CN113378102A - Data missing preprocessing method, medium and application for short-term load prediction - Google Patents
Data missing preprocessing method, medium and application for short-term load prediction Download PDFInfo
- Publication number
- CN113378102A CN113378102A CN202110597836.4A CN202110597836A CN113378102A CN 113378102 A CN113378102 A CN 113378102A CN 202110597836 A CN202110597836 A CN 202110597836A CN 113378102 A CN113378102 A CN 113378102A
- Authority
- CN
- China
- Prior art keywords
- point
- load
- missing
- value
- slope
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000007781 pre-processing Methods 0.000 title claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 15
- 238000007620 mathematical function Methods 0.000 claims abstract description 13
- 238000012935 Averaging Methods 0.000 claims abstract description 9
- 238000012217 deletion Methods 0.000 claims abstract description 8
- 230000037430 deletion Effects 0.000 claims abstract description 8
- 238000012937 correction Methods 0.000 claims description 11
- 230000008901 benefit Effects 0.000 abstract description 3
- 230000000295 complement effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 239000000126 substance Substances 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Computational Mathematics (AREA)
- Pure & Applied Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Algebra (AREA)
- Game Theory and Decision Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a short-term load prediction data missing preprocessing method, medium and application, wherein the preprocessing method comprises the following steps: a classification step, namely acquiring historical data, and classifying time acquisition points of the historical data according to set indexes; averaging, namely calculating the load average value of similar day acquisition points according to the local deletion area to obtain a series of time acquisition point-load relation pairs; a relation fitting step, namely selecting a plurality of time acquisition point-load relation pairs in the neighborhood of the local deletion area to be fitted to obtain a multi-item mathematical function formula; and a filling step, namely calculating the derivative of the missing point in the polynomial mathematical function formula as the slope of the missing point, and filling the load value of the missing point in different calculation modes according to the slope. Compared with the prior art, the invention has the advantages of high reliability and the like.
Description
Technical Field
The invention belongs to the technical field of electric power data mining, relates to an electric power load data processing method, and particularly relates to a short-term load prediction data missing preprocessing method, medium and application.
Background
The power load prediction is one of important works of management departments such as power system scheduling, power utilization, planning and planning. Accurate short-term load prediction can economically and reasonably arrange the start and stop of the generator set in the power grid, maintain the safety and stability of the operation of the power grid, reduce unnecessary rotary reserve capacity, reasonably arrange a unit maintenance plan, ensure normal production and life of the society, effectively reduce the power generation cost and improve the economic benefit and the social benefit. Accurate medium and long-term load prediction is beneficial to determining the installed capacity, the location and the time of a new future generator set and determining the capacity increase and reconstruction of a power grid and the construction and development of the power grid. Therefore, highly accurate load prediction is required.
The power load prediction operation depends on the collected historical data, and whether the historical data is complete or accurate or not will have a serious influence on the load prediction. Particularly, in medium-and long-term load prediction, due to historical reasons, mainly negligence of data collection or custody departments, some archival data may be left or lost without record, so that data is vacant, and the collected historical data is often defective. The data loss is data vacancy and data distortion.
If the vacancy data can not be effectively compensated, calculated and corrected, the vacancy data are provided for load prediction as reference in a mode of pseudo information and pseudo change rules, so that the establishment of a load prediction model is inevitably misled, the accuracy and reliability of a prediction result are influenced, and the prediction result is directly influenced.
The conventional method for complementing the vacancy of the intermediate data mainly includes a non-adjacent mean generation method and a recursive non-adjacent mean generation method, wherein the non-adjacent mean generation method is an averaging method, and the recursive non-adjacent mean generation method is a method for using the data at two ends of the vacancy data to obtain the middle vacancy data by using the non-adjacent mean generation method and then using the data at two ends and the obtained vacancy data one by one to finally complement the vacancy data under the condition that the intermediate data has more vacancies.
The biggest defect of the methods is that the situations of continuous data loss and continuous data mutation cannot be processed, the latter load data must be utilized when the first abnormal data is corrected, and the correction effect is seriously influenced if the latter load data is still abnormal data. In summary, the prediction results in large relative errors, and the overall prediction is not accurate.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a high-reliability data missing preprocessing method, medium and application for short-term load prediction.
The purpose of the invention can be realized by the following technical scheme:
a method for short-term load prediction data loss preprocessing, comprising:
a classification step, namely acquiring historical data, and classifying time acquisition points of the historical data according to set indexes;
averaging, namely calculating the load average value of similar day acquisition points according to the local deletion area to obtain a series of time acquisition point-load relation pairs;
a relation fitting step, namely selecting a plurality of time acquisition point-load relation pairs in the neighborhood of the local deletion area to be fitted to obtain a multi-item mathematical function formula;
and a filling step, namely calculating the derivative of the missing point in the polynomial mathematical function formula as the slope of the missing point, and filling the load value of the missing point in different calculation modes according to the slope.
Further, the set index comprises a collection day type and an external climate condition of a collection point.
Further, a polyfit fitting method is adopted to obtain the polynomial mathematical function.
Further, the filling up the load value at the missing point by adopting different calculation methods according to the magnitude of the slope specifically includes:
if the slope at the missing point is less than 0, calculating to obtain a load value at the missing point by adopting an equation (1):
if the slope at the missing point is greater than 0, calculating to obtain a load value at the missing point by adopting an equation (2):
if the slope at the missing point is equal to 0, the load value at the missing point is obtained by adopting the calculation of the formula (3):
wherein, PiIndicating the load value, P, at the point of absencestaIndicating a recorded point load value, P, preceding the missing pointendIndicating that there is a recorded point load value, g, after the point of absenceimaxIs the maximum slope value, k, in the i missing point historyiR is a random number for the characteristic coefficient.
Further, a characteristic coefficient k in the above formula (1)iThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2Is a correction factor.
Further, a characteristic coefficient k in the above formula (2)iThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2Is a correction factor.
Further, the correction coefficient α1Has a value range of 0.90-0.99 and a correction coefficient alpha2The value range of (A) is 0.97-0.99.
Further, the value range of the random number R is 0.4-0.6.
The present invention also provides a computer readable storage medium comprising one or more programs for execution by one or more processors of an electronic device, the one or more programs including instructions for performing the method for short term load prediction data loss preprocessing described above.
The invention also provides an application of the short-term load prediction data missing preprocessing method, which is used for realizing the short-term load prediction of the power system based on the historical data after the load values at the missing points are supplemented.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method, time collection points of historical data are classified according to set indexes, missing data is supplemented based on similar days of local missing areas, high-quality missing sample data can be effectively supplemented, the short-term power load prediction accuracy can be improved to a certain extent, and stable operation of a power grid is facilitated.
2. The invention adopts different calculation modes to obtain the load value at the missing point according to the difference of the slope of the missing point, and has higher reliability.
3. The method of the invention can overcome the defect of large error when the non-adjacent mean generating method is used for processing continuous data.
Drawings
FIG. 1 is a schematic flow diagram of the present invention;
FIG. 2 is a T1-T12 polynomial fit plot for curve A;
FIG. 3 is a comparison of actual load curves and two complementary load curves from T1-T12 at a given day;
FIG. 4 is a T31-T46 polynomial fit plot for curve B;
FIG. 5 is a comparison of actual load curves and two complementary load curves from T31-T46 at a given day;
FIG. 6 is a T54-T66 polynomial fit plot for curve C;
FIG. 7 is a comparison of the actual load curve and the two complementary load curves from T54-T66 at a given day.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Referring to fig. 1, the present invention provides a data missing preprocessing method for short-term load prediction, including:
s101, a classification step, namely acquiring historical data, and classifying time acquisition points of the historical data according to set indexes;
s102, an averaging step, namely calculating the load average value of similar day acquisition points according to the local missing area to obtain a series of time acquisition point-load relation pairs;
s103, a relation fitting step, namely selecting a plurality of time acquisition point-load relation pairs in the neighborhood of the local deletion area, and fitting to obtain a polynomial mathematical function formula;
s104, a filling step, namely calculating the derivative of the missing point in the polynomial mathematical function formula as the slope of the missing point, and filling the load value of the missing point in different calculation modes according to the slope.
The set indexes comprise the type of the collection day and the external climate condition of the collection point. Specifically, the method classifies 96 or 288 collected daily historical data points according to the working day, the rest day, the holiday, and the corresponding external climate conditions, such as temperature, humidity, wind speed, cloud cover, sunshine hours, sunshine intensity and other indexes.
