CN111522808B - Abnormal operation data processing method for wind turbine generator - Google Patents
Abnormal operation data processing method for wind turbine generator Download PDFInfo
- Publication number
- CN111522808B CN111522808B CN202010358500.8A CN202010358500A CN111522808B CN 111522808 B CN111522808 B CN 111522808B CN 202010358500 A CN202010358500 A CN 202010358500A CN 111522808 B CN111522808 B CN 111522808B
- Authority
- CN
- China
- Prior art keywords
- data
- power
- wind speed
- wind
- abnormal
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 63
- 238000003672 processing method Methods 0.000 title description 2
- 238000000034 method Methods 0.000 claims abstract description 83
- 230000008030 elimination Effects 0.000 claims abstract description 13
- 238000003379 elimination reaction Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000004364 calculation method Methods 0.000 claims description 23
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 description 13
- 238000011160 research Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 6
- 238000010248 power generation Methods 0.000 description 4
- 238000012217 deletion Methods 0.000 description 3
- 230000037430 deletion Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- 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
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Quality & Reliability (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Wind Motors (AREA)
Abstract
The invention discloses a method for processing abnormal operation data of a wind turbine, which comprises the following steps: preprocessing data: for wind speed-power abnormal data with wind speed greater than cut-in wind speed and less than cut-out wind speed and power less than or equal to zero, eliminating the data of the part; the preprocessed wind speed-power abnormal data is firstly identified and removed by using a Thompson tau method; performing secondary identification and elimination on the wind speed-power abnormal data after the primary identification and elimination by using a quartile method; adopting a four-point interpolation subdivision algorithm to carry out interpolation reconstruction on the misdeleted and missing data; the method solves the technical problems that the abnormal operation data processing of the wind turbine generator is inapplicable under the condition of more wind power plant fan types, the dependence on available data of adjacent wind power plants is too high, the error is gradually increased along with the increase of the distance of the fans, the accuracy is reduced, the reconstruction result depends on the correlation degree of the wind power plants, the model solving is complex when the data quantity is more, and the like.
Description
Technical Field
The invention belongs to the technical field of power systems, and particularly relates to a method for processing abnormal operation data of a wind turbine.
Background
Along with the increasing of the wind power installation capacity in China in recent years, the country has corresponding policy support for wind power generation, the power generation cost is reduced, the technology is mature day by day, and in order to better utilize wind energy, the establishment of a high-efficiency and accurate wind power monitoring model is urgent. However, the randomness and the volatility of the wind power generation generate non-negligible influence on the safety and the stability of the power system, a large amount of abnormal data can be generated in the actual operation process of the wind turbine due to factors such as measurement, transmission, control, wind disposal and the like, the data subjected to interference and influence cannot reflect the actual performance of the wind turbine, and the data are required to be identified and removed.
The existing research identification cleaning method has the defects that the cleaning time is too long, engineering applicability is not realized when the number of wind turbine sets is large, the operation data of a plurality of fans are required to be continuously corrected for an equivalent power boundary line, the proportion of data elimination is high, the phenomenon of over-identification exists on some sets, and the normal data deletion is possibly excessive.
By identifying and eliminating the wind speed-power abnormal data, certain damage to the integrity of the operation data is unavoidable, and meanwhile, partial normal operation data can be deleted, so that the follow-up research application is not facilitated, and interpolation reconstruction is needed for the eliminated data. The existing research interpolation reconstruction method has the defects that the method is inapplicable under the condition of more wind power plant fan types, the dependence on available data of adjacent wind power plants is too high, the error is gradually increased along with the increase of the distance of the fans, the accuracy is reduced, the reconstruction result depends on the correlation degree of the wind power plants, the model solving is complex when the data quantity is more, and the like.
Disclosure of Invention
The invention aims to solve the technical problems that: the method for processing the abnormal operation data of the wind turbine generator is provided to solve the technical problems that in the prior art, the abnormal operation data processing of the wind turbine generator is inapplicable under the condition of more wind power plant fan types, the dependence on available data of adjacent wind power plants is too high, the error is gradually increased along with the increase of the distance of the wind power plants, the accuracy is reduced, the reconstruction result depends on the correlation degree of the wind power plants, the model solving is complex when the data quantity is more, and the like.
The technical scheme of the invention is as follows:
a method for processing abnormal operation data of a wind turbine generator comprises the following steps:
step S201, preprocessing data: for wind speed-power abnormal data with wind speed greater than cut-in wind speed and less than cut-out wind speed and power less than or equal to zero, eliminating the data of the part;
step S202, the preprocessed wind speed-power abnormal data is firstly identified and removed by utilizing a Thompson tau method;
step S203, performing secondary identification and elimination on the wind speed-power abnormal data after the primary identification and elimination by using a quartile method;
and S204, carrying out interpolation reconstruction on the data deleted by mistake and missing by adopting a four-point interpolation subdivision algorithm.
The power calculation method in step S201 is as follows:
wherein C is P Is the wind energy utilization coefficient; ρ 0 Is the reference air density; a is the swept area of the wind wheel; v is wind speed; p (P) e The actual power of the wind turbine generator system.
The method for first identifying and eliminating the preprocessed wind speed-power abnormal data by using the thompson tau method in the step S202 includes: firstly dividing wind speed into n sections according to the size, and recording a power data sample of an ith wind speed-power section as P i ＝[P i,1 ,P i,2 ,P i,3 ,…,P i,m ]Where i=1, 2, …, n, P i,1 ≤P i,2 ≤P i,3 ≤…≤P i,m ；
The method for carrying out the first recognition of the wind speed-power abnormal data by using the Thompson tau method comprises the following steps:
calculating the mean value of the power data in the interval
Wherein m is the number of power points in a wind speed-power interval; p (P) j J=1, 2, …, m for each power value in the interval;
the standard deviation of the power data in the calculation interval is as follows:
the absolute value of the data bias for each power sample is:
if delta is maximum, this is the outlier to be removed;
thompson tau values are calculated from the T distribution values of the power sample data:
wherein α=0.01;
when delta>When τ is S, the detected power value is an abnormal point; conversely, if delta<When τ is S, the detected power value is a normal point; using Thompson tau method, 1 outlier can be detected each time in each interval if delta j If the power value is detected as an abnormal value, the power value is removed, and the average value and the standard deviation are recalculated until no new power abnormal value is found.
The method for performing secondary identification and elimination on the wind speed-power abnormal data after the primary identification and elimination by using the quartile method in the step S203 comprises the following steps:
firstly, after the first recognition and elimination are recorded, the power data sample of the ith wind speed-power interval is P' i ＝[P' i,1 ,P' i,2 ,P' i,3 ,…,P' i,m ]Where i=1, 2, …, n, P' i,1 ≤P' i,2 ≤P' i,3 ≤…≤P' i,m The method comprises the steps of carrying out a first treatment on the surface of the Second oneQuantile M i Representing sample data P' i Is a median of (2);
the first, third quantile represents P' i Values represented by the positions of 25% of data points before and after the middle segmentation; when M is an even number, M i Will P' i The two subsequences with the same length are marked as P' i_1 ＝[P' i,1 ,P' i,2 ,…,P' i,(m-1)/2 ]And P' i_2 ＝[P' i,(m+1)/2 ,P' i,(m+3)/2 ,…,P' i,m ]，Q 1,i 、Q 3,i Respectively represent the subsequences P' i_1 And P' i_2 Is a median of (2);
when m=4k+3 (k=0, 1,2, …), the calculation formula is
When m=4k+1 (k=0, 1,2, …), the calculation formula is
From the calculation results of equation 6 and equation 7, P 'can be obtained' i Is of the quartile range IQR i Is that
IQR i ＝Q 3,i -Q 1,i
… … … … … … … (equation 8) in the quartile method, the upper limit W is used u,i Lower limit W d,i To eliminate abnormal value in the data, the calculation formula is that
According to the calculation result of formula 9, a power data sample P 'is defined' i Is at W u,i And W is d,i The data in between are normal values, and the data points outside are outliers.
In step S204, the method for performing interpolation reconstruction on the missing data by adopting the four-point interpolation subdivision algorithm includes: and carrying out missing data interpolation by searching four wind speed points closest to the wind speed to be interpolated and utilizing a four-point interpolation algorithm.
The formula of the four-point interpolation algorithm is as follows:
wherein P is 2i+1 The power point to be interpolated; p (P) i-1 ,P i ,P i+1 ,P i+2 The power values corresponding to four wind speeds closest to the power point to be interpolated; omega is tensor parameter, omega e [0,0.125 ]]。
Taking ω=1/16, and finishing to obtain:
order the
Substituting equation 12 into equation 11 yields:
second quantile M i The calculation formula of (2) is
Wherein m is the number of power points in a wind speed-power interval.
The invention has the beneficial effects that:
the method is based on Thompson tau-quartile method to identify and reject abnormal data, and then based on historical wind speed-power operation data of the wind turbine, interpolation and reconstruction are carried out on the misdelete data by adopting a four-point interpolation subdivision algorithm. Compared with the common treatment method, the proposed cleaning method is simple to realize and higher in cleaning efficiency, and the proposed reconstruction method can effectively improve the quality of reconstruction data under the condition that data of a nearby wind farm is missing and unavailable, is better in reconstruction effect, improves the integrity of the data to a certain extent, and is beneficial to reutilization of subsequent research; the method solves the technical problems that the prior art is not applicable to the abnormal operation data processing of the wind turbine generator under the condition of more wind power plant fan types, the dependence on available data of adjacent wind power plants is too high, the error is gradually increased along with the increase of the distance of the fans, the accuracy is reduced, the reconstruction result depends on the correlation degree of the wind power plants, and the model solving is complex when the data quantity is more.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
FIG. 2 is a flow chart of the abnormal data identification cleaning process of the present invention;
FIG. 3 shows the cleaning effect of the machine set data 1, 14 and 23 in the specific embodiment;
fig. 4 shows recognition results of abnormal data of the No. 1, no. 14 and No. 23 machine sets in the specific embodiment.
Detailed Description
The power curve of the wind turbine is the most visual and most commonly used description mode for the power characteristics of the wind turbine, accurately describes the correlation between the power and the wind speed of the wind turbine, and essentially reflects the performance of the wind turbine. The accurate actually measured wind power curve can provide important reference basis for wind turbine generator performance evaluation, wind power curve monitoring, wind power prediction, wind power plant numerical modeling and other works. In the actual running process of the wind turbine, because of the difference of external environments and control strategies, the power curve actually tested by the wind turbine has larger deviation from the standard power curve. When the power grid is in wind-abandoning electricity limitation, the abnormal data often show up as transverse pile-up data in a wind speed-power scatter diagram, the power is lower than the value in a normal running state, and scattered points distributed in a scattered manner are caused by data acquisition, measurement and the like. The data subjected to interference and influence cannot reflect the real performance of the wind turbine, the abnormal data are not identified and removed, and the research on the related problems of the subsequent wind turbine is seriously influenced, so that the research on the identification and removal method of the wind speed-power abnormal data of the wind turbine is very important.
Fig. 2 shows a method for processing abnormal operation data of a wind turbine based on Thompson tau-quartile method and multipoint interpolation, which comprises the following steps:
(1) Preprocessing data: for points with wind speeds greater than the cut-in wind speed and less than the cut-out wind speed and power less than or equal to zero, the points are represented as bottom accumulation type abnormal data clusters in a wind speed-power scatter diagram, and the data of the points are removed.
(2) The Thompson tau method is utilized to identify and reject abnormal data for the first time.
(3) And performing secondary identification and eliminating abnormal data by using a quartile method.
(4) And carrying out interpolation reconstruction on the erroneously deleted data by adopting a four-point interpolation subdivision algorithm based on the historical wind speed-power operation data of the wind turbine.
Step S101, firstly dividing the wind speed into n sections according to a certain size, and recording the power data sample of the ith wind speed-power section as P i ＝[P i,1 ,P i,2 ,P i,3 ,…,P i,m ]Where i=1, 2, …, n, P i,1 ≤P i,2 ≤P i,3 ≤…≤P i,m The first recognition of wind speed-power anomaly data using Thompson tau method is as follows:
calculating the mean value of the power data in the interval
Wherein m is the number of power points in a wind speed-power interval; p (P) j J=1, 2, …, m for each power value in the interval.
The standard deviation of the power data in the calculation interval is as follows:
the absolute value of the data bias for each power sample is:
if delta is the largest, this is likely to be the outlier to be culled; d, d j Representing the absolute value of the data bias for each power sample.
Thompson tau values are calculated from the T distribution values of the power sample data:
wherein m is the number of power data samples in the interval; t is a t distribution value of the power sample data, alpha is a significance level, the value of the t distribution value influences the abundance of the power sample data, and the invention rejects the level alpha=0.01.
When delta>When τ is S, the detected power value is an abnormal point; conversely, if delta<And τ×s, the detected power value is a normal point. Using Thompson tau method, 1 outlier can be detected each time in each interval if delta j If the power value is detected as an abnormal value, the power value is removed, and the average value and the standard deviation are recalculated until no new power abnormal value is found.
Step S102, performing secondary identification on the residual data after the primary identification by using a quartile method. Firstly, after the first recognition is recorded, the power data sample of the ith wind speed-power interval is P' i ＝[P' i,1 ,P' i,2 ,P' i,3 ,…,P' i,m ]Where i=1, 2, …, n, P' i,1 ≤P' i,2 ≤P' i,3 ≤…≤P' i,m . Second quantile M i Representing sample data P' i The median of (2) is calculated as
Wherein m is the number of power points in a wind speed-power interval.
The first, third quantile represents P' i The position of the 25% data points before and after the middle segmentation. The total amount m of data points in the interval is different, and the calculation formulas are slightly different.
When M is an even number, M i Will P' i The two subsequences with the same length are marked as P' i_1 ＝[P' i,1 ,P' i,2 ,…,P' i,(m-1)/2 ]And P' i_2 ＝[P' i,(m+1)/2 ,P' i,(m+3)/2 ,…,P' i,m ]，Q 1,i 、Q 3,i Respectively represent the subsequences P' i_1 And P' i_2 Is a median of (a).
When m=4k+3 (k=0, 1,2, …), the calculation formula is
When m=4k+1 (k=0, 1,2, …), the calculation formula is
From the calculation results of equation 6 and equation 7, P 'can be obtained' i Is of the quartile range IQR i Is that
IQR i ＝Q 3,i -Q 1,i
… … … … … … … (equation 8)
In the quartile method, the upper limit W is used u,i Lower limit W d,i To eliminate abnormal value in the data, the calculation formula is that
According to the calculation result of formula 9, a power data sample P 'is defined' i Is at W u,i And W is d,i The data in between are normal values, and the data points outside are outliers.
Step S103, after abnormal data is identified and removed by the Thompson tau-quartile method, normal data which is deleted by mistake may exist, and the integrity of the data is destroyed, so that the cleaned data needs to be subjected to data reconstruction. The four-point interpolation subdivision algorithm uses four adjacent control points to calculate a new control point, using the same algorithm every time it "splits".
The method has theoretical feasibility by searching four wind speed points similar to the wind speed to be interpolated and utilizing a four-point interpolation algorithm to interpolate missing data. The specific formula of the four-point interpolation algorithm is as follows:
wherein P is 2i+1 The power point to be interpolated; p (P) i-1 ,P i ,P i+1 ,P i+2 The power values corresponding to four wind speeds closest to the power point to be interpolated; omega is tensor parameter, when omega is E [0,0.125 ]]In this case, a more satisfactory interpolation point can be obtained, and the invention takes the interval midpoint ω=1/16, and has the best Holder regularities.
And (3) finishing to obtain:
order the
Substituting equation 12 into equation 11 yields:
in order to verify the effectiveness of the method and the process for identifying and reconstructing the wind speed-power abnormal data of the provided wind turbine, taking a wind turbine of a certain domestic wind farm as an example, 10 min-level wind speed-power data from 8 months in 2017 to 7 months in 2018 are taken for analysis. The cut-in wind speed of the wind turbine generator is 3.5m/s, the rated wind speed is 12m/s, the cut-out wind speed is 25m/s, the rated power is 2000kW, the 1, 14 and 23-number wind turbine generator is taken as a research object, and the test provides a method for identifying and cleaning the wind speed-power abnormal data of each wind turbine generator.
The result of cleaning the wind speed-power abnormal data of the No. 1, no. 14 and No. 23 units by using the Thompson tau method is shown as a figure, and the identified and removed abnormal data is shown as a figure 4.
As can be seen from fig. 4, the Thompson tau-quartile method can be used for well identifying and eliminating the abnormal wind speed-power data of 3 units, so that the operation data of the units in the conventional power generation state are completely reserved, the abnormal data caused by other reasons are eliminated, an effective research data set can be provided for the subsequent analysis of the actual operation state of the fan, and the practicability is high. To further illustrate the greater engineering applicability of the present disclosure, the method is compared with the existing variable point grouping-quartile method in terms of data deletion rate and cleaning time, and the recognition and comparison effects of abnormal data are shown in table 1. As can be seen from table 1, the method in this document is equivalent to the variable-point grouping-quartile method in terms of abnormal data deletion rate, but has a great improvement in cleaning efficiency, and the cleaning time of three units is about 1 s. In practical application, the number of wind turbines in the wind power plant is often more, and the normal operation data set of each wind turbine in the wind power plant can be obtained more quickly by using the method, so that the method is more suitable for popularization and application in practical engineering.
Table 1 data cleaning effect of different methods
As can be seen from fig. 4, performing the wind speed-power abnormal data identification cleaning inevitably causes the erroneous erasure of part of the normal data, so that the integrity of the normal data is destroyed. Therefore, the rejected data needs to be reconstructed to restore the integrity of the data.
And deleting part of wind speed-power data in a 10-min-level wind speed and power data sequence acquired by the No. 1 wind turbine generator, simulating the missing condition of the power data, and reconstructing the missing power data to verify the effectiveness of the method. Average relative error E is selected MRE And the reconstruction accuracy r is used as an evaluation index of data reconstruction.
Wherein P is ip The reconstructed power at the moment i; p (P) i The actual power at the moment i; p (P) N Rated power of the wind turbine generator; n is the number of missing power data.
10 min-level wind speed-power data of 3 th month, 3 th month and 0 th day of 2017 to 6 th month and 0 th day of No. 1 fans are selected, and 433 wind speed-power data are taken in total. 50, 100, 200 wind speed-power data were randomly deleted, data missing conditions were simulated, and the methods herein were compared to piecewise linear interpolation, cubic spline interpolation, and cubic Hermite interpolation. The results of the reconstitution comparisons are shown in Table 2.
Table 2 comparison of reconstruction methods
As can be seen from table 2, the data reconstruction method presented herein is superior to the remaining three methods in terms of average relative error and reconstruction accuracy in the absence of wind speed-power data 50, 100, 200. The piecewise linear interpolation effect is closer to the method, the cubic Hermite interpolation is inferior, and the cubic spline interpolation reconstruction effect is the worst. The method reconstruct data based on single machine historical operation data, does not depend on operation data of other wind power plants or wind turbines, and has certain practical engineering applicability.
Claims (5)
1. A method for processing abnormal operation data of a wind turbine generator comprises the following steps:
step S201, preprocessing data: for wind speed-power abnormal data with wind speed greater than cut-in wind speed and less than cut-out wind speed and power less than or equal to zero, eliminating the data of the part;
step S202, the preprocessed wind speed-power abnormal data is firstly identified and removed by utilizing a Thompson tau method;
the method for first identifying and eliminating the preprocessed wind speed-power abnormal data by using the thompson tau method in the step S202 includes: firstly dividing wind speed into n sections according to the size, and recording a power data sample of an ith wind speed-power section as P i ＝[P i,1 ,P i,2 ,P i,3 ,…,P i,m ]Where i=1, 2, …, n, P i,1 ≤P i,2 ≤P i,3 ≤…≤P i,m ；
The method for carrying out the first recognition of the wind speed-power abnormal data by using the Thompson tau method comprises the following steps: calculating the mean value of the power data in the interval
Wherein m is the number of power points in a wind speed-power interval; p (P) j J=1, 2, …, m for each power value in the interval;
the standard deviation of the power data in the calculation interval is as follows:
the absolute value of the data bias for each power sample is:
if delta is maximum, this is the outlier to be removed;
thompson tau values are calculated from the T distribution values of the power sample data:
wherein α=0.01;
when (when)When the detected power value is an abnormal point; on the contrary, if->When the detected power value is the normal point; using Thompson tau method, 1 outlier can be detected each time in each interval if delta j If the power value is detected as an abnormal value, the power value is removed, and the average value and the standard deviation are recalculated until no new power abnormal value is found;
step S203, performing secondary identification and elimination on the wind speed-power abnormal data after the primary identification and elimination by using a quartile method;
step S204, adopting a four-point interpolation subdivision algorithm to carry out interpolation reconstruction on the data deleted by mistake and missing data;
in step S204, the method for performing interpolation reconstruction on the missing data by adopting the four-point interpolation subdivision algorithm includes: the method comprises the steps of carrying out missing data interpolation by searching four wind speed points closest to a wind speed to be interpolated and utilizing a four-point interpolation algorithm;
the formula of the four-point interpolation algorithm is as follows:
wherein P is 2i+1 The power point to be interpolated; p (P) i-1 ,P i ,P i+1 ,P i+2 The power values corresponding to four wind speeds closest to the power point to be interpolated; omega is tensor parameter, omega e [0,0.125 ]]。
2. The method for processing abnormal operation data of a wind turbine generator according to claim 1, wherein the method comprises the following steps: taking ω=1/16, and finishing to obtain:
order the
Substituting equation 12 into equation 11 yields:
3. the method for processing abnormal operation data of a wind turbine generator according to claim 1, wherein the method comprises the following steps: the power calculation method in step S201 is as follows:wherein C is P Is the wind energy utilization coefficient; ρ 0 Is the reference air density; a is the swept area of the wind wheel; v is wind speed; p (P) e The actual power of the wind turbine generator system.
4. The method for processing abnormal operation data of a wind turbine generator according to claim 1, wherein the method comprises the following steps: the method for performing secondary identification and elimination on the wind speed-power abnormal data after the primary identification and elimination by using the quartile method in the step S203 comprises the following steps:
firstly, after the first recognition and elimination are recorded, the power data sample of the ith wind speed-power interval is P' i ＝[P' i,1 ,P' i,2 ,P' i,3 ,…,P' i,m ]Where i=1, 2, …, n, P' i,1 ≤P' i,2 ≤P' i,3 ≤…≤P' i,m The method comprises the steps of carrying out a first treatment on the surface of the Second quantile M i Representing sample data P' i Is a median of (2);
the first, third quantile represents P' i Values represented by the positions of 25% of data points before and after the middle segmentation; when M is an even number, M i Will P' i The two subsequences with the same length are marked as P' i_1 ＝[P' i,1 ,P' i,2 ,…,P' i,(m-1)/2 ]And P' i_2 ＝[P' i,(m+1)/2 ,P' i,(m+3)/2 ,…,P' i,m ]，Q 1,i 、Q 3,i Respectively represent the subsequences P' i_1 And P' i_2 Is a median of (2);
when m=4k+3 (k=0, 1,2, …), the calculation formula is
When m=4k+1 (k=0, 1,2, …), the calculation formula is
From the calculation results of equation 6 and equation 7, P 'can be obtained' i Is of the quartile range IQR i Is that
IQR i ＝Q 3,i -Q 1,i
… … … … … … … (equation 8)
In the quartile method, the upper limit W is used u,i Lower limit W d,i To eliminate abnormal value in the data, the calculation formula is that
According to the calculation result of formula 9, a power data sample P 'is defined' i Is at W u,i And W is d,i The data in between are normal values, and the data points outside are outliers.
5. The method for processing abnormal operation data of a wind turbine generator according to claim 1, wherein the method comprises the following steps: second quantile M i The calculation formula of (2) is
Wherein m is the number of power points in a wind speed-power interval.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010358500.8A CN111522808B (en) | 2020-04-29 | 2020-04-29 | Abnormal operation data processing method for wind turbine generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010358500.8A CN111522808B (en) | 2020-04-29 | 2020-04-29 | Abnormal operation data processing method for wind turbine generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111522808A CN111522808A (en) | 2020-08-11 |
| CN111522808B true CN111522808B (en) | 2023-07-28 |
Family
ID=71903278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010358500.8A Active CN111522808B (en) | 2020-04-29 | 2020-04-29 | Abnormal operation data processing method for wind turbine generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111522808B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112362179A (en) * | 2020-11-05 | 2021-02-12 | 甄十信息科技（上海）有限公司 | Method and device for detecting temperature of back cover of watch case of wearable device |
| CN113536198B (en) * | 2021-07-13 | 2022-12-13 | 中国华能集团清洁能源技术研究院有限公司 | System and method for identifying abnormal scattered points of power curve of wind turbine generator |
| CN113569399A (en) * | 2021-07-20 | 2021-10-29 | 华能四平风力发电有限公司 | Wind turbine generator operation data processing method, system, equipment and storage medium |
| CN115951088B (en) * | 2023-03-10 | 2023-08-25 | 南京南自华盾数字技术有限公司 | Wind turbine anemometer anomaly analysis method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008217280A (en) * | 2007-03-02 | 2008-09-18 | Toshiba Corp | Manufacturing method of product, and process management program |
| CN106014858A (en) * | 2016-07-21 | 2016-10-12 | 浙江运达风电股份有限公司 | Automatic calibration method and device for air aligning errors of wind generation set |
| CN107330183A (en) * | 2017-06-29 | 2017-11-07 | 华北电力大学 | A kind of wind power utilization computational methods based on service data |
| CN107885959A (en) * | 2017-12-06 | 2018-04-06 | 华北电力大学 | A kind of wind-powered electricity generation modeling and performance estimating method based on confidence equivalent power curve belt |
| CN108171400A (en) * | 2017-12-06 | 2018-06-15 | 浙江大学 | A kind of power of fan curve data preprocess method based on abnormal point and outlier detection |
| CN108412710A (en) * | 2018-01-30 | 2018-08-17 | 同济大学 | A kind of Wind turbines wind power data cleaning method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10261839B2 (en) * | 2016-11-02 | 2019-04-16 | International Business Machines Corporation | Outlier and root cause determination of excessive resource usage in a virtual machine environment |
-
2020
- 2020-04-29 CN CN202010358500.8A patent/CN111522808B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008217280A (en) * | 2007-03-02 | 2008-09-18 | Toshiba Corp | Manufacturing method of product, and process management program |
| CN106014858A (en) * | 2016-07-21 | 2016-10-12 | 浙江运达风电股份有限公司 | Automatic calibration method and device for air aligning errors of wind generation set |
| CN107330183A (en) * | 2017-06-29 | 2017-11-07 | 华北电力大学 | A kind of wind power utilization computational methods based on service data |
| CN107885959A (en) * | 2017-12-06 | 2018-04-06 | 华北电力大学 | A kind of wind-powered electricity generation modeling and performance estimating method based on confidence equivalent power curve belt |
| CN108171400A (en) * | 2017-12-06 | 2018-06-15 | 浙江大学 | A kind of power of fan curve data preprocess method based on abnormal point and outlier detection |
| CN108412710A (en) * | 2018-01-30 | 2018-08-17 | 同济大学 | A kind of Wind turbines wind power data cleaning method |
Non-Patent Citations (3)
| Title |
|---|
| 基于时空相关性的大规模风电功率短期预测方法研究;赵永宁;《中国博士学位论文全文数据库工程科技Ⅱ辑》;20190215(第2期);全文 * |
| 我国医药上市公司财务欺诈信息辨识研究 ——基于几种不同方法的比较;申晴;《中国硕士学位论文全文经济与管理科学辑》;20180315(第3期);全文 * |
| 风电场输出功率异常数据识别与重构方法研究;朱倩雯等;《电力***保护与控制》;20150201;第43卷(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111522808A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111522808B (en) | Abnormal operation data processing method for wind turbine generator | |
| CN106570790B (en) | Wind power plant output data restoration method considering wind speed data segmentation characteristics | |
| CN105205569B (en) | State of fan gear box online evaluation method for establishing model and online evaluation method | |
| CN107944175B (en) | Method for acquiring real power curve of fan by considering wind turbulence intensity | |
| CN107728059B (en) | Pitch system state evaluation method | |
| CN109918364B (en) | Data cleaning method based on two-dimensional probability density estimation and quartile method | |
| CN111275570A (en) | Wind turbