CN113568898A - Electric power data leakage point completion method, device, equipment and readable storage medium - Google Patents
Electric power data leakage point completion method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN113568898A CN113568898A CN202110873162.6A CN202110873162A CN113568898A CN 113568898 A CN113568898 A CN 113568898A CN 202110873162 A CN202110873162 A CN 202110873162A CN 113568898 A CN113568898 A CN 113568898A
- Authority
- CN
- China
- Prior art keywords
- data
- historical
- window
- power data
- leakage point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 230000001502 supplementing effect Effects 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 abstract description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 5
- 238000005070 sampling Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000009469 supplementation Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for supplementing leakage points of electric power data, which comprises traversing the electric power data to be supplemented, and inquiring the position of the leakage point data; selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window; similarity calculation is carried out on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively, and a historical data window corresponding to the highest similarity is obtained; and according to the historical power data in the historical data window corresponding to the highest similarity, completing the power data at the corresponding position of the leakage point data. According to the method and the device, the data of the leakage points determined according to the approximate historical electric power data are more accurate and effective, the accurate reliability of electric power information analysis according to the electric power data after completion is guaranteed, and the reliable operation of the intelligent power grid is promoted. The application also provides a device, equipment and computer readable storage medium for completing the electric power data leakage points, and the device, equipment and computer readable storage medium have the beneficial effects.
Description
Technical Field
The present invention relates to the field of power data processing technologies, and in particular, to a method, an apparatus, a device, and a computer-readable storage medium for completing a power data leakage point.
Background
With the comprehensive development of intelligent power grid construction in China, big electric power data are generated at the same time and become an indispensable part in the intelligent process of a power grid. The basis for supporting the reliable operation of the smart power grid is the panoramic real-time data acquisition, transmission and storage of the power grid and the analysis of accumulated massive multi-source power data. In the face of various large-size structured, semi-structured and unstructured power data, loss in different degrees is usually inevitable, and the stable and reliable power data can support upper-layer application based on large measurement data, so that the existence of the loss data not only affects an analysis result, but also affects effective development of subsequent work of workers. Therefore, it is a significant and troublesome task to effectively complement the measured data.
The conventional missing data completion method is an interpolation method, and the interpolation method comprises simple filling, mean filling, random filling, regression model filling, nearest neighbor filling, a random forest filling method and the like; however, any method for filling missing data by completion has certain limitations, so that reliability of the supplemented power data is poor, and reliable operation of the smart grid is affected.
Disclosure of Invention
The invention aims to provide a method, a device and equipment for supplementing leakage points of electric power data and computer readable storage equipment, which are beneficial to improving the reliability of the electric power data after being supplemented and improving the reliability of only power grid operation.
In order to solve the above technical problem, the present invention provides a method for completing a leakage point of power data, including:
traversing the electric power data to be compensated, and inquiring the position of the data with the leakage point;
selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window;
similarity calculation is carried out on the historical power data in each historical data window and adjacent data corresponding to the leakage point data respectively, and a historical data window corresponding to the highest similarity is obtained; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and completing the power data of the corresponding position of the leakage point data according to the historical power data in the historical data window corresponding to the highest similarity.
Optionally, performing similarity calculation on the historical power data in each historical data window and the neighboring data corresponding to the leakage point data respectively includes:
respectively calculating a first average value, a first standard deviation and a first forward differential inclination of the historical power data in each historical data window;
calculating a second mean, a second standard deviation, and a second forward differential inclination of the neighborhood data;
according to the similarity
Calculating the similarity of the proximity data and the historical power data in each historical data window; wherein, mui、σi、
The first average, the first standard deviation and the first forward differential inclination respectively correspond to the historical power data in an ith historical data window; mu.s0、σ0、
The second mean, the second standard deviation, and the second forward differential inclination, respectively; alpha, beta and gamma are respectively a first weight value, a second weight value and a third weight value; i is a positive integer.
Optionally, after selecting historical power data in a plurality of historical data windows in a historical power database by a preset window size, before performing similarity calculation on the historical power data in each historical data window, the method further includes:
removing a historical data window corresponding to the historical power data missing at the head and tail positions;
the process of calculating the first forward differential inclination comprises:
according to the formula of forward differential inclination
Calculating the first forward differential inclination corresponding to each historical data window; wherein,
is a forward differential inclination; lambda and eta are the first historical power data and the last historical power data in the same historical data window respectively.
Optionally, after selecting historical power data in a plurality of historical data windows in a historical power database by a preset window size, before performing similarity calculation on the historical power data in each historical data window, the method further includes:
and eliminating a window with the deletion rate of the corresponding historical power data being larger than a preset deletion rate threshold.
Optionally, the process of determining the preset window size includes:
and according to the number N of continuous missing points corresponding to the position of the current missing point data, setting the size of the preset window and the size of the continuous missing point data N in positive correlation, wherein the size of the preset window is not less than 2N +1, and N is a positive integer.
Optionally, when the number of the power data in the window corresponding to the preset window size is 2M and the number of the continuous leakage points is 2Q +1, the determining the adjacent data process includes:
determining first adjacent data corresponding to a first window and second adjacent data corresponding to a second window according to the preset window size; the Q +1 th leakage point power data in the power data of the continuous leakage points are located at the Mth position point of the first window and at the M +1 th position point of the second window; wherein M is a positive integer greater than 3, and Q is a natural number;
correspondingly, according to the historical power data in the historical data window corresponding to the highest similarity, the power data of the corresponding position of the leakage point data is supplemented, and the method comprises the following steps:
determining first completion data according to historical power data in a historical data window with the highest similarity corresponding to the first adjacent data;
determining second completion data according to historical power data in a historical data window with the highest similarity corresponding to the second adjacent data;
and completing the electric power data of the position corresponding to the leakage point data by using the average value of the first completion data and the second completion data.
Optionally, after determining the number of consecutive missing points corresponding to the position where the missing point data currently exists, the method further includes:
and when the number of the continuous leakage points is greater than a preset number threshold value, sending an alarm.
The application also provides a device is mended to electric power data leak source, includes:
the leakage point query module is used for traversing the to-be-compensated electric power data and querying the position of the leakage point data;
the historical data module is used for selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window;
the similarity operation module is used for performing similarity operation on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively and obtaining a historical data window corresponding to the highest similarity; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and the missing point completion module is used for completing the electric power data of the corresponding position of the missing point data according to the historical electric power data in the historical data window corresponding to the highest similarity.
The application also provides a power data leak source completion equipment, includes:
a memory for storing a computer program;
a processor for implementing the steps of the power data leakage point completing method as described in any one of the above when the computer program is executed.
The present application further provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the power data leakage point completion method as described in any one of the above.
The electric power data leakage point completion method provided by the invention comprises traversing electric power data to be completed and inquiring the position of the leakage point data; selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window; similarity calculation is carried out on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively, and a window corresponding to the highest similarity is obtained; the adjacent data is power data except the leakage point data in a window with the size of a preset window by taking the leakage point data as a center; and according to the historical power data in the historical data window corresponding to the highest similarity, completing the power data at the corresponding position of the leakage point data.
When the leakage point data with data missing exist in the collected electric power data are supplemented, certain regularity and certain similarity exist according to the electric power data, a group of historical electric power data which are most similar to the electric power data adjacent to the leakage point data are determined in the historical electric power database and serve as a basis source of the supplemented leakage point data, and compared with the method that the supplemented leakage point data are estimated only according to the adjacent data of the missing data in an interpolation method, the leakage point data determined according to the similar historical electric power data are more accurate and effective, the accuracy and the reliability of electric power information analysis according to the supplemented electric power data are guaranteed, and the reliability operation of the intelligent power grid is promoted.
The application also provides a device, equipment and computer readable storage medium for completing the electric power data leakage points, and the device, equipment and computer readable storage medium have the beneficial effects.
Drawings
In order to more clearly illustrate the embodiments or technical solutions of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a power data leakage point completion method according to an embodiment of the present disclosure;
fig. 2 is a schematic diagram of a framework of a power data leakage point completion apparatus according to an embodiment of the present disclosure.
Detailed Description
The power data is typical time series data, and data sampling is performed at fixed time intervals. However, in practical applications, due to a failure of a sampling device or a failure of a data storage device, a situation that part of data is missing is likely to occur, and the missing data is also missing point data.
Interpolation method for complementing electric power data is a commonly used method for complementing leakage point data at present, and the basic principle is to estimate the leakage point data according to the non-missing electric power data adjacent to the leakage point data, and finally set a relatively reasonable leakage point data. However, the missing point data complementing method is more accurate for complementing the missing point data with a small amount of continuously missing data, and once more continuously missing data exists, it is difficult to ensure the accuracy of complementing the data.
For this reason, in this application, it is considered that there is generally a certain regularity and repeatability in using electric devices by users, for example, an air conditioner is always turned on during a time period when a user goes home from work in summer, and power consumption is increased when a factory is always turned on. Accordingly, the power data for the power consumption situation with a certain regularity also necessarily has a certain regularity, for example, the power consumption data of the user collected at the off-duty time point in summer has a very strong approximation. Then a section of the user power utilization curve can be found in the historical power utilization curve and is basically consistent with the section of the power utilization curve needing to be completed.
Based on the principle, the data completion can be realized according to the historical power utilization curve with higher approximation degree when the missing data is completed, and the accuracy and the reliability of the data completion are ensured.
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, fig. 1 is a schematic flow chart of a power data leakage point completing method provided in the embodiment of the present application, where the method may include:
s11: and traversing the electric power data to be compensated, and inquiring the position of the data with the leakage point.