In the averaging step, the ones having the missing points do not participate in the average calculation.
In the invention, a polyfit fitting method can be adopted to obtain a polynomial, and the obtained polynomial can be expressed as:
P(x)＝A+Bx+Cx2+Dx3+......
wherein A, B, C and D are constants to be determined; p (x) represents the time point load of x, wherein x is the time point of load operation and ranges from 0 to 24 hours.
And (2) calculating a derivative of a time point at a plurality of (generally not more than 5) missing points by using the polynomial function formula P (x), and judging the slope at the time point. Let its slope be g, then its slope magnitude is divided into three kinds:
P′(x)＝g＜0
P′(x)＝g＞0
P′(x)＝g＝0
the concrete steps of filling up the load value at the missing point by adopting different calculation modes according to the slope are as follows:
if the slope at the missing point is less than 0, calculating to obtain a load value at the missing point by adopting an equation (1):
if the slope at the missing point is greater than 0, calculating to obtain a load value at the missing point by adopting an equation (2):
if the slope at the missing point is equal to 0, the load value at the missing point is obtained by adopting the calculation of the formula (3):
wherein, PiIndicating the load value, P, at the point of absencestaIndicating a recorded point load value, P, preceding the missing pointendIndicating that there is a recorded point load value, g, after the point of absenceimaxIs the maximum slope value, k, in the i missing point historyiAnd R is a random number and has a value range of 0.4-0.6.
Characteristic coefficient k in formula (1)iThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2To correct the coefficient, α1Has a value range of 0.90-0.99, alpha2The value range of (A) is 0.97-0.99.
Characteristic coefficient k in the formula (2)iThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2To correct the coefficient, α1Has a value range of 0.90-0.99, alpha2The value range of (A) is 0.97-0.99.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Load data of 10kV power main equipment of a regional power grid for three years and 1095 days are obtained, and 2372865 load curves are obtained. Each load curve recorded 96 points throughout the day for a total of 227795040 data points, for which 1 curve was artificially manufactured with partially missing data.
The data set S is all 2372865 load curve sets and is divided into two load curve sets S1 and S2; s1 is three curve sets with missing values, namely curve A, curve B and curve C, and S2 is the rest curve set without missing values.
Example 1
The artificially produced curve a had 2 deletion values, 6 at T, and 7 as shown in table 1.
TABLE 1 load data for the artificial manufacturing curve A with 2 missing values
Note: bracketed data as missing data
The data set S2 is categorized according to the working day, the holiday, and the corresponding external climate conditions, such as temperature, humidity, wind speed, cloud cover, sunshine duration, sunshine intensity, etc., and the curve a belongs to the working day. Data set S2 was averaged for each acquisition point on similar days of the same class (points with missing data did not participate in the averaging calculation). Thus, a series of time acquisition point-load value relationship pairs are obtained. Then, after 6 points before and after 7 points T, the embodiment selects T1-T12, and a polynomial mathematical function expression is fitted by polyfit according to the relation between the time acquisition point and the load value, so that the time-load relation at any time point of the section can be represented. The result of the polynomial fit is shown in fig. 2.
Calculating the derivative of the time point at the missing point of T6 and T7 according to the polynomial function obtained by polyfit, and judging the magnitude of the slope at the point, wherein the slope P' (6) of T6 and T7 is g6< 0 and P' (7) ═ g7< 0, then insert the formula:
wherein the content of the first and second substances,
wherein the content of the first and second substances,
in the above formula, g6maxAnd g7maxThe maximum slope values in the historical data of the points T6 and T7 respectively, if the correction coefficient is preferentially selected by machine learning from the historical data, alpha1Take 0.975, alpha2Taking 0.980, P is calculated therefrom64591.5, similarly, the daily load value at point T7 can be calculated.
By contrast, as shown in fig. 3, the dashed line represents the filling situation of the missing value of the data (T ═ 6,7) on a certain day of the 10kV power master of a regional power grid in this embodiment by the method of the present invention, and the thick solid line represents the filling situation of the non-adjacent mean. Obviously, for the 2 discrete missing values, the filling effect of the invention is obviously better than that of the non-adjacent mean filling mode.
Example 2
Curve B was artificially created with 4 consecutive missing values and data were missing at T-37-40 as shown in table 2.
