CN109408767A - A kind of complementing method towards power grid missing data - Google Patents
A kind of complementing method towards power grid missing data Download PDFInfo
- Publication number
- CN109408767A CN109408767A CN201811209336.3A CN201811209336A CN109408767A CN 109408767 A CN109408767 A CN 109408767A CN 201811209336 A CN201811209336 A CN 201811209336A CN 109408767 A CN109408767 A CN 109408767A
- Authority
- CN
- China
- Prior art keywords
- data
- value
- sample
- missing
- power grid
- 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 24
- 238000012417 linear regression Methods 0.000 claims description 12
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 claims description 7
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000006467 substitution reaction Methods 0.000 claims description 2
- 238000004458 analytical method Methods 0.000 abstract description 8
- 238000007405 data analysis Methods 0.000 abstract description 6
- 238000012423 maintenance Methods 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000012502 risk assessment Methods 0.000 abstract description 2
- 239000013589 supplement Substances 0.000 abstract 1
- 230000005611 electricity Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
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
Abstract
The invention discloses a kind of complementing method towards power grid missing data, the monitoring of the Various types of data in electric system and acquisition be electric power system dispatching operation, security and stability analysis, equipment state and risk assessment basis.However, in the actual motion of electric system, due to data acquisition channel mistake, remote-terminal unit failure and other reasons will lead to observation data and unusual situation occur, it is inconsistent with most of observations, additionally due to the maintenance of route, cutting load has a power failure and major issue impact is possible to will lead to observation data and goes against the established rules, electric power system data is set to cause difficulty to the analysis of smart electric grid system, therefore it needs to supplement the data of missing using a kind of data compensation process before data analysis complete, improve the accuracy of electric power system data analysis and the availability of class of a curve data, support is provided for subsequent analysis.
Description
Technical field
The present invention relates to a kind of data filling methods, and in particular to a kind of complementing method towards power grid missing data.
Background technique
With the arrival of big data era and the continuous improvement of electric power information degree, electric power big data is in explosive
Increase, and type is also more and more.The data that acquisition is mostly used in existing Power System Analysis establish model realization user power utilization point
Analysis is innovated and is developed particularly important to smart grid business model to offers helps such as subsequent stealing detection, load predictions.
The monitoring and acquisition of Various types of data in electric system are electric power system dispatching operation, security and stability analysis, equipment
The basis of state and risk assessment.However, in the actual motion of electric system, due to data acquisition channel mistake, long-range end
End unit failure and other reasons will lead to observation data and unusual situation occur, inconsistent with most of observations, additionally due to route
Maintenance, cutting load has a power failure and major issue impact is possible to will lead to observation data and goes against the established rules, make electric power system data to intelligence
The analysis of energy network system causes difficulty, therefore needs to mend the data of missing using a kind of data compensation process before data analysis
Charge it is whole, improve electric power system data analysis accuracy and class of a curve data availability, provide support for subsequent analysis.
Summary of the invention
If the technical problem to be solved by the present invention is to the primary datas needed before data analysis to be lost, there is no a kind of
The mode of filling up can be made up to greatest extent, and the application is supplemented the data of missing completely as far as possible by data compensation process,
It is designed to provide a kind of complementing method towards power grid missing data, is solved the problem above-mentioned.
The present invention is achieved through the following technical solutions:
A kind of complementing method towards power grid missing data, which is characterized in that described method includes following steps:
S1: being obtained using arest neighbors and lack the nearest sample of sample, uses the value of this sample as filling up value y1;
For giving sample xi=(xi1, xi2..., xin) and xj=(xj1, xj2..., xjn), wherein xik(0 < k≤n)
Missing,
Formula is filled up by power grid missing data:It finds out
The corresponding y of the smallest di, y1=yi;
S2: another value of filling up y is obtained using linear regression2;The given example by d attribute description, x=(x1;
x2;...;xd), wherein xiIt is value of the x in ith attribute,
S3: it finally fills up value 1 by what is obtained and fills up that value 2 is adaptive weighted to be obtained final filling up value Y.
In the step S2, another value of filling up y is obtained using linear regression2;
The given example by d attribute description, x=(x1;x2;...;xd), wherein xiIt is x taking in ith attribute
Value carries out value using formula, which is y=(x1l,x2l,...xtl,...xdl) t ≠ i, w=(XTX)-1XTY, y2=wTxi。
In the step S3, finally by obtain fill up value 1 and fill up value 2 it is adaptive weighted obtain it is final fill up value Y,
Its calculating process are as follows: k1=y2/(y1+y2), k2=y1/(y1+y2), Y=k1*y1+k2*y2。
Further, the step S1 is before obtaining missing sample recently, if without missing sample, by original initial
Sample carries out step S2 and is finally filled up value.
Further, the value Y that finally fills up in the step S3 is substituted into power grid missing data sample, carries out filling up meter
It calculates.
Compared with prior art, the present invention having the following advantages and benefits:
1, a kind of complementing method towards power grid missing data of the present invention, can quickly and easily find shape by nearest neighbor algorithm
Immediate area's moment of state;Electricity consumption behavior classification need not be carried out to user day curve, can directly match most like day curve
It carries out missing power data to fill up, effectively improves the accuracy and actual effect of data filling, provided for electric power system data analysis
Data supporting;
Detailed description of the invention
Attached drawing described herein is used to provide to further understand the embodiment of the present invention, constitutes one of the application
Point, do not constitute the restriction to the embodiment of the present invention.In the accompanying drawings:
Fig. 1 is the method for the present invention flow diagram.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below with reference to embodiment and attached drawing, to this
Invention is described in further detail, and exemplary embodiment of the invention and its explanation for explaining only the invention, are not made
For limitation of the invention.
Embodiment one
It is a kind of towards power grid missing data as shown in Figure 1, a kind of complementing method towards power grid missing data of the present invention
Complementing method, which is characterized in that described method includes following steps:
S1: being obtained using arest neighbors and lack the nearest sample of sample, uses the value of this sample as filling up value y1;
For giving sample xi=(xi1, xi2..., xin) and xj=(xj1, xj2..., xjn), wherein xik(0 < k≤n)
Missing,
Formula is filled up by power grid missing data:It finds out
The corresponding y of the smallest di, y1=yi;
S2: another value of filling up y is obtained using linear regression2;The given example by d attribute description, x=(x1;
x2;...;xd), wherein xiIt is value of the x in ith attribute,
S3: it finally fills up value 1 by what is obtained and fills up that value 2 is adaptive weighted to be obtained final filling up value Y.
In the step S2, another value of filling up y is obtained using linear regression2;
The given example by d attribute description, x=(x1;x2;...;xd), wherein xiIt is x taking in ith attribute
Value carries out value using formula, which is y=(x1l,x2l,...xtl,...xdl) t ≠ i, w=(XTX)-1XTY, y2=wTxi。
In the step S3, finally by obtain fill up value 1 and fill up value 2 it is adaptive weighted obtain it is final fill up value Y,
Its calculating process are as follows: k1=y2/(y1+y2), k2=y1/(y1+y2), Y=k1*y1+k2*y2。
The step S1 is before obtaining missing sample recently, if being carried out without missing sample by original initial sample
Step S2 is finally filled up value.
The value Y that finally fills up in the step S3 is substituted into power grid missing data sample, carries out filling up calculating.
Embodiment two
The present embodiment carries out actual optimization and obtains, net certain years data of certain provincial company with state on the basis of embodiment one
For research object, missing is carried out to the electric network data of each ground company, cities and counties under it and is filled up, and to have complete certain year electricity
For network data and wherein 7 cities, to show the data based on nearest neighbor algorithm and linear regression algorithm Weight number adaptively
Complementing method.
7 cities that certain provincial company is netted by state are chosen herein, are named as c1, c2, c3, c4, c5, c6, c7;Choose 4 dimension power grids
Feature of the data as training, names x1, x2, x3, x4.The value of this paper missing at random sample c6 feature x1, missing at random sample
The value of c7 feature x4,
x1 | x2 | x3 | x4 | |
c1 | 5551.33 | 1374.26 | 933.8 | 807 |
c2 | 960.22 | 379.36 | 180.55 | 228 |
c3 | 552.25 | 360 | 141.1 | 126 |
c4 | 921.3 | 389.2 | 277.4 | 195 |
c5 | 657.9 | 494.2 | 0 | 0 |
c6 | 647.7 | 327.2 | 96.85 | 93 |
c7 | 870.85 | 520 | 155.75 | 198 |
The complete electric network data table of table 1
The tables of data of the missing of table 2 sample c6 feature x1
The feature x1 of missing at random sample c6 calculates separately sample c6 at a distance from other samples according to formula (1) first,
Show that missing sample c6 is minimum at a distance from sample c3, then sampling this c3 feature x1 is to fill up value y1, y1=552.25;By formula
(4) y is obtained2=691.79;Y=614.19 is finally obtained by formula (7).
x1 | x2 | x3 | x4 | |
c1 | 5551.33 | 1374.26 | 933.8 | 807 |
c2 | 960.22 | 379.36 | 180.55 | 228 |
c3 | 552.25 | 360 | 141.1 | 126 |
c4 | 921.3 | 389.2 | 277.4 | 195 |
c5 | 657.9 | 494.2 | 0 | 0 |
c6 | 647.7 | 327.2 | 96.85 | 93 |
c7 | 870.85 | 520 | 155.75 |
The tables of data of the missing of table 3 sample c7 feature x4
The feature x4 of missing at random sample c7 calculates separately sample c7 at a distance from other samples according to formula (1) first,
Show that missing sample c7 is minimum at a distance from sample c2, then sampling this c2 feature x2 is to fill up value y1, y1=228;By formula (4)
Obtain y2=139.42;Y=173.03 is finally obtained by formula (7).
In order to which more intuitive see is filled up as a result, result is expressed as table 4, table 5:
Error | Percentage error | |
Arest neighbors is filled up | 95.45 | 14.74% |
Linear regression is filled up | 44.09 | 6.81% |
Arest neighbors and linear regression Weight number adaptively | 33.51 | 5.17% |
Table 4 lacks the comparison that sample c6 feature x1 fills up result
Error | Percentage error | |
Arest neighbors is filled up | 30 | 15.16% |
Linear regression is filled up | 58.58 | 29.59% |
Arest neighbors and linear regression Weight number adaptively | 24.97 | 12.61% |
Table 5 lacks the comparison that sample c7 feature x4 fills up result
As can be seen that nearest neighbor algorithm and linear regression algorithm Weight number adaptively are filled up than commonly most from table 4, table 5
The filling up of neighbour, mean value is filled up, linear regression fills up that result is good, error is small, is more nearly initial data, is shown of the invention
Accuracy.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (3)
1. a kind of complementing method towards power grid missing data, which is characterized in that described method includes following steps:
S1: being obtained using arest neighbors and lack the nearest sample of sample, uses the value of this sample as filling up value y1;
For giving sample xi=(xi1, xi2..., xin) and xj=(xj1, xj2..., xjn), wherein xik(0 < k≤n) missing,
Formula is filled up by power grid missing data:Find out minimum
The corresponding y of di, y1=yi;
S2: another value of filling up y is obtained using linear regression2;The given example by d attribute description, x=(x1;x2;...;
xd), wherein xiIt is value of the x in ith attribute,
S3: it finally fills up value 1 by what is obtained and fills up that value 2 is adaptive weighted to be obtained final filling up value Y.
2. a kind of complementing method towards power grid missing data according to claim 1, which is characterized in that the step S1
Before obtaining missing sample recently, if carrying out step S2 without missing sample by original initial sample and finally being filled up
Value.
3. according to a kind of complementing method towards power grid missing data described in claim 1, which is characterized in that in the step S3
Finally fill up value Y substitution power grid missing data sample in, carry out filling up calculating.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811209336.3A CN109408767A (en) | 2018-10-17 | 2018-10-17 | A kind of complementing method towards power grid missing data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811209336.3A CN109408767A (en) | 2018-10-17 | 2018-10-17 | A kind of complementing method towards power grid missing data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109408767A true CN109408767A (en) | 2019-03-01 |
Family
ID=65467358
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811209336.3A Pending CN109408767A (en) | 2018-10-17 | 2018-10-17 | A kind of complementing method towards power grid missing data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109408767A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109933582A (en) * | 2019-03-11 | 2019-06-25 | 国家电网有限公司 | Data processing method and device |
CN110046152A (en) * | 2019-04-19 | 2019-07-23 | 国网河南省电力公司经济技术研究院 | A method of processing electricity consumption data missing values |
CN110274995A (en) * | 2019-06-18 | 2019-09-24 | 深圳市美兆环境股份有限公司 | Fill the determination method, apparatus and computer equipment of data |
CN110781176A (en) * | 2019-11-06 | 2020-02-11 | 国网山东省电力公司威海供电公司 | Power grid data quality improvement method based on data correlation |
CN110879328A (en) * | 2019-11-29 | 2020-03-13 | 国网辽宁省电力有限公司锦州供电公司 | Method for processing power data at fault moment based on reverse distance weighting |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103440283A (en) * | 2013-08-13 | 2013-12-11 | 江苏华大天益电力科技有限公司 | Vacancy filling system for measured point data and vacancy filling method |
CN104298893A (en) * | 2014-09-30 | 2015-01-21 | 西南交通大学 | Imputation method of genetic expression deletion data |
CN105049460A (en) * | 2014-12-19 | 2015-11-11 | 国网电力科学研究院 | Smart preservation technology for power quality data |
CN106096324A (en) * | 2016-08-26 | 2016-11-09 | 清华大学 | The power transmission and transformation main equipment load data disappearance returned based on k neighbour fills up algorithm |
CN106651651A (en) * | 2016-12-12 | 2017-05-10 | 全球能源互联网研究院 | Data filling method and device for utilization power curve of grid user |
CN107862409A (en) * | 2017-11-06 | 2018-03-30 | 重庆大学 | A kind of a large amount of missing data complementing methods of transformer station's power transmission and transforming equipment based on regression analysis |
-
2018
- 2018-10-17 CN CN201811209336.3A patent/CN109408767A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103440283A (en) * | 2013-08-13 | 2013-12-11 | 江苏华大天益电力科技有限公司 | Vacancy filling system for measured point data and vacancy filling method |
CN104298893A (en) * | 2014-09-30 | 2015-01-21 | 西南交通大学 | Imputation method of genetic expression deletion data |
CN105049460A (en) * | 2014-12-19 | 2015-11-11 | 国网电力科学研究院 | Smart preservation technology for power quality data |
CN106096324A (en) * | 2016-08-26 | 2016-11-09 | 清华大学 | The power transmission and transformation main equipment load data disappearance returned based on k neighbour fills up algorithm |
CN106651651A (en) * | 2016-12-12 | 2017-05-10 | 全球能源互联网研究院 | Data filling method and device for utilization power curve of grid user |
CN107862409A (en) * | 2017-11-06 | 2018-03-30 | 重庆大学 | A kind of a large amount of missing data complementing methods of transformer station's power transmission and transforming equipment based on regression analysis |
Non-Patent Citations (2)
Title |
---|
侯贺: "缺失数据处理方法的研究及其在软测量技术中的应用", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
朱晓峰: "缺失值填充的若干问题研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109933582A (en) * | 2019-03-11 | 2019-06-25 | 国家电网有限公司 | Data processing method and device |
CN110046152A (en) * | 2019-04-19 | 2019-07-23 | 国网河南省电力公司经济技术研究院 | A method of processing electricity consumption data missing values |
CN110274995A (en) * | 2019-06-18 | 2019-09-24 | 深圳市美兆环境股份有限公司 | Fill the determination method, apparatus and computer equipment of data |
CN110781176A (en) * | 2019-11-06 | 2020-02-11 | 国网山东省电力公司威海供电公司 | Power grid data quality improvement method based on data correlation |
CN110879328A (en) * | 2019-11-29 | 2020-03-13 | 国网辽宁省电力有限公司锦州供电公司 | Method for processing power data at fault moment based on reverse distance weighting |
CN110879328B (en) * | 2019-11-29 | 2021-10-19 | 国网辽宁省电力有限公司锦州供电公司 | Method for processing power data at fault moment based on reverse distance weighting |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109408767A (en) | A kind of complementing method towards power grid missing data | |
Frysztacki et al. | The strong effect of network resolution on electricity system models with high shares of wind and solar | |
Yan et al. | Wind power grouping forecasts and its uncertainty analysis using optimized relevance vector machine | |
CN102945223B (en) | Method for constructing joint probability distribution function of output of a plurality of wind power plants | |
CN103020423B (en) | The method of output of wind electric field correlation properties is obtained based on copula function | |
Goh et al. | Wind energy assessment considering wind speed correlation in Malaysia | |
Feng et al. | A taxonomical review on recent artificial intelligence applications to PV integration into power grids | |
Choksi et al. | Feature based clustering technique for investigation of domestic load profiles and probabilistic variation assessment: Smart meter dataset | |
Chiodo et al. | Inverse Burr distribution for extreme wind speed prediction: Genesis, identification and estimation | |
CN105550275B (en) | A kind of microblogging transfer amount prediction technique | |
Wang et al. | Evolution of coordination degree of eco-economic system and early-warning in the Yangtze River Delta | |
CN104700118A (en) | Pulmonary nodule benignity and malignancy predicting method based on convolutional neural networks | |
CN103996074A (en) | CFD and improved PSO based microscopic wind-farm site selection method of complex terrain | |
CN104376371B (en) | A kind of distribution based on topology is layered load forecasting method | |
Azzopardi et al. | Decision support system for ranking photovoltaic technologies | |
CN105574541A (en) | Compactness sorting based network community discovery method | |
CN102801629A (en) | Traffic matrix estimation method | |
Fleischer | Minimising the effects of spatial scale reduction on power system models | |
Sun et al. | Spatial modelling the location choice of large-scale solar photovoltaic power plants: Application of interpretable machine learning techniques and the national inventory | |
Frysztacki et al. | A comparison of clustering methods for the spatial reduction of renewable electricity optimisation models of Europe | |
CN103049609A (en) | Wind power multi-stage scene simulation method | |
Miraftabzadeh et al. | K-means and alternative clustering methods in modern power systems | |
Tian et al. | Multi-scale solar radiation and photovoltaic power forecasting with machine learning algorithms in urban environment: A state-of-the-art review | |
CN104751253B (en) | Distribution power flow Forecasting Methodology based on B- spline Basis bottom developed curve cluster | |
Adkins et al. | A geospatial framework for electrification planning in developing countries |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190301 |