As with the conventional power data, the power data to be compensated in the present embodiment is also time series data collected at fixed time intervals. When the data of the leakage points is traversed, the data of the leakage points can be sequentially traversed according to the time point sequence of the acquisition of the power data, and once the position of the data of the leakage points is found, the data completion is carried out on the position of the data of the leakage points.
For example, a group of power data obtained by collecting is A1, A2, A3, A4, A5, A6, A7, A8, A9, A10, A11 and A12. Wherein. If the A4, the A7, the A8 and the A9 are all missing, two missing leakage point data position points exist in the group of power data, the missing leakage point data of one position point is A4, and the number of the continuous missing leakage point data is 1; the missing data of the second position is A7, A8 and A9, and the number of the missing data of the second position is 3.
S12: and selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window.
The preset window size in this embodiment may be regarded as a time period window, and the preset window size may be determined based on the number of missing point data, where the larger the number of missing point data is, the larger the preset window size is. At least the preset window size should be not less than 2N +1, where N is a positive integer representing the number of missing data.
When selecting the historical power data in the plurality of historical data windows, the historical power data may be selected by sliding the windows, for example, there is a group of historical power data of a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, a11, and a 12.
The preset window size is just equal to the duration of continuously acquiring 5 power sizes, so that 8 groups of historical power data corresponding to 8 historical data windows can be determined, wherein the historical power data are as follows:
the first window: a1, a2, a3, a4, a 5;
the second window: a2, a3, a4, a5, a 6;
the third window: a3, a4, a5, a6, a 7;
the fourth window: a4, a5, a6, a7, a 8;
the fifth window: a5, a6, a7, a8, a 9;
a sixth window: a6, a7, a8, a9, a 10;
a seventh window: a7, a8, a9, a10, a 11;
the eighth window: a8, a9, a10, a11 and a 12.
Because the historical power data of each historical data window is to be taken as a basis for complementing missing leakage point data, the reliability of the historical power data in each historical data window is relatively important. The historical power data in this embodiment is the original power data actually collected before that, and obviously, there is inevitable data missing therein. Therefore, in an optional embodiment of the present application, a historical data window in which the corresponding historical power data is missing may be further removed. For example, a preset loss rate threshold may be set, and when the loss rate of the corresponding historical power data in the historical data window is greater than the preset loss rate threshold, the historical data window may be directly removed.
For example, the preset missing rate threshold may be set to 30%, if exactly three historical data of a6, a7, and a8 are missing in the above 8 historical data windows, obviously, the missing rate of the historical data in the fourth window, the fifth window, and the sixth window is 60%; the historical data missing rate of the third window and the seventh window is 40%; the historical data missing rate of the second window and the historical data missing rate of the eighth window are both 20%, and the historical data missing rate of the first window is both 0; thus, only the first window, the second window, and the eighth window may be retained.
Of course, in the actual application process, the historical missing threshold may also be directly set to 0, that is, all historical data windows in which historical power data exist are removed, so as to further ensure the accuracy and reliability of the historical power data as a sample for completing the missing power data.
In addition, when the amount of the continuously missing leakage point data is too large, it is obvious that the corresponding preset window size also needs to be set to be a larger historical data window, but in reality, when the amount of the continuously missing leakage point data is too large, no matter how the data completion is performed, the power data after completion is unreliable, so that in practical application, if the amount of the continuously missing power data is too large, the preset window size is too large, an alarm can be directly sent out.
In addition, as described above, the larger the number of consecutive missing dot data is, the larger the corresponding preset window size needs to be set. In order to simplify the operation, all the electric power data to be supplemented can be traversed, the same preset window size is adopted for the leakage point positions with the same quantity of continuous leakage point data, the historical electric power data of a plurality of historical data windows are obtained, and the historical data window with the highest similarity corresponding to each leakage point position is determined in the plurality of historical data windows corresponding to the same preset window size, so that the electric power data supplementation of the leakage point positions is realized.
S13: and respectively carrying out similarity calculation on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data, and obtaining the window corresponding to the highest similarity.
The adjacent data is power data except the leakage point data in a window with the size of a preset window by taking the leakage point data as a center.
The specific vicinity data is a plurality of power data that are adjacent to each other before and after the leak data. As described above, for the leakage point data a4, a window of 5 power data including the leakage point data is divided by a preset window size, and the power data a2, A3, a5, a6 in the window except the leakage point data a4 are corresponding adjacent data.
It can be understood that, when the window based on the preset window size divides the adjacent data corresponding to the missing point data, the missing point data should be located at the center of the window as much as possible.
As described above, based on the preset window size, historical power data corresponding to a plurality of historical data windows may be partitioned in the historical power database. Therefore, similarity calculation can be carried out between the historical power data and the adjacent data corresponding to the leakage point data in each historical data window, so that the historical data window with the most approximate change rule of the corresponding historical power data and the adjacent data corresponding to the leakage point data is determined.