TABLE 2 load data for artificial curve B with 4 consecutive missing values
Note: bracketed data as missing data
Curve B the day belongs to the working day. Data set S2 was averaged for each acquisition point on similar days of the same class (points with missing data did not participate in the averaging calculation). Thus, a series of time acquisition point-load value relationship pairs are obtained. Then, about 6 points before and after the T-37-40, the embodiment selects T31-T46, and a polynomial mathematical function expression is fitted by the time acquisition point-load value relations, so that the time-load relation at any time point of the section can be represented. The result of the polynomial fit is shown in fig. 4.
According to a polynomial function obtained by polyfit, the time point of each missing point T37-40 is differentiated to judge the slope of the point, and the slope P' (37) of T37-T40 is g37＞0，P′(38)＝g38> 0 up to P' (40) ═ g40> 0, then substituting the formula:
wherein the content of the first and second substances,
wherein the content of the first and second substances,
in the above formula (P)38And P39Similarly, omit), g37max、g38max、g39maxAnd g40maxT37, T38, T39 and T40 respectively, the maximum slope value in the historical data of the point, such as the correction coefficient is preferentially selected by the historical data through machine learning, and alpha1Take 0.90, alpha2Taking 0.97, P was calculated therefrom376433.6, similarly, the daily load values of T38, T39 and T40 can be obtained.
By contrast with the conventional recursive non-adjacent mean value filling method, as shown in fig. 5, the dotted line indicates that the data (T ═ 37 to 40) is missing from the local grid 10kV power master device on a certain day in the present embodiment, and the thick solid line indicates that the recursive non-adjacent mean value is filled. Obviously, for 4 continuous discrete missing values, the filling effect of the method is obviously superior to that of a recursive non-adjacent mean filling mode.
Example 3
The artificial curve C had 1 missing value, missing at T59, as shown in table 3.
TABLE 3 load data for artificial manufacturing curve C with 1 missing value
Note: bracketed data as missing data
Curve C this day belongs to the working day. Data set S2 was averaged for each acquisition point on similar days of the same class (points with missing data did not participate in the averaging calculation). Thus, a series of time acquisition point-load value relationship pairs are obtained. Then, about 6 points before and after T59, T54-T65 is selected in the embodiment, and a polynomial mathematical function expression is fitted by using polyfit according to the relation between the time acquisition point and the load value, so that the time-load relation at any time point of the section can be expressed. The result of the polynomial fit is shown in fig. 6.
Calculating the derivative of the time point at the missing point of T59 according to the polynomial equation obtained by polyfit, and judging the size of the slope at the point, wherein P' (59) g 590 and slope P' (58) of T58, T60 is g58> 0 and P' (60) ═ g60< 0, then insert the formula:
in the above formula, R is a random number between 0.4 and 0.6, and 0.5 is selected here, whereby it can be calculated that the daily load value at point T59 is 6346.3.
By contrast, as shown in fig. 7, the dashed line represents the filling situation of the missing value of the data (T59) of a certain day of the 10kV power master of a regional power grid in this embodiment by the method of the present invention, and the thick solid line represents the filling situation of the non-adjacent mean value. Obviously, for the 1 discrete missing value, the filling effect of the invention is obviously better than that of the non-adjacent mean filling mode.
After the supplemented historical data is obtained by the method, more accurate and reliable short-term load prediction of the power system can be carried out.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (10)
1. A method for short-term load prediction data loss preprocessing is characterized by comprising the following steps:
a classification step, namely acquiring historical data, and classifying time acquisition points of the historical data according to set indexes;
averaging, namely calculating the load average value of similar day acquisition points according to the local deletion area to obtain a series of time acquisition point-load relation pairs;
a relation fitting step, namely selecting a plurality of time acquisition point-load relation pairs in the neighborhood of the local deletion area to be fitted to obtain a multi-item mathematical function formula;
and a filling step, namely calculating the derivative of the missing point in the polynomial mathematical function formula as the slope of the missing point, and filling the load value of the missing point in different calculation modes according to the slope.
2. The method of claim 1, wherein the set indicators include a type of collection day and climatic conditions outside the collection point.
3. The method for short-term load prediction data loss preprocessing as claimed in claim 1, wherein the polynomial mathematical function is obtained by using a polyfit fitting method.
4. The method as claimed in claim 1, wherein the filling up the load value at the missing point by different calculation methods according to the slope magnitude specifically includes:
if the slope at the missing point is less than 0, calculating to obtain a load value at the missing point by adopting an equation (1):
if the slope at the missing point is greater than 0, calculating to obtain a load value at the missing point by adopting an equation (2):
if the slope at the missing point is equal to 0, the load value at the missing point is obtained by adopting the calculation of the formula (3):
wherein, PiIndicating the load value, P, at the point of absencestaIndicating a recorded point load value, P, preceding the missing pointendIndicating that there is a recorded point load value, g, after the point of absenceimaxIs the maximum slope value, k, in the i missing point historyiR is a random number for the characteristic coefficient.
5. The method for short term load prediction data loss preprocessing as claimed in claim 4, wherein the formula (1)Characteristic coefficient k iniThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2Is a correction factor.
6. The method as claimed in claim 4, wherein the coefficient k in the formula (2) is a characteristic coefficientiThe formula is adopted to calculate and obtain:
wherein alpha is1And alpha2Is a correction factor.
7. The method as claimed in claim 5 or 6, wherein the correction factor α is a1Has a value range of 0.90-0.99 and a correction coefficient alpha2The value range of (A) is 0.97-0.99.
8. The method for short-term load prediction data loss preprocessing as claimed in claim 5, wherein the value of the random number R ranges from 0.4 to 0.6.
9. A computer-readable storage medium comprising one or more programs for execution by one or more processors of an electronic device, the one or more programs including instructions for performing the method for short-term load prediction data loss preprocessing as claimed in any of claims 1-8.
10. Use of a method according to any of claims 1-8 for short term load forecast data loss preprocessing, characterized in that the power system short term load forecast is implemented based on historical data after filling up the load values at the points of loss.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110597836.4A CN113378102A (en) | 2021-05-31 | 2021-05-31 | Data missing preprocessing method, medium and application for short-term load prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110597836.4A CN113378102A (en) | 2021-05-31 | 2021-05-31 | Data missing preprocessing method, medium and application for short-term load prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113378102A true CN113378102A (en) | 2021-09-10 |
Family
ID=77574936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110597836.4A Pending CN113378102A (en) | 2021-05-31 | 2021-05-31 | Data missing preprocessing method, medium and application for short-term load prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113378102A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779109A (en) * | 2021-09-15 | 2021-12-10 | 南京才华智能科技有限公司 | Power data preprocessing method based on context environment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20040054906A (en) * | 2002-12-18 | 2004-06-26 | 황병기 | The System & Method For Managing Sewer-Pipe |
| US9104964B1 (en) * | 2012-11-16 | 2015-08-11 | Amazon Technologies, Inc. | Data estimation of historical data that identifies fine grained time period data and coarse grained time period data and applies correlated slopes to time periods |
| CN109145949A (en) * | 2018-07-19 | 2019-01-04 | 山东师范大学 | Non-intrusive electrical load monitoring and decomposition method and system based on integrated study |
| CN109636035A (en) * | 2018-12-12 | 2019-04-16 | 北京天诚同创电气有限公司 | Load forecasting model creation method and device, Methods of electric load forecasting and device |
| US20190347590A1 (en) * | 2014-03-12 | 2019-11-14 | Bahwan CyberTek Private Limited | Intelligent Decision Synchronization in Real Time for both Discrete and Continuous Process Industries |
| US10485489B1 (en) * | 2018-12-11 | 2019-11-26 | Fifth Eye Inc. | System and method for assessing and monitoring the hemodynamic condition of a patient |
| CN110533738A (en) * | 2019-09-02 | 2019-12-03 | 上海联影医疗科技有限公司 | Rebuild data processing method, device, medical image system 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 |
-
2021
- 2021-05-31 CN CN202110597836.4A patent/CN113378102A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20040054906A (en) * | 2002-12-18 | 2004-06-26 | 황병기 | The System & Method For Managing Sewer-Pipe |
| US9104964B1 (en) * | 2012-11-16 | 2015-08-11 | Amazon Technologies, Inc. | Data estimation of historical data that identifies fine grained time period data and coarse grained time period data and applies correlated slopes to time periods |
| US20190347590A1 (en) * | 2014-03-12 | 2019-11-14 | Bahwan CyberTek Private Limited | Intelligent Decision Synchronization in Real Time for both Discrete and Continuous Process Industries |
| CN109145949A (en) * | 2018-07-19 | 2019-01-04 | 山东师范大学 | Non-intrusive electrical load monitoring and decomposition method and system based on integrated study |
| US10485489B1 (en) * | 2018-12-11 | 2019-11-26 | Fifth Eye Inc. | System and method for assessing and monitoring the hemodynamic condition of a patient |
| CN109636035A (en) * | 2018-12-12 | 2019-04-16 | 北京天诚同创电气有限公司 | Load forecasting model creation method and device, Methods of electric load forecasting and device |
| CN110533738A (en) * | 2019-09-02 | 2019-12-03 | 上海联影医疗科技有限公司 | Rebuild data processing method, device, medical image system 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 |
|---|
| 姚建刚;龙立波;钱卫华;张凯;: "基于.NET的负荷分析与预测系统的研究与实现", 湖南大学学报(自然科学版), no. 09, 25 September 2007 (2007-09-25) * |
| 李世玺;孙宪坤;尹玲;张仕森;: "一种基于混沌理论和LSTM的GPS高程时间序列预测方法", 导航定位学报, no. 01, 20 February 2020 (2020-02-20) * |
| 王晨霖;杨洁;居文军;顾复;陈芨熙;纪杨建;: "基于智能家电的短期电力负荷预测与削峰填谷优化", 浙江大学学报(工学版), no. 07 * |
| 肖红军: ""数据驱动的污水处理过程故障诊断与多步预测"", 《中国博士学位论文全文数据库》, 15 May 2017 (2017-05-15) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113779109A (en) * | 2021-09-15 | 2021-12-10 | 南京才华智能科技有限公司 | Power data preprocessing method based on context environment |
| CN113779109B (en) * | 2021-09-15 | 2024-05-24 | 南京才华智能科技有限公司 | Electric power data preprocessing method based on context environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110648249B (en) | Annual power balance measuring and calculating method, device and equipment | |
| CN107730076B (en) | Method for analyzing comprehensive utilization efficiency of regional distribution network equipment | |
| CN106570790B (en) | Wind power plant output data restoration method considering wind speed data segmentation characteristics | |
| CN102426674A (en) | Power system load prediction method based on Markov chain | |
| CN111049193B (en) | Standby demand dynamic evaluation method for multiple scheduling scenes of wind power system | |
| CN110909958A (en) | Short-term load prediction method considering photovoltaic grid-connected power | |
| CN102509173A (en) | Markov chain based method for accurately forecasting power system loads | |
| CN110866658A (en) | Method for predicting medium and long term load of urban power grid | |
| CN114372360A (en) | Method, terminal and storage medium for power load prediction | |
| CN117691645A (en) | Energy storage system for intelligent micro-grid | |
| CN113378102A (en) | Data missing preprocessing method, medium and application for short-term load prediction | |
| CN110991747A (en) | Short-term load prediction method considering wind power plant power | |
| CN111799798B (en) | Method and system for improving accuracy of future state load flow calculation result | |
| CN116562398B (en) | Power load prediction model training method, prediction method, electronic device and medium | |
| CN115511156A (en) | Main transformer overload early warning method and system based on load prediction | |
| CN115877145A (en) | Transformer overload working condition big data cross evaluation system and method | |
| CN112633642B (en) | Method, system, device and storage medium for predicting standby demand of power system | |
| CN115204944A (en) | Energy storage optimal peak-to-valley price difference measuring and calculating method and device considering whole life cycle | |
| CN114971427A (en) | Entropy weight-analytic hierarchy process-based energy storage power station grid-connected safety evaluation method | |
| CN114782001A (en) | Power grid infrastructure project optimization method and system based on life cycle cost | |
| CN114168662A (en) | Power distribution network problem combing and analyzing method and system based on multiple data sources | |
| CN110175705B (en) | Load prediction method and memory and system comprising same | |
| CN113691174A (en) | Distribution transformer gear identification method and system based on exponential nonlinear regression | |
| CN110991748A (en) | Short-term load prediction method for urban power grid | |
| CN111368257A (en) | Method and device for analyzing and predicting coal-to-electricity load characteristics |
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 |