream device.
[0113] It should be noted that the real-time telemetry data of the upstream key node equipment is outlier calculation statistics or the first The calculation process of outlier calculation statistics of real-time telemetry data of downstream devices is as follows:
[0114]
[0115] in, Calculate statistics or outliers for real-time telemetry data of upstream key node devices Real-time telemetry data outlier calculation statistics of downstream devices, or ; Real-time telemetry data of upstream key node equipment or real-time telemetry data of downstream equipment; The real-time telemetry data of the upstream key node equipment is the previous telemetry data or the first telemetry data in the valid telemetry data sorting result. The real-time telemetry data of a downstream device is the previous telemetry data in the sorting result of the valid telemetry data; The maximum or the least significant value among the latest 10 valid telemetry data of the upstream key node equipment. The maximum value among the latest 10 valid telemetry data of downstream devices; The minimum value or the first value among the latest 10 valid telemetry data of the upstream key node equipment. The minimum value among the most recent 10 valid telemetry data of the downstream devices.
[0116] In the data outlier judgment matrix, the trust value of the upstream key node equipment The value of is determined according to the outlier calculation statistics of the real-time telemetry data of the upstream key node device; specifically, if the outlier calculation statistics of the real-time telemetry data of the upstream key node device is greater than the preset outlier calculation threshold of the real-time telemetry data of the upstream key node device, then , which means that the real-time telemetry data of the upstream key node device is unreliable; if the outlier calculation statistic of the real-time telemetry data of the upstream key node device is less than the preset outlier calculation threshold of the real-time telemetry data of the upstream key node device, then , which means that the real-time telemetry data of upstream key node equipment is credible.
[0117] Similarly, in the data outlier determination matrix, Trusted value for downstream devices The value of The outlier calculation statistics of the real-time telemetry data of the downstream device are determined; specifically, if the The outlier calculation statistics of the real-time telemetry data of the downstream devices are greater than the preset The outlier calculation threshold of the real-time telemetry data of downstream devices is , which means The real-time telemetry data of the downstream device is unreliable; if the The outlier calculation statistics of the real-time telemetry data of the downstream devices are less than the preset The outlier calculation threshold of the real-time telemetry data of downstream devices is , which means Real-time telemetry data from downstream devices can be trusted.
[0118] Step 114: From the data outlier determination matrix, according to the trust value of the upstream key node device and Trusted value for downstream devices , obtain the unreliable data in the real-time telemetry data of the device to be analyzed, and obtain the equipment list corresponding to the unreliable real-time telemetry data, and output the power grid data quality verification result; it should be noted that, at this time, the power grid data quality verification result is the equipment list corresponding to the unreliable real-time telemetry data.
[0119] The power grid data quality verification method provided in this embodiment 1 uses the topological relationship of the power grid to determine the equipment to be analyzed, and realizes abnormal data analysis based on the physical connection and logical relationship of the power grid, so that the data quality verification is closer to the actual operation of the power grid; by introducing a linear regression function model as a judgment condition for data quality verification, it can effectively identify and filter abnormal data that does not meet the expectations of the model; for real-time telemetry data that does not meet the constraints of the linear regression model, the Dixon criterion is used to perform data outlier judgment to improve the accuracy and reliability of power grid data quality verification; the present invention can fully consider the impact of switch displacement and changes in power grid current laws on the power grid state, meet the quality verification operations of different professional data in the power system, and fundamentally alleviate the heavy workload caused by the power grid data anomaly detection task, and significantly improve the processing time, greatly improving the efficiency of fault detection and repair.
[0120] Example 2
[0121] As attached Figure 2 As shown, this embodiment 2 provides a power grid data quality verification system, including a device determination module, a historical data acquisition module, a first matrix to be verified generation module, a consistency verification module, a second matrix to be verified generation module, a normal distribution verification module, a linear regression model construction module, a real-time data acquisition module and a data quality verification module.
[0122] The device determination module is used to determine the device to be analyzed based on the grid topology relationship. The historical data acquisition module is used to obtain the historical measurement data of the device to be analyzed; the first matrix to be verified generation module is used to generate a matrix of historical measurement data to be verified with the time sequence of the acquisition points based on the grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed; the consistency verification module is used to perform consistency verification on the data in the matrix of historical measurement data to be verified with the time sequence of the acquisition points to obtain a consistency verification result matrix; the second matrix to be verified generation module is used to generate a matrix of historical telemetry data to be verified with the time sequence of the natural day based on the consistency verification result matrix; the normal distribution verification module is used to perform consistency verification on the historical telemetry data to be verified with the time sequence of the natural day The data in the kernel matrix are normally distributed checked to obtain a normal distribution check result matrix; a linear regression model construction module is used to construct a linear regression function model according to the normal distribution check result matrix to obtain a pre-constructed linear regression function model; a real-time data acquisition module is used to obtain the real-time telemetry data of the device to be analyzed; a data quality check module is used to determine whether the real-time telemetry data of the device to be analyzed meets the constraint conditions of the pre-constructed linear regression function model; if so, the power grid data quality check result is output; if not, the real-time telemetry data of the device to be analyzed is judged as outlier by using a preset judgment criterion, and the power grid data quality check result is output.
[0123] Example 3
[0124] As attached Figure 3 As shown, this embodiment 3 provides a power grid data quality verification device, including: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a power grid data quality verification method program; wherein the attached Figure 3 The communication interface in is used to access external devices to obtain data.
[0125] When the processor executes the computer program, the steps of the above-mentioned power grid data quality verification method are implemented, for example:
[0126] Obtain historical measurement data of the equipment to be analyzed; based on the power grid topology relationship between the equipment to be analyzed and the historical measurement data of the equipment to be analyzed, generate a historical measurement data to-be-verified matrix with the time sequence of the acquisition points; perform consistency verification on the data in the historical measurement data to-be-verified matrix with the time sequence of the acquisition points to obtain a consistency verification result matrix; based on the consistency verification result matrix, generate a historical telemetry data to-be-verified matrix with the natural day time sequence; perform normal distribution verification on the data in the historical telemetry data to-be-verified matrix with the natural day time sequence to obtain a normal distribution verification result matrix; based on the normal distribution verification result matrix, construct a linear regression function model to obtain a pre-constructed linear regression function model; obtain real-time telemetry data of the equipment to be analyzed; determine whether the real-time telemetry data of the equipment to be analyzed meets the constraints of the pre-constructed linear regression function model; if so, output the power grid data quality verification result; if not, use the preset judgment criteria to perform data outlier judgment on the real-time telemetry data of the equipment to be analyzed, and output the power grid data quality verification result.
[0127] Alternatively, when the processor executes the computer program, the functions of each module in the above system are realized, for example:
[0128] The device determination module is used to determine the device to be analyzed based on the grid topology relationship. The historical data acquisition module is used to obtain the historical measurement data of the device to be analyzed; the first matrix to be verified generation module is used to generate a matrix of historical measurement data to be verified with the time sequence of the acquisition points based on the grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed; the consistency verification module is used to perform consistency verification on the data in the matrix of historical measurement data to be verified with the time sequence of the acquisition points to obtain a consistency verification result matrix; the second matrix to be verified generation module is used to generate a matrix of historical telemetry data to be verified with the time sequence of the natural day based on the consistency verification result matrix; the normal distribution verification module is used to perform consistency verification on the historical telemetry data to be verified with the time sequence of the natural day The data in the kernel matrix are normally distributed checked to obtain a normal distribution check result matrix; a linear regression model construction module is used to construct a linear regression function model according to the normal distribution check result matrix to obtain a pre-constructed linear regression function model; a real-time data acquisition module is used to obtain the real-time telemetry data of the device to be analyzed; a data quality check module is used to determine whether the real-time telemetry data of the device to be analyzed meets the constraint conditions of the pre-constructed linear regression function model; if so, the power grid data quality check result is output; if not, the real-time telemetry data of the device to be analyzed is judged as outlier by using a preset judgment criterion, and the power grid data quality check result is output.
[0129] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the power grid data quality verification device.
[0130] For example, the computer program can be divided into a device determination module, a historical data acquisition module, a first matrix generation module to be verified, a consistency verification module, a second matrix generation module to be verified, a normal distribution verification module, a linear regression model construction module, a real-time data acquisition module and a data quality verification module, and the specific functions of each module are as follows:
[0131] The device determination module is used to determine the device to be analyzed based on the grid topology relationship. The historical data acquisition module is used to obtain the historical measurement data of the device to be analyzed; the first matrix to be verified generation module is used to generate a matrix of historical measurement data to be verified with the time sequence of the acquisition points based on the grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed; the consistency verification module is used to perform consistency verification on the data in the matrix of historical measurement data to be verified with the time sequence of the acquisition points to obtain a consistency verification result matrix; the second matrix to be verified generation module is used to generate a matrix of historical telemetry data to be verified with the time sequence of the natural day based on the consistency verification result matrix; the normal distribution verification module is used to perform consistency verification on the historical telemetry data to be verified with the time sequence of the natural day The data in the kernel matrix are normally distributed checked to obtain a normal distribution check result matrix; a linear regression model construction module is used to construct a linear regression function model according to the normal distribution check result matrix to obtain a pre-constructed linear regression function model; a real-time data acquisition module is used to obtain the real-time telemetry data of the device to be analyzed; a data quality check module is used to determine whether the real-time telemetry data of the device to be analyzed meets the constraint conditions of the pre-constructed linear regression function model; if so, the power grid data quality check result is output; if not, the real-time telemetry data of the device to be analyzed is judged as outlier by using a preset judgment criterion, and the power grid data quality check result is output.
[0132] The power grid data quality check device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The power grid data quality check device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the attached Figure 3 The schematic diagram is merely an example of a power grid data quality verification device and does not constitute a limitation on the power grid data quality verification device. The power grid data quality verification device may include more or fewer components than shown in the diagram, or a combination of certain components, or different components. For example, the power grid data quality verification device may also include input and output devices, network access devices, buses, etc.
[0133] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the power grid data quality check device, and uses various interfaces and lines to connect various parts of the entire power grid data quality check device.
[0134] The memory can be used to store the computer program and / or module, and the processor implements various functions of the power grid data quality verification device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.
[0135] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMediaCard, SMC), a secure digital (SecureDigital, SD) card, a flash card (FlashCard), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0136] Example 4
[0137] This embodiment 4 also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the power grid data quality verification method are implemented.
[0138] If the module / unit integrated in the power grid data quality verification device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
[0139] Based on this understanding, the present embodiment 4 implements all or part of the processes in the above-mentioned power grid data quality verification method, and can also be completed by instructing related hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned power grid data quality verification method can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.
[0140] The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.
[0141] It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media does not include electrical carrier signals and telecommunication signals.
[0142] The above embodiment is only one of the implementation methods that can realize the technical solution of the present invention. The scope of protection claimed by the present invention is not limited only to this embodiment, but also includes changes, replacements and other implementation methods that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention.
Claims
1. A method for checking the quality of power grid data, characterized in that: include: Based on the power grid topology relationship, determine the equipment to be analyzed; wherein the equipment to be analyzed includes an upstream key node device and a number of downstream devices, and there is a power grid topology relationship between the several downstream devices and the upstream key node device; Obtain real-time telemetry data from the equipment to be analyzed; Determine whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model; if so, output the power grid data quality verification result; if not, use the preset judgment criteria to perform data outlier judgment on the real-time telemetry data of the device to be analyzed, and output the power grid data quality verification result; The construction process of the pre-built linear regression function model is as follows: Obtain historical measurement data of the equipment to be analyzed; wherein the historical measurement data of the equipment to be analyzed includes historical telesignaling data of upstream key node equipment, historical telemetry data of upstream key node equipment, and historical telemetry data of several downstream equipment; Based on the power grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed, a historical measurement data verification matrix with the time sequence of the acquisition points is generated; Perform consistency check on the data in the to-be-checked matrix of historical measurement data with the time sequence of acquisition points to obtain a consistency check result matrix; According to the consistency check result matrix, a check matrix of historical telemetry data with natural day time series is generated; Perform normal distribution calibration on the data in the to-be-calibrated matrix of historical telemetry data with natural daily time series to obtain a normal distribution calibration result matrix; According to the normal distribution calibration result matrix, a linear regression function model is constructed to obtain a pre-constructed linear regression function model; The process of constructing a linear regression function model according to the normal distribution calibration result matrix and obtaining a pre-constructed linear regression function model is as follows: Complete the missing data in the normal distribution calibration result matrix to obtain the equipment effective data matrix; Taking the historical telemetry data of upstream key node equipment in the equipment effective data matrix as the dependent variable, and the historical telemetry data of several downstream equipment in the equipment effective data matrix as the independent variable, a linear regression function is constructed for the same collection point on different natural days to obtain a pre-constructed linear regression function model.
2. A method for checking the quality of power grid data according to claim 1, characterized in that: The historical measurement data to be verified matrix with the time sequence of the acquisition points includes a plurality of historical measurement data columns, and the plurality of historical measurement data columns are arranged according to the time sequence of the acquisition points; Each historical measurement data column includes a historical telesignaling data row of an upstream key node device, a historical telemetry data row of an upstream key node device, and a historical telemetry data row of several downstream devices.
3. A method for checking the quality of power grid data according to claim 2, characterized in that: The process of performing consistency check on the data in the historical measurement data to be checked matrix with the time sequence of the acquisition points to obtain the consistency check result matrix comprises: executing a preset check criterion on each column of the historical measurement data to be checked matrix with the time sequence of the acquisition points to obtain the consistency check result matrix; Among them, the preset verification criteria are as follows: When the data in the historical telemetering data row of the upstream key node device is 0, determine whether more than 1 / 3 of the data in the historical telemetering data rows of several downstream devices are not 0; if so, modify the data that is 0 in the historical telemetering data row of the upstream key node device to 1; if not, modify the data that is not 0 in the historical telemetering data rows of several downstream devices to 0; When the data in the historical telemetering data row of the upstream key node device is 1, determine whether the data in the historical telemetry data row of the upstream key node device is greater than the sum of all data in the historical telemetry data rows of several downstream devices; if not, all data in the historical measurement data column will be eliminated.
4. A method for checking the quality of power grid data according to claim 2, characterized in that: The historical telemetry data to be verified matrix with natural day time series includes several historical telemetry data columns, and the several historical telemetry data columns are arranged according to the natural day time series; wherein each historical telemetry data column includes historical telemetry data rows of upstream key node equipment and historical telemetry data rows of several downstream equipment.
5. A method for checking the quality of power grid data according to claim 1, characterized in that: The default judgment criterion is the Dixon criterion.
6. A power grid data quality verification system, characterized in that: include: The device determination module is used to determine the device to be analyzed based on the power grid topology relationship; wherein the device to be analyzed includes an upstream key node device and a plurality of downstream devices, and there is a power grid topology relationship between the plurality of downstream devices and the upstream key node device; A real-time data acquisition module is used to obtain real-time telemetry data of the equipment to be analyzed; The data quality check module is used to determine whether the real-time telemetry data of the device to be analyzed meets the constraints of the pre-built linear regression function model; if so, the power grid data quality check result is output; if not, the real-time telemetry data of the device to be analyzed is judged as outliers by using the preset judgment criteria, and the power grid data quality check result is output; The construction process of the pre-built linear regression function model is as follows: Obtain historical measurement data of the equipment to be analyzed; wherein the historical measurement data of the equipment to be analyzed includes historical telesignaling data of upstream key node equipment, historical telemetry data of upstream key node equipment, and historical telemetry data of several downstream equipment; Based on the power grid topology relationship between the devices to be analyzed and the historical measurement data of the devices to be analyzed, a historical measurement data verification matrix with the time sequence of the acquisition points is generated; Perform consistency check on the data in the to-be-checked matrix of historical measurement data with the time sequence of acquisition points to obtain a consistency check result matrix; According to the consistency check result matrix, a check matrix of historical telemetry data with natural day time series is generated; Perform normal distribution calibration on the data in the to-be-calibrated matrix of historical telemetry data with natural daily time series to obtain a normal distribution calibration result matrix; According to the normal distribution calibration result matrix, a linear regression function model is constructed to obtain a pre-constructed linear regression function model; The process of constructing a linear regression function model according to the normal distribution calibration result matrix and obtaining a pre-constructed linear regression function model is as follows: Complete the missing data in the normal distribution calibration result matrix to obtain the equipment effective data matrix; Taking the historical telemetry data of upstream key node equipment in the equipment effective data matrix as the dependent variable, and the historical telemetry data of several downstream equipment in the equipment effective data matrix as the independent variable, a linear regression function is constructed for the same collection point on different natural days to obtain a pre-constructed linear regression function model.
7. A power grid data quality verification device, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the power grid data quality verification method as described in any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the power grid data quality verification method according to any one of claims 1 to 5 are implemented.
