A Power Grid Data Analysis and Management System

Through the power grid data analysis and management system, a variety of data processing and analysis modules are used to solve the challenges of data management and analysis in the smart grid, and the accurate judgment of the operating status of the power grid is achieved, the failure risk is reduced, and the data processing efficiency is improved. Security.

CN113869664BActive Publication Date: 2024-11-12STATE GRID ANHUI ELECTRIC POWER CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111050949.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-09-08
Publication Date
2024-11-12
Estimated Expiration
2041-09-08

AI Technical Summary

Technical Problem

The existing technology cannot effectively manage and analyze the power grid data in the smart grid, resulting in the inability to accurately judge the operating status of the power grid, increasing the risk of failure.

Method used

A power grid data analysis management system was designed, through data mining modules, data marking modules, data processing modules, data distribution modules, data modeling modules, historical data analysis modules, distribution boundary map construction modules and feature distribution map construction modules. , using data mining model, source coding, K nearest neighbor algorithm, random matrix modeling and parallelized k-means algorithm to conduct directional mining, correction, classification, modeling and analysis of power grid data, and construct distribution boundary maps and feature distribution maps , realize accurate judgment of the operating status of the power grid.

Benefits of technology

It effectively improves the accuracy and perfection of power grid data, can accurately analyze the operating status of the power grid, reduces the risk of failure, and improves the efficiency and safety of data processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN113869664B_ABST
    Figure CN113869664B_ABST
Patent Text Reader

Abstract

The present invention relates to data analysis management, and in particular to a power grid data analysis management system, comprising a server, wherein the server performs directional mining of power grid data based on a data mining model through a data mining module, the server sets different type labels for different types of power grid data based on source coding through a data marking module, the server performs data correction on different types of power grid data using a data processing module, and the server distributes the corrected power grid data to corresponding data transmission channels based on the type labels using a data distribution module, and stores the corrected power grid data in different data storage nodes in a data storage module. The technical solution provided by the present invention can effectively overcome the defects of the prior art that the power grid data cannot be effectively improved and the power grid operation status cannot be accurately analyzed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to data analysis management, and in particular to a power grid data analysis management system. Background Art

[0002] Realizing information-driven intelligent control of large power grids is of great practical significance for the safe and stable operation of power grids. At present, energy security issues are prominent and environmental pollution problems are severe in countries around the world. Vigorously developing new energy such as wind power and photovoltaics and realizing the transformation of energy production to renewable energy are major needs for the sustainable development of global energy and economy. Due to the continuous grid connection and operation of renewable energy, a large number of power electronic equipment and their control devices are connected to the power system, making the power system develop towards a high proportion of renewable energy and a high proportion of electronics. The new power system will show the characteristics of multi-time scale interaction, strong nonlinearity, more complex structure, and higher order of state variables. These characteristics bring severe challenges to the stability analysis and control of the new generation of power systems.

[0003] With the development of smart grids, the power quality provided by traditional grids can no longer meet the needs of users. According to statistics of grid failures, about 80% of grid failures come from distribution networks. With the access of new energy nodes, voltage fluctuations and other factors in active distribution networks will increase the risk of failures. Different from the data management and status analysis of traditional grids, smart grids need to collect a large amount of related data and perform real-time data query and analysis on the related data. Traditional data storage methods and data analysis methods are no longer applicable to big data analysis of smart grids. Summary of the invention

[0004] 1. Technical issues to be solved

[0005] In view of the above-mentioned shortcomings of the prior art, the present invention provides a power grid data analysis and management system, which can effectively overcome the defects of the prior art that the power grid data cannot be effectively improved and the power grid operation status cannot be accurately analyzed.

[0006] (II) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] A power grid data analysis and management system, comprising a server, wherein the server performs directional mining of power grid data based on a data mining model through a data mining module, wherein the server sets different type labels for different types of power grid data based on source coding through a data marking module, wherein the server performs data correction on different types of power grid data using a data processing module, wherein the server distributes the corrected power grid data to corresponding data transmission channels based on the type labels using a data distribution module, and stores the corrected power grid data in different data storage nodes in a data storage module;

[0009] The server models the power grid synchronization data using a random matrix through a data modeling module, the server analyzes and processes the recorded and stored historical power grid synchronization data through a historical data analysis module, the server constructs a distribution limit map about the characteristic values ​​of the historical power grid synchronization data based on the analysis and processing results through a distribution limit map construction module, the server constructs a characteristic distribution map about the characteristic values ​​of the real-time power grid synchronization data through a characteristic distribution map construction module, the server compares and analyzes the characteristic distribution map and the distribution limit map through a comparative analysis module, and determines the current power grid operation status based on the analysis results.

[0010] Preferably, the data processing module screens the missing data in various types of power grid data, and calculates the replacement data corresponding to the missing data based on a K-nearest neighbor algorithm in combination with K power grid data near the missing data.

[0011] Preferably, the data processing module selects N marker data closest to the missing data, and respectively traverses from near to far starting from the marker data, and records the minimum value a in each traversal data. 1 and the average value a 2 , use the K nearest neighbor algorithm twice for each missing data, K is a 1 、a 2 , and determine whether to proceed with the estimation process based on the calculation results.

[0012] Preferably, after the data processing module uses the K nearest neighbor algorithm twice on the missing data, if the results obtained are consistent, the results are used as replacement data; otherwise, the estimation process is performed.

[0013] Preferably, the data processing module performs an estimation process, including:

[0014] Through a 2 =a 2 -x to the average value a 2 Perform assignment operation, and then take K as a 1 、a 2 Use the 2nd K nearest neighbor algorithm. If the results are consistent, the results are used as replacement data.

[0015] If the results are inconsistent, 1 =a 1 +y for minimum value a 1 Perform assignment operation, and then take K as a 1 、a 2 Use the 2nd K nearest neighbor algorithm. If the results are consistent, the results are used as replacement data.

[0016] If the results are still inconsistent, repeat the above process until the same result is obtained;

[0017] Where x = (a 2 -a 1 ) / N, y=x-1, and x, y are both integers not less than 0.

[0018] Preferably, the data distribution module adopts SparkR's parallelized k-means algorithm to classify the power grid data according to type labels, and matches the corresponding data transmission channel for the power grid data based on the classification results.

[0019] Preferably, it also includes a data association module, a permission allocation module and a data calling module, and the server searches for a suitable location for the power grid data in the data storage node through the data allocation module and performs data association on the power grid data;

[0020] The server sets operation permissions for different types of power grid data based on type tags through a permission allocation module;

[0021] The server verifies the data operation performed by the user through the data calling module, executes the verified data operation based on the calling conditions input by the user, and returns the calling data to the user.

[0022] Preferably, the historical data analysis module divides the recorded and stored historical power grid synchronization data into multiple time period data on average and performs normalization processing;

[0023] The distribution limit diagram construction module calculates the characteristic value distribution function of the historical power grid synchronization data, obtains the upper distribution limit function and the lower distribution limit function of the characteristic value distribution function, and establishes a distribution limit diagram about the characteristic value of the historical power grid synchronization data.

[0024] Preferably, the characteristic distribution graph building module calculates a characteristic value distribution function of the real-time power grid synchronization data, and builds a characteristic distribution graph about the characteristic values ​​of the real-time power grid synchronization data.

[0025] Preferably, the comparative analysis module merges the distribution boundary map and the characteristic distribution map according to a unified standard. If the characteristic distribution map is located inside the distribution boundary map, the operation state of the power grid is stable; otherwise, the operation state of the power grid is abnormal.

[0026] (III) Beneficial effects

[0027] Compared with the prior art, the present invention provides a power grid data analysis and management system, in which a data processing module can screen missing data in various types of power grid data, and calculate the replacement data corresponding to the missing data based on the K nearest neighbor algorithm combined with K power grid data near the missing data, and can fill in the missing data with replacement data, effectively improving the accuracy and completeness of mined power grid data; a distribution boundary map construction module establishes a distribution boundary map for the characteristic values ​​of historical power grid synchronization data, and a characteristic distribution map construction module establishes a characteristic distribution map for the characteristic values ​​of real-time power grid synchronization data, and through the combined comparison of the distribution boundary map and the characteristic distribution map, an accurate analysis of the power grid operation status is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0029] Figure 1 It is a system schematic diagram of the present invention;

[0030] Figure 2 It is a schematic diagram of the process of analyzing the operation status of the power grid in the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] A power grid data analysis and management system, such as Figure 1 and Figure 2 As shown, it includes a server, which performs targeted mining of power grid data based on a data mining model through a data mining module, sets different type labels for different types of power grid data based on source coding through a data marking module, and corrects different types of power grid data using a data processing module. The server uses a data allocation module to allocate the corrected power grid data to corresponding data transmission channels based on the type label, and stores them in different data storage nodes in the data storage module.

[0033] The technical solution of the present application also includes a data association module, a permission allocation module and a data call module. The server uses the data allocation module to find a suitable location for the power grid data in the data storage node and performs data association on the power grid data;

[0034] The server sets operation permissions for different types of power grid data based on type tags through the permission allocation module;

[0035] The server verifies the data operation performed by the user through the data calling module, executes the verified data operation based on the calling conditions input by the user, and returns the calling data to the user.

[0036] In the technical solution of this application, different type labels are set for different types of power grid data based on source coding, and the corrected power grid data is allocated to the corresponding data transmission channel based on the type label, and stored in different data storage nodes in the data storage module. At the same time, a suitable location is found for the power grid data in the data storage node, and the power grid data is associated with data, thereby realizing optimized storage of the power grid data. In addition, the setting of the authority allocation module improves the security of users calling power grid data through the data calling module.

[0037] The data processing module screens the missing data in various types of power grid data, and calculates the replacement data corresponding to the missing data based on the K nearest neighbor algorithm combined with the K power grid data near the missing data. The following is a detailed description of the process of the data processing module calculating the missing replacement data:

[0038] 1) The data processing module selects the N marker data closest to the missing data, and traverses from near to far starting from the marker data, records the minimum value a1 and the average value a2 in each traversal data, uses the K nearest neighbor algorithm twice for each missing data, K takes a1 and a2 respectively, and determines whether to perform the estimation process based on the calculation results;

[0039] 2) After the data processing module uses the K nearest neighbor algorithm twice on the missing data, if the results are consistent, then the result is the replacement data; otherwise, the estimation process is carried out;

[0040] 3) The data processing module performs the estimation process, including:

[0041] Through a 2 =a 2 -x to the average value a 2 Perform assignment operation, and then take K as a 1 、a 2 Use the 2nd K nearest neighbor algorithm. If the results are consistent, the results are used as replacement data.

[0042] If the results are inconsistent, 1 =a1 +y for minimum value a 1 Perform assignment operation, and then take K as a 1 、a 2 Use the 2nd K nearest neighbor algorithm. If the results are consistent, the results are used as replacement data.

[0043] If the results are still inconsistent, repeat the above process until the same result is obtained;

[0044] Where x = (a 2 -a 1 ) / N, y=x-1, and x, y are both integers not less than 0.

[0045] In the technical solution of the present application, the value of K is gradually adjusted in the above manner until the final result is obtained, which can largely ensure the accuracy of the replacement data and is of great value for large-scale data processing. At the same time, the adjustment range is from large to small, which also reduces the number of calculations and takes into account the calculation efficiency.

[0046] The data allocation module uses SparkR's parallelized k-means algorithm to classify power grid data according to type labels, and matches the corresponding data transmission channels for the power grid data based on the classification results.

[0047] The server uses a random matrix to model the power grid synchronization data through the data modeling module, and analyzes and processes the recorded and stored historical power grid synchronization data through the historical data analysis module. The server constructs a distribution limit map about the characteristic values ​​of the historical power grid synchronization data based on the analysis and processing results through the distribution limit map construction module. The server constructs a feature distribution map about the characteristic values ​​of the real-time power grid synchronization data through the feature distribution map construction module. The server compares and analyzes the feature distribution map and the distribution limit map through the comparative analysis module, and judges the current power grid operation status based on the analysis results.

[0048] The data modeling module establishes the relationship between the observed value of the grid synchronization data and the grid operation status according to the grid power flow equation, that is, the random matrix representation of the grid synchronization data. According to the operation status of the grid, the grid can be divided into four types: topology unchanged, topology changed, power supply unchanged, and power supply changed.

[0049] The historical data analysis module divides the recorded and stored historical power grid synchronization data into multiple time period data on average and performs normalization processing.

[0050] The distribution limit diagram construction module calculates the characteristic value distribution function of the historical power grid synchronization data, obtains the upper distribution limit function and the lower distribution limit function of the characteristic value distribution function, and establishes the distribution limit diagram of the characteristic value of the historical power grid synchronization data.

[0051] The characteristic distribution graph building module calculates the characteristic value distribution function of the real-time power grid synchronization data and builds a characteristic distribution graph about the characteristic value of the real-time power grid synchronization data.

[0052] The comparative analysis module merges the distribution boundary map and the characteristic distribution map according to a unified standard. If the characteristic distribution map is located inside the distribution boundary map, the grid operation state is stable; otherwise, the grid operation state is abnormal.

[0053] In the technical solution of the present application, the distribution limit map construction module establishes a distribution limit map for the characteristic values ​​of historical power grid synchronization data, and the characteristic distribution map construction module establishes a characteristic distribution map for the characteristic values ​​of real-time power grid synchronization data. By combining and comparing the distribution limit map and the characteristic distribution map, accurate analysis of the power grid operation status is achieved.

[0054] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will 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 power grid data analysis and management system, characterized in that: The server comprises a server, wherein the server performs directional mining of power grid data based on a data mining model through a data mining module, the server sets different type labels for different types of power grid data based on source coding through a data marking module, the server performs data correction on different types of power grid data using a data processing module, and the server distributes the corrected power grid data to corresponding data transmission channels based on the type labels using a data distribution module, and stores the corrected power grid data in different data storage nodes in a data storage module; The server models the power grid synchronization data using a random matrix through a data modeling module, analyzes and processes the recorded and stored historical power grid synchronization data through a historical data analysis module, constructs a distribution limit map about the characteristic values ​​of the historical power grid synchronization data based on the analysis and processing results through a distribution limit map construction module, constructs a characteristic distribution map about the characteristic values ​​of the real-time power grid synchronization data through a characteristic distribution map construction module, and compares and analyzes the characteristic distribution map and the distribution limit map through a comparative analysis module, and determines the current power grid operation status based on the analysis results; The data processing module screens the missing data in various types of power grid data, and calculates the replacement data corresponding to the missing data based on the K nearest neighbor algorithm combined with K power grid data near the missing data; The data processing module selects N marker data closest to the missing data, and traverses from near to far starting from the marker data, records the minimum value a1 and the average value a2 in each traversal data, uses the K nearest neighbor algorithm twice for each missing data, K takes a1 and a2 respectively, and determines whether to perform the estimation process according to the calculation results; If the data processing module uses the K nearest neighbor algorithm twice on the missing data and the results are consistent, then the result is the replacement data, otherwise the estimation process is performed; The data processing module performs an estimation process, including: The average value a2 is assigned by a2=a2-x, and then K is respectively taken as a1 and a2 to use the K nearest neighbor algorithm twice. If the results are consistent, the result is the replacement data; If the results are inconsistent, the minimum value a1 is assigned by a1=a1+y, and then K is taken as a1 and a2 respectively and the K nearest neighbor algorithm is used twice. If the results are consistent, the result is the replacement data; If the results are still inconsistent, repeat the above process until the same result is obtained; Wherein, x=(a2-a1) / N, y=x-1, and both x and y are integers not less than 0; The data distribution module adopts SparkR's parallelized k-means algorithm to classify the power grid data according to the type label, and matches the corresponding data transmission channel for the power grid data based on the classification result; The historical data analysis module divides the recorded and stored historical power grid synchronization data into multiple time period data on average and performs normalization processing; The distribution limit diagram construction module calculates the characteristic value distribution function of the historical power grid synchronization data, obtains the upper distribution limit function and the lower distribution limit function of the characteristic value distribution function, and establishes a distribution limit diagram of the characteristic value of the historical power grid synchronization data; The characteristic distribution map building module calculates the characteristic value distribution function of the real-time power grid synchronization data and builds a characteristic distribution map about the characteristic value of the real-time power grid synchronization data; The comparative analysis module merges the distribution boundary map and the characteristic distribution map according to a unified standard. If the characteristic distribution map is located inside the distribution boundary map, the operation state of the power grid is stable; otherwise, the operation state of the power grid is abnormal. The power grid data analysis and management system further comprises a data association module, a permission allocation module and a data calling module. The server searches for a suitable location for the power grid data in the data storage node through the data allocation module and performs data association on the power grid data. The server sets operation permissions for different types of power grid data based on type tags through a permission allocation module; The server verifies the data operation performed by the user through the data calling module, executes the verified data operation based on the calling conditions input by the user, and returns the calling data to the user.