S14: and according to the historical power data in the historical data window corresponding to the highest similarity, completing the power data at the corresponding position of the leakage point data.
After the historical data window with the most similar change rule of the corresponding historical power data and the change rule of the adjacent data is determined, the historical power data in the historical data window can be used as sample data for determining the leakage point data, and the completion of the leakage point data is achieved.
There may be a variety of completion methods for the missing dot data. For example, the historical power data can be directly subjected to linear fitting, and the slope of a first linear fitting equation of the historical power data changing along with the acquisition time is determined; and fitting a second linear fitting equation which meets the slope to the adjacent data by taking the slope as a basis, and finally determining the complemented power data corresponding to the missing point data by taking the missing point data which meets the second linear fitting equation as a basis.
For example, the historical power data and the adjacent data of the corresponding position in the historical data window can be calculated, and the change curve of the difference value along with the sampling time can be determined. For example, for the missing point data a4 and the adjacent data a2, A3, a5 and a6, the historical data window corresponding to the historical power data with the highest similarity is the first window, i.e., the window containing a1, a2, A3, a4 and a 5. The sizes of a1-A2, a2-A3, a4-A5 and a5-A6 are respectively determined, 4 groups of differences are fitted with a difference curve changing along with sampling time, and obviously, the difference between A3 and A4 also satisfies the difference curve, so that the size of the missing point data a4 can be determined.
Other similar methods of determining missing dot data may also exist in the present application and are not listed here.
To sum up, there is certain regularity and repeatability based on the user power consumption in this application for this principle of certain regularity and repeatability also exists in corresponding electric power data, cuts apart a section of electric power data that the leak source data is adjacent through the intercepting and regards as adjacent data, and regards this as a set of historical electric power data that the inquiry changes law and adjacent data change law in historical electric power database is the most approximate, and regards this as sample data and determines the completion data that the leak source data corresponds, makes the electric power data after the completion more accurate reliable, is favorable to the reliable effectual operation of smart power grids.
Based on any of the above embodiments, in another optional embodiment of the present application, the process of determining the similarity between the historical power data and the neighboring data corresponding to the leakage point data of each window may include:
s21: and respectively calculating a first average value, a first standard deviation and a first forward differential inclination of each historical power data in each historical data window.
For the first average value and the first standard deviation, the conventional calculation formula for calculating the average value and the standard deviation may be adopted. The average value may reflect a central trend of the power data, such as a level of power consumption of a user per day; the standard deviation may reflect the degree of dispersion of the power data, such as the distribution of power consumption by a user during a day.
Optionally, the calculation process for the first forward differential inclination may include:
according to the formula of forward differential inclination
Calculating a first forward differential inclination corresponding to each historical data window; wherein,
is a forward differential inclination; lambda and eta are the first historical power data and the last historical power data in the same historical data window respectively.
Based on a forward differential inclination formula, it can be known that the calculation of the forward differential inclination can be realized only when the first historical power data and the last historical power data are ensured not to be missing in each historical data window; therefore, before the first forward differential inclination is carried out, historical data windows with missing historical power data at head and tail positions in historical data windows corresponding to various groups of historical power data can be removed.
In this embodiment, the arctangent of the head-to-tail data difference value is used as the forward differential inclination, and the arctangent function has a definite boundary and monotonicity, so that the trend of data can be reflected, a positive value is a rise, and a negative value is a fall, for example, the power consumption of the same user is increased or decreased.
S22: a second mean, a second standard deviation, and a second forward differential inclination of the neighboring data are calculated.
Obviously, the manner of determining the second average value, the second standard deviation and the second forward differential inclination of the neighboring data is the same as the manner of determining the first average value, the first standard deviation and the first forward differential inclination of each window, and detailed description thereof is omitted.
S23: according to the similarity
Calculating the similarity between the adjacent data and the historical power data in each historical data window;
wherein,μi、σi、
respectively corresponding to the historical power data in the ith historical data window, a first average value, a first standard deviation and a first forward difference inclination; mu.s0、σ0、
Respectively a second average value, a second standard deviation and a second forward differential inclination; alpha, beta and gamma are respectively a first weight value, a second weight value and a third weight value; i is a positive integer.
The sum of the three weight values of α, β, and γ may be equal to 1, and the specific magnitude of each weight value may be obtained based on statistical principles and training based on a large amount of historical data, or determined by a worker according to experience, which is not limited in this application.
In the present application, the similarity between the historical power data and the neighboring data is determined based on the comprehensive similarity between three statistical indexes, i.e., the average value, the standard deviation and the forward difference inclination, but in practical applications, the similarity between the historical power data and the neighboring data is not determined based on other statistical indexes, such as the slope of a fitted curve, the difference between the maximum value and the minimum value, and the like.
As described above, when determining the neighboring data of the missing point data by dividing according to the preset window size, the missing point data should be at the center of the window containing the neighboring data, but if the number of the power data in the window corresponding to the preset window size is an even number and the number of the consecutive missing point data is a base number, it is obviously impossible to determine the window with the missing point data at the center. To this end, in an alternative embodiment of the present application,
when the number of the power data in the window corresponding to the preset window size is 2M and the number of the continuous leakage points is 2Q +1, determining the adjacent data process comprises:
determining first adjacent data corresponding to a first window and second adjacent data corresponding to a second window according to the size of a preset window; the Q +1 th leakage point power data in the power data of the continuous leakage points are located at the Mth position point of the first window and at the M +1 th position point of the second window; wherein M is a positive integer greater than 3, and Q is a natural number;
correspondingly, according to the historical power data in the historical data window corresponding to the highest similarity, the power data of the corresponding position of the leakage point data is supplemented, and the method comprises the following steps:
determining first completion data according to historical power data in a historical data window with the highest similarity corresponding to the first adjacent data;
determining second completion data according to historical power data in a historical data window with the highest similarity corresponding to the second adjacent data;
and completing the electric power data of the corresponding position of the leakage point data by using the average value of the first completion data and the second completion data.
Taking the number of continuous leakage points as 1 and the number of power data in a window corresponding to a preset window size as 4 as an example, for power data a1, a2, A3, a4, a5 and a6, if a4 is the leakage point data a4, a2, A3 and a5 are first leakage point smart power data; a3, A5 and A6 are second adjacent data;
the first neighborhood data a2, A3, a5 and the second neighborhood data A3, a5, a6 may respectively determine the historical power data in the historical data window with the highest similarity, accordingly, two sets of completion data may be respectively determined based on the historical power data of the two historical data windows, and the two sets of completion data are averaged to obtain the final completion data of the leakage point data.
It should be noted that, if the continuous missing-point data includes more than 1, taking three as an example, the average value of the completion data at the corresponding position in the two sets of completion data may be directly used as the final completion data of the position point.
Furthermore, in the practical application process, even though only one set of adjacent data is only corresponded to the leakage point data of each position, when the historical power data serving as the sample is determined according to the highest similarity, the historical power data in two or even a plurality of historical data windows can be determined to have the basically same and the maximum similarity, at this time, a set of completion data can be respectively determined according to each set of historical power data, then, a plurality of sets of completion data are averaged, and the averaged operation result is used as the final completion data, so that the completion of the power data is realized.
In the following, the electric power data leakage point completing device provided by the embodiment of the present invention is introduced, and the electric power data leakage point completing device described below and the electric power data leakage point completing method described above may be referred to correspondingly.
Fig. 2 is a block diagram of a power data leakage point completing device according to an embodiment of the present invention, where the power data leakage point completing device shown in fig. 2 may include:
the leakage point query module 100 is configured to traverse the to-be-compensated power data and query a position where the leakage point data exists;
a historical data module 200, configured to select historical power data in multiple historical data windows from a historical power database according to a preset window size;
the similarity operation module 300 is configured to perform similarity operation on the historical power data in each historical data window and the neighboring data corresponding to the leakage point data, and obtain a window corresponding to the highest similarity; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and a missing point completion module 400, configured to complete the power data at the position corresponding to the missing point data according to the historical power data in the historical data window corresponding to the highest similarity.
The power data missing point completion apparatus of this embodiment is used to implement the aforementioned power data missing point completion method, and therefore specific embodiments of the power data missing point completion apparatus may be found in the foregoing embodiments of the power data missing point completion method, for example, the missing point query module 100, the historical data module 200, the similarity operation module 300, and the missing point completion module 400 are respectively used to implement steps S11, S12, S13, and S14 in the aforementioned power data missing point completion method, so that specific embodiments thereof may refer to descriptions of corresponding partial embodiments, and are not repeated herein.
The present application further provides an embodiment of a device for supplementing a leakage point of power data, which may include:
a memory for storing a computer program;
a processor for implementing the steps of the power data leakage point completing method as described in any one of the above when the computer program is executed.
The step of the method for completing the leakage point of the power data pointed by the processor in the embodiment may include:
traversing the electric power data to be compensated, and inquiring the position of the data with the leakage point;
selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window;
similarity calculation is carried out on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively, and a historical data window corresponding to the highest similarity is obtained; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and completing the power data of the corresponding position of the leakage point data according to the historical power data in the historical data window corresponding to the highest similarity.
The present application further provides an embodiment of a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the power data leakage point completion method as described in any one of the above.
The computer-readable storage medium may include Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Furthermore, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include elements inherent in the list. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element. In addition, parts of the above technical solutions provided in the embodiments of the present application, which are consistent with the implementation principles of corresponding technical solutions in the prior art, are not described in detail so as to avoid redundant description.
The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts. It should be noted that, for those skilled in the art, it is possible to make various improvements and modifications to the present invention without departing from the principle of the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
Claims (10)
1. A method for supplementing leakage points of power data is characterized by comprising the following steps:
traversing the electric power data to be compensated, and inquiring the position of the data with the leakage point;
selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window;
similarity calculation is carried out on the historical power data in each historical data window and adjacent data corresponding to the leakage point data respectively, and a historical data window corresponding to the highest similarity is obtained; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and completing the power data of the corresponding position of the leakage point data according to the historical power data in the historical data window corresponding to the highest similarity.
2. The method for completing the leakage points of the power data according to claim 1, wherein performing similarity calculation between the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively comprises:
respectively calculating a first average value, a first standard deviation and a first forward differential inclination of the historical power data in each historical data window;
calculating a second mean, a second standard deviation, and a second forward differential inclination of the neighborhood data;
according to the similarity
Calculating the similarity of the proximity data and the historical power data in each historical data window; wherein, mui、σi、
The first average, the first standard deviation and the first forward differential inclination respectively correspond to the historical power data in an ith historical data window; mu.s0、σ0、
The second mean, the second standard deviation, and the second forward differential inclination, respectively; alpha, beta and gamma are respectively a first weight value, a second weight value and a third weight value; i is a positive integer.
3. The method for supplementing the leakage point of the power data according to claim 2, wherein after selecting the historical power data in a plurality of historical data windows in the historical power database by a preset window size, before performing similarity calculation on the historical power data in each historical data window, the method further comprises:
removing a historical data window corresponding to the historical power data missing at the head and tail positions;
the process of calculating the first forward differential inclination comprises:
according to the formula of forward differential inclination
Calculating the first forward differential inclination corresponding to each historical data window; wherein,
is a forward differential inclination; lambda and eta are the first historical power data and the last historical power data in the same historical data window respectively.
4. The method for supplementing the leakage point of the power data according to claim 1, wherein after selecting the historical power data in a plurality of historical data windows in the historical power database by a preset window size, before performing similarity calculation on the historical power data in each historical data window, the method further comprises:
and eliminating a window with the deletion rate of the corresponding historical power data being larger than a preset deletion rate threshold.
5. The power data leakage point completion method of claim 1, wherein determining the preset window size comprises:
and according to the number N of continuous missing points corresponding to the position of the current missing point data, setting the size of the preset window and the size of the continuous missing point data N in positive correlation, wherein the size of the preset window is not less than 2N +1, and N is a positive integer.
6. The method according to claim 5, wherein when the number of power data in the window corresponding to the preset window size is 2M and the number of consecutive leakage points is 2Q +1, the determining the neighboring data process comprises:
determining first adjacent data corresponding to a first window and second adjacent data corresponding to a second window according to the preset window size; the Q +1 th leakage point power data in the power data of the continuous leakage points are located at the Mth position point of the first window and at the M +1 th position point of the second window; wherein M is a positive integer greater than 3, and Q is a natural number;
correspondingly, according to the historical power data in the historical data window corresponding to the highest similarity, the power data of the corresponding position of the leakage point data is supplemented, and the method comprises the following steps:
determining first completion data according to historical power data in a historical data window with the highest similarity corresponding to the first adjacent data;
determining second completion data according to historical power data in a historical data window with the highest similarity corresponding to the second adjacent data;
and completing the electric power data of the position corresponding to the leakage point data by using the average value of the first completion data and the second completion data.
7. The method for supplementing the leakage points of the power data according to claim 5, wherein after determining the number of the continuous leakage points corresponding to the position where the leakage point data currently exists, the method further comprises:
and when the number of the continuous leakage points is greater than a preset number threshold value, sending an alarm.
8. An electric power data leakage point completion device, comprising:
the leakage point query module is used for traversing the to-be-compensated electric power data and querying the position of the leakage point data;
the historical data module is used for selecting historical power data in a plurality of historical data windows from a historical power database according to the size of a preset window;
the similarity operation module is used for performing similarity operation on the historical power data in each historical data window and the adjacent data corresponding to the leakage point data respectively and obtaining a historical data window corresponding to the highest similarity; the adjacent data is power data except the leakage point data in a window with the size of the preset window by taking the leakage point data as a center;
and the missing point completion module is used for completing the electric power data of the corresponding position of the missing point data according to the historical electric power data in the historical data window corresponding to the highest similarity.
9. An electric power data leakage point completion device, comprising:
a memory for storing a computer program;
a processor for implementing the steps of the power data leakage point completion method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, having stored thereon a computer program which, when being executed by a processor, carries out the steps of the power data leakage point complementing method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873162.6A CN113568898B (en) | 2021-07-30 | 2021-07-30 | Method, device, equipment and readable storage medium for supplementing electric power data leakage points |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873162.6A CN113568898B (en) | 2021-07-30 | 2021-07-30 | Method, device, equipment and readable storage medium for supplementing electric power data leakage points |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113568898A true CN113568898A (en) | 2021-10-29 |
| CN113568898B CN113568898B (en) | 2024-07-09 |
Family
ID=78169543
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110873162.6A Active CN113568898B (en) | 2021-07-30 | 2021-07-30 | Method, device, equipment and readable storage medium for supplementing electric power data leakage points |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113568898B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115329907A (en) * | 2022-10-14 | 2022-11-11 | 杭州致成电子科技有限公司 | Electric load completion method and system based on DBSCAN clustering |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080208888A1 (en) * | 2007-02-28 | 2008-08-28 | Kevin Mitchell | Historical data management |
| CN101976301A (en) * | 2010-10-09 | 2011-02-16 | 广东省电力设计研究院 | Method and device for preprocessing historical data in yearly load forecasting |
| CN104484673A (en) * | 2014-12-05 | 2015-04-01 | 南京大学 | Data complementation method for pattern recognition application of real-time data flow |
| US20170046426A1 (en) * | 2015-08-14 | 2017-02-16 | Nasdaq, Inc. | Computer-implemented systems and methods for intelligently retrieving, analyzing, and synthesizing data from databases |
| US20180012141A1 (en) * | 2016-07-11 | 2018-01-11 | Conduent Business Services, Llc | Method of trip prediction by leveraging trip histories from neighboring users |
| CN108461130A (en) * | 2018-01-18 | 2018-08-28 | 邱堃 | Medical task intelligent dispatching method and system |
| US10102559B1 (en) * | 2014-09-30 | 2018-10-16 | Amazon Technologies, Inc. | Diversification of recommendations |
| US20190067946A1 (en) * | 2017-08-28 | 2019-02-28 | General Electric Company | System and method for distribution load forecasting in a power grid |
| CN110020122A (en) * | 2017-10-16 | 2019-07-16 | Tcl集团股份有限公司 | A kind of video recommendation method, system and computer readable storage medium |
| CN110244099A (en) * | 2019-06-24 | 2019-09-17 | 河南工业大学 | Stealing detection method based on user's voltage |
| CN110502751A (en) * | 2019-08-09 | 2019-11-26 | 国网山西省电力公司 | Bulk power grid operation situation cognitive method, terminal device and storage medium |
| CN110780150A (en) * | 2019-10-21 | 2020-02-11 | 江苏轶一电力科技有限公司 | Transmission line fault positioning device and method based on transmission tower leakage current |
| CN111291280A (en) * | 2020-03-10 | 2020-06-16 | 中国科学院计算技术研究所 | Method, medium, and apparatus for fast predicting trajectory of large-scale moving object |
| CN111309718A (en) * | 2020-02-19 | 2020-06-19 | 南方电网科学研究院有限责任公司 | Distribution network voltage data missing filling method and device |
| CN111507412A (en) * | 2020-04-20 | 2020-08-07 | 南京工程学院 | Voltage missing value filling method based on historical data auxiliary scene analysis |
| CN112162878A (en) * | 2020-09-30 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Database fault discovery method and device, electronic equipment and storage medium |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
| US20210097060A1 (en) * | 2019-10-01 | 2021-04-01 | Institute For Information Industry | Missing data compensation method, missing data compensation system, and non-transitory computer-readable medium |
| US20210167852A1 (en) * | 2018-08-15 | 2021-06-03 | Huawei Technologies Co., Ltd. | Method and apparatus for obtaining odn logical topology information, device, and storage medium |
-
2021
- 2021-07-30 CN CN202110873162.6A patent/CN113568898B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080208888A1 (en) * | 2007-02-28 | 2008-08-28 | Kevin Mitchell | Historical data management |
| CN101976301A (en) * | 2010-10-09 | 2011-02-16 | 广东省电力设计研究院 | Method and device for preprocessing historical data in yearly load forecasting |
| US10102559B1 (en) * | 2014-09-30 | 2018-10-16 | Amazon Technologies, Inc. | Diversification of recommendations |
| CN104484673A (en) * | 2014-12-05 | 2015-04-01 | 南京大学 | Data complementation method for pattern recognition application of real-time data flow |
| US20170046426A1 (en) * | 2015-08-14 | 2017-02-16 | Nasdaq, Inc. | Computer-implemented systems and methods for intelligently retrieving, analyzing, and synthesizing data from databases |
| US20180012141A1 (en) * | 2016-07-11 | 2018-01-11 | Conduent Business Services, Llc | Method of trip prediction by leveraging trip histories from neighboring users |
| US20190067946A1 (en) * | 2017-08-28 | 2019-02-28 | General Electric Company | System and method for distribution load forecasting in a power grid |
| CN110020122A (en) * | 2017-10-16 | 2019-07-16 | Tcl集团股份有限公司 | A kind of video recommendation method, system and computer readable storage medium |
| CN108461130A (en) * | 2018-01-18 | 2018-08-28 | 邱堃 | Medical task intelligent dispatching method and system |
| US20210167852A1 (en) * | 2018-08-15 | 2021-06-03 | Huawei Technologies Co., Ltd. | Method and apparatus for obtaining odn logical topology information, device, and storage medium |
| CN110244099A (en) * | 2019-06-24 | 2019-09-17 | 河南工业大学 | Stealing detection method based on user's voltage |
| CN110502751A (en) * | 2019-08-09 | 2019-11-26 | 国网山西省电力公司 | Bulk power grid operation situation cognitive method, terminal device and storage medium |
| US20210097060A1 (en) * | 2019-10-01 | 2021-04-01 | Institute For Information Industry | Missing data compensation method, missing data compensation system, and non-transitory computer-readable medium |
| CN110780150A (en) * | 2019-10-21 | 2020-02-11 | 江苏轶一电力科技有限公司 | Transmission line fault positioning device and method based on transmission tower leakage current |
| CN111309718A (en) * | 2020-02-19 | 2020-06-19 | 南方电网科学研究院有限责任公司 | Distribution network voltage data missing filling method and device |
| CN111291280A (en) * | 2020-03-10 | 2020-06-16 | 中国科学院计算技术研究所 | Method, medium, and apparatus for fast predicting trajectory of large-scale moving object |
| CN111507412A (en) * | 2020-04-20 | 2020-08-07 | 南京工程学院 | Voltage missing value filling method based on historical data auxiliary scene analysis |
| CN112162878A (en) * | 2020-09-30 | 2021-01-01 | 深圳前海微众银行股份有限公司 | Database fault discovery method and device, electronic equipment and storage medium |
| AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
Non-Patent Citations (4)
| Title |
|---|
| 乔文俞等: "基于曲线相似与低秩矩阵的缺失电量数据补全方法", 电力建设, no. 1, pages 36 - 42 * |
| 机器不学习能咋地: "缺失填补方法汇总", Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/88811158> * |
| 李通等: "一种基于SVD的交通流量数据补全算法", 数字技术与应用, no. 1, 15 January 2016 (2016-01-15), pages 129 - 130 * |
| 贾梓健等: "基于傅里叶变换和kNNI的周期性时序数据缺失值补全算法", 软件工程, vol. 20, no. 3, 5 March 2017 (2017-03-05), pages 9 - 13 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115329907A (en) * | 2022-10-14 | 2022-11-11 | 杭州致成电子科技有限公司 | Electric load completion method and system based on DBSCAN clustering |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113568898B (en) | 2024-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110830450A (en) | Abnormal flow monitoring method, device and equipment based on statistics and storage medium | |
| CN106649832B (en) | Estimation method and device based on missing data | |
| CN109215338B (en) | Gridding-based Beidou gps vehicle track management system and method thereof | |
| CN107610021A (en) | The comprehensive analysis method of environmental variance spatial and temporal distributions | |
| CN106682763B (en) | Power load optimization prediction method for large amount of sample data | |
| CN111784093B (en) | Enterprise reworking auxiliary judging method based on power big data analysis | |
| CN114819374B (en) | Regional new energy ultra-short term power prediction method and system | |
| CN102609501B (en) | Data cleaning method based on real-time historical database | |
| CN113568898A (en) | Electric power data leakage point completion method, device, equipment and readable storage medium | |
| CN117078048A (en) | Digital twinning-based intelligent city resource management method and system | |
| CN111639060A (en) | Thermal power plant time sequence data processing method, device, equipment and medium | |
| CN118035710A (en) | Extraction method of power generation characteristics of typical scene | |
| Wang et al. | Stull: Unbiased online sampling for visual exploration of large spatiotemporal data | |
| CN111966733A (en) | Hotspot knowledge generation method and device | |
| CN116295506A (en) | Method, device, equipment and medium for predicting vehicle remaining mileage | |
| CN113554079B (en) | Power load abnormal data detection method and system based on secondary detection method | |
| CN115759250A (en) | Attribution analysis method, attribution analysis device, electronic equipment and storage medium | |
| CN112445835B (en) | Business data processing method and device, network management server and storage medium | |
| CN110244096B (en) | Method for automatically discovering and processing electric meter full code in electric energy metering system | |
| CN112732694A (en) | Cleaning method for power utilization time sequence data | |
| CN109086309B (en) | Index dimension relation definition method, server and storage medium | |
| CN113673759A (en) | Real-time marshalling method and terminal for hydrological data | |
| CN112561734A (en) | Transformer area line loss analysis system and analysis method thereof | |
| CN115858895B (en) | Multi-source heterogeneous data processing method and system for smart city | |
| CN109542510B (en) | Software V & V effectiveness measurement method based on Bayesian theory |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |