CN105224733B - The method that wind power abandons wind data feature recognition - Google Patents
The method that wind power abandons wind data feature recognition Download PDFInfo
- Publication number
- CN105224733B CN105224733B CN201510599490.6A CN201510599490A CN105224733B CN 105224733 B CN105224733 B CN 105224733B CN 201510599490 A CN201510599490 A CN 201510599490A CN 105224733 B CN105224733 B CN 105224733B
- Authority
- CN
- China
- Prior art keywords
- wind
- power
- wind speed
- data
- probability
- 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.)
- Active
Links
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The present invention is a kind of recognition methods that wind data feature is abandoned based on wind power, is characterized in:For wind speed power data, differentiate that it is present by scatter diagram and abandon wind data, it is then rejected by viscous interval method and abandons wind point, remaining air speed data is Wind turbines normal output data, and the foundation in viscous section is as follows:Its scatter diagram is drawn according to without the actual wind speed power for abandoning wind blower fan;Wind speed interval is divided with interval 1m/s;Obtain the difference power (DP) that the power corresponding to each section wind speed and this wind speed obtain from calibration power curve;Difference power (DP) is fitted with normal distribution, obtains its parameter μ, σ;According to the 3 viscous sections of σ rule definitions [L, R], the wherein σ of L=μ 3 σ, R=μ+3, blower fan can be rejected according to it abandons wind data.With scientific and reasonable, the advantages of strong applicability, the wind data of abandoning of Wind turbines can be rejected, ensure the precision that actual wind speed power curve is established.
Description
Technical field
The present invention relates to technical field of wind power, is that a kind of wind power abandons wind data characteristic recognition method.
Background technology
Wind-power electricity generation is removed water beyond energy in field of renewable energy, and technology is most ripe, exploit condition on the largest scale and business
One of generation mode of industry development prospect.The typical problem of wind power plant modeling is structure reflection wind farm wind velocity and its output work
The mathematical modeling of relation between rate.The document wherein having characterizes Wind turbines entirety power coefficient and equivalent wind by proposing
Speed is modeled to curve.Wherein the selection of equivalent wind speed then needs to probe into a large amount of wind speed power datas.Some documents
Power curve is modeled with the fuzzy logic method based on cluster centre, used clustering method then needs to enter data
Row cluster analysis.The conventional modeling method of some document analysises is such as:On the basis of Bean's method, most probable number method and maximum value process
Have also been proposed a kind of new method:Maximum likelihood estimate based on mathematic expectaion.And the presence for abandoning wind data can be to wind turbine
Output analysis, status monitoring and the control of group have an impact, so as to influence Equivalent Model and wind power prediction model establishes
Accuracy, therefore the rejecting to abandoning wind data is particularly important.The present invention passes through analysis on the basis of document above is analyzed
The characteristics of calibration power curve annex historical data, the creative definition for giving viscous section, wind number is abandoned to reject with this
According to, and demonstrate by probability density fitting the validity of the method.
The content of the invention
The object of the present invention is to overcome the deficiencies in the prior art, there is provided a kind of scientific and reasonable, strong applicability, can be to wind
The wind data of abandoning of group of motors is rejected, and the wind power that guarantee actual wind speed power curve establishes precision abandons the knowledge of wind data feature
Other method.
The purpose of the present invention is realized by following technical scheme:A kind of wind power abandons wind data feature recognition side
Method, it is characterised in that it comprises the following steps:
1) data acquisition
The whole field Wind turbines of wind power plant are gathered, the rated power of every single unit, wind speed, rated wind speed is cut and cuts out
The data of wind speed, and certain blower fan is actually investigated in certain whole month in month without wind is abandoned, its power swing is only by fluctuations in wind speed and its wind-powered electricity generation
Machine internal conversion is determined as precondition;
2) definition of progress of disease error
It is bent in calibration power under equal wind speed that progress of disease error (DP) refers to that the actual power corresponding to actual wind speed subtracts
Power under line,
DP=P (i)-f (s (i)) (1)
Wherein s (i) is the wind speed at the i-th moment, and P (i) is the actual wind power at the i-th moment, and f (s (i)) is based on now
The progress of disease power that the wind speed at quarter is obtained using calibration power curve;
3) 3 σ criterions
σ represents standard deviation in normal distribution, and μ represents average, and x=μ are the symmetry axis of image, and 3 σ criterions are numerical value
The probability for being distributed in (μ-σ, μ+σ) is 0.6526;The probability for being distributed in (μ -2 σ, μ+2 σ) is 0.9544;It is distributed in (μ -3 σ, μ+3
Probability σ) is 0.9974;
4) probability paper
Probability paper is a kind of graph paper made according to specific probability distribution, for each continuous distribution function, all
A kind of graph paper is designed, makes the distribution function in figure thereon be in straight line, using probability paper, according to sample to overall point
The type of cloth is tested, and distributed constant is estimated, and carries out other simple and rapid statistical inferences, is obeyed really just
State is distributed;
5) viscous section
For a certain wind speed time series, the integer part of the maximum of its wind speed adds 1 to be max, wind speed minimum value it is whole
Number part is min, the wind speed range min, max of this sequence is carried out into interval division, each siding-to-siding block length is 1m/s:[min, min+
1) ..., [max-1, max), the scope that each wind speed power sample is pressed to each wind speed interval carries out returning section;
Ask for step in each viscous section in section:
(1) DP of contained all sample points in the segment is obtained;
(2) probability density fitting is carried out to the DP in this section with normal distribution, and is entered with KS and normal distribution probability paper
Row normal distribution-test;
(3) μ under the fitting, σ are obtained;
With 3 σ rule definition viscosity sections [μ -3 σ, μ+3 σ], the σ of L=μ -3 σ, R=μ+3 are made, interval of definition [L, R] is the section
Viscous section, judge whether to abandon the index of wind, when the DP Bu Ci areas corresponding to the power of ultra-long time continuity sample point
Between in [L, R], just define the point to abandon wind.
The wind power of the present invention, which abandons wind data characteristic recognition method, has the advantages that scientific and reasonable, strong applicability, can
The wind data of abandoning of Wind turbines is rejected, ensures that actual wind speed power curve establishes precision.
Brief description of the drawings
Fig. 1 is that certain blower fan has the wind speed power scatterplot schematic diagram that two classes abandon wind point;
Fig. 2 is the wind speed power scatterplot schematic diagram before certain blower fan is rejected;
Fig. 3 is the wind speed power scatterplot schematic diagram after certain blower fan is rejected.
Embodiment
Wind data characteristic recognition method is abandoned below with drawings and examples to the wind power of the present invention to carry out specifically
It is bright.
Reference picture 1- Fig. 3, wind power of the invention are abandoned wind data characteristic recognition method, comprised the following steps:
1) data acquisition
By taking the wind power plant F of northeast as an example, the whole typhoon group of motors of field 177, single unit rated power is 1500KW, cuts wind
Speed is 3m/s, rated wind speed 10.5m/s, cut-out wind speed 25m/s, and through actual verification, wherein certain blower fan T is in August part whole month
Without wind is abandoned, its power swing is only determined by fluctuations in wind speed and its wind turbine internal conversion;
2) definition of progress of disease error
It is bent in calibration power under equal wind speed that progress of disease error (DP) refers to that the actual power corresponding to actual wind speed subtracts
Power under line,
DP=P (i)-f (s (i)) (1)
Wherein s (i) is the wind speed at the i-th moment, and P (i) is the actual wind power at the i-th moment, and f (s (i)) is based on now
The progress of disease power that the wind speed at quarter is obtained using calibration power curve;
3) 3 σ criterions
σ represents standard deviation in normal distribution, and μ represents average, and x=μ are the symmetry axis of image.3 σ criterions are numerical value
The probability for being distributed in (μ-σ, μ+σ) is 0.6526;The probability for being distributed in (μ -2 σ, μ+2 σ) is 0.9544;It is distributed in (μ -3 σ, μ+3
Probability σ) is 0.9974;
4) probability paper
Probability paper is a kind of graph paper made according to specific probability distribution, for each continuous distribution function, all
A kind of graph paper is designed, makes the distribution function in figure thereon be in straight line, using probability paper, according to sample to overall point
The type of cloth is tested, and distributed constant is estimated, and carries out other simple and rapid statistical inferences, for example, for
Sometime sequence, to examine its whether Normal Distribution, its normal distribution probability paper can be drawn, then examine scatterplot
Whether satisfaction is distributed in linear edge, and is substantially in straight line, and this time sequential test is qualified if meeting, obeys really just
State is distributed;
5) viscous section
For a certain wind speed time series, the integer part of the maximum of its wind speed adds 1 to be max, wind speed minimum value it is whole
Number part is min, the wind speed range min, max of this sequence is carried out into interval division, each siding-to-siding block length is 1m/s:[min, min+
1) ..., [max-1, max), the scope that each wind speed power sample is pressed to each wind speed interval carries out returning section.
Ask for step in each viscous section in section:
(1) DP of contained all sample points in the segment is obtained;
(2) probability density fitting is carried out to the DP in this section with normal distribution, and is entered with KS and normal distribution probability paper
Row normal distribution-test;
(3) μ under the fitting, σ are obtained;
With 3 σ rule definition viscosity sections [μ -3 σ, μ+3 σ], the σ of L=μ -3 σ, R=μ+3 are made, interval of definition [L, R] is the section
Viscous section (judging whether to abandon the index of wind), when the DP Bu Ci areas corresponding to the power of ultra-long time continuity sample point
Between in [L, R], just define the point to abandon wind.
The present invention embodiment it is not exhaustive, those skilled in the art without creative work simple copy
And improvement, the protection domains of the claims in the present invention should be belonged to.
Claims (1)
1. a kind of wind power abandons wind data characteristic recognition method, it is characterised in that it comprises the following steps:
1) data acquisition
The whole field Wind turbines of wind power plant are gathered, rated power, incision wind speed, rated wind speed and cut-out wind speed per single unit
Data, be chosen at the Wind turbines do not abandoned under wind state in some month, and the fluctuation of the power output of the Wind turbines is special
Property is caused by the fluctuation of wind speed and unit error component itself;
2) definition of progress of disease power error
Progress of disease power error refers to that the actual power corresponding to actual wind speed is subtracted under equal wind speed under calibration power curve
Power, DP=P (i)-f (s (i)) (1)
Wherein, DP is progress of disease power error, and s (i) is the wind speed at the i-th moment, and P (i) is the actual wind power at the i-th moment, f (s
(i)) the progress of disease power to be obtained based on the wind speed at this moment using calibration power curve;
3) 3 σ criterions σ in normal distribution represent standard deviation, and μ represents average, and x=μ are the symmetry axis of image, and 3 σ criterions are
Numeric distribution is 0.6526 in the probability of (μ-σ, μ+σ);The probability for being distributed in (μ -2 σ, μ+2 σ) is 0.9544;It is distributed in (μ -3
The σ of σ, μ+3) probability be 0.9974;
4) probability paper
Probability paper is a kind of graph paper made according to specific probability distribution, for each continuous distribution function, is all designed
A kind of graph paper, the distribution function is set in figure to be thereon in straight line, using probability paper, according to sample to overall distribution
Type is tested, and distributed constant is estimated, shows the certain Normal Distribution of air speed data through estimation of distribution parameters;
5) viscous section
For a certain wind speed time series, the integer part of the maximum of its wind speed adds 1 to be max, the integer portion of wind speed minimum value
It is divided into min, the wind speed range min, max of this sequence is subjected to interval division, each siding-to-siding block length is 1m/s:[min, min+
1) ..., [max-1, max), the scope that each wind speed power sample is pressed to each wind speed interval carries out returning section;
Ask for step in each viscous section of segment:
(1) DP of contained all sample points in the segment is obtained;
(2) probability density fitting is carried out to the DP in this section with normal distribution, and is examined with K-S with normal distribution probability paper to enter
Row normal distribution-test;
(3) μ under the fitting, σ are obtained;With the 3 viscous sections of σ rule definitions [μ -3 σ, μ+3 σ], the σ of L=μ -3 σ, R=μ+3 are made, it is fixed
Adopted section [L, R] is the viscous section of the segment, judges whether to abandon the index of wind, when the work(of ultra-long time continuity sample point
DP corresponding to rate just defines the point to abandon wind not in this section [L, R].
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510599490.6A CN105224733B (en) | 2015-09-19 | 2015-09-19 | The method that wind power abandons wind data feature recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510599490.6A CN105224733B (en) | 2015-09-19 | 2015-09-19 | The method that wind power abandons wind data feature recognition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105224733A CN105224733A (en) | 2016-01-06 |
CN105224733B true CN105224733B (en) | 2018-02-02 |
Family
ID=54993699
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510599490.6A Active CN105224733B (en) | 2015-09-19 | 2015-09-19 | The method that wind power abandons wind data feature recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105224733B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108734367A (en) * | 2017-04-25 | 2018-11-02 | 中国电力科学研究院 | It is a kind of to be used to calculate the method and system that wind power plant abandons wind-powered electricity generation amount |
CN107291927B (en) * | 2017-06-29 | 2019-11-05 | 同济大学 | A kind of air speed data cleaning method for Wind turbines wind speed correlation analysis |
CN107732962B (en) * | 2017-09-29 | 2020-12-01 | 国网辽宁省电力有限公司 | Abandoned wind reduction method based on ultra-short term abandoned wind curve prediction |
CN108415979B (en) * | 2018-02-09 | 2022-03-29 | 自然资源部第三海洋研究所 | Method for calculating microwave scattering count data search range in sea surface wind field inversion |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102664409A (en) * | 2012-05-14 | 2012-09-12 | 东北电力大学 | Real-time prediction calculation method based on measured data for wind power of wind power station |
CN103956773A (en) * | 2014-05-09 | 2014-07-30 | 河海大学 | Standby configuration optimization method adopting wind power system unit |
CN104201712A (en) * | 2014-08-10 | 2014-12-10 | 东北电力大学 | Wind power real-time prediction calculation method base on spatial average wind speed |
-
2015
- 2015-09-19 CN CN201510599490.6A patent/CN105224733B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102664409A (en) * | 2012-05-14 | 2012-09-12 | 东北电力大学 | Real-time prediction calculation method based on measured data for wind power of wind power station |
CN103956773A (en) * | 2014-05-09 | 2014-07-30 | 河海大学 | Standby configuration optimization method adopting wind power system unit |
CN104201712A (en) * | 2014-08-10 | 2014-12-10 | 东北电力大学 | Wind power real-time prediction calculation method base on spatial average wind speed |
Non-Patent Citations (1)
Title |
---|
基于运行数据的风电机组建模方法;苏勋文等;《电力系统保护与控制》;20100501;第38卷(第9期);第50-54页 * |
Also Published As
Publication number | Publication date |
---|---|
CN105224733A (en) | 2016-01-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105224733B (en) | The method that wind power abandons wind data feature recognition | |
CN111104981B (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
CN108375476B (en) | Hydroelectric generating set health assessment method | |
CN107016235B (en) | Equipment running state health degree evaluation method based on multi-feature adaptive fusion | |
CN110926809A (en) | Big data analysis-based wind turbine generator transmission chain fault early warning method | |
Cappugi et al. | Machine learning-enabled prediction of wind turbine energy yield losses due to general blade leading edge erosion | |
CN106897945A (en) | The clustering method and equipment of wind power generating set | |
CN111091298B (en) | Wind power plant flow field coupling characteristic evaluation and intelligent grouping method and system | |
CN104134013B (en) | Wind turbine blade modal analysis method | |
CN106529018A (en) | Fatigue crack growth prognostics method based on Gaussian weight-hybrid particle filtering | |
CN105930900A (en) | Method and system for predicting hybrid wind power generation | |
CN111237646B (en) | Automatic identification and positioning method for leakage of water supply pipe network | |
CN106611243A (en) | Residual correction method for wind speed prediction based on GARCH (Generalized ARCH) model | |
Cevasco et al. | O&M cost-Based FMECA: Identification and ranking of the most critical components for 2-4 MW geared offshore wind turbines | |
CN115526258A (en) | Power system transient stability evaluation method based on Spearman correlation coefficient feature extraction | |
CN115859099A (en) | Sample generation method and device, electronic equipment and storage medium | |
CN109376937B (en) | Self-adaptive scheduling end-of-term water level prediction method based on ensemble empirical mode decomposition | |
CN111412959A (en) | Flow online monitoring calculation method, monitor and monitoring system | |
CN113988398A (en) | Wind turbine generator power prediction method and device, electronic equipment and storage medium | |
CN114021610A (en) | Fan fault recognition model training method and system based on transfer learning | |
CN109376990B (en) | Method and system for determining critical removal time of power system based on Simese network model | |
Bao et al. | Iterative modeling of wind turbine power curve based on least‐square B‐spline approximation | |
CN114818815A (en) | Method and system for acquiring timed arrival time of blade tip for blade vibration measurement | |
CN114330141A (en) | Fan main shaft bearing service life prediction method based on GRU (generalized regression Unit) super-parameter optimization | |
CN110555543B (en) | Uncertainty analysis method and system for wind power plant prediction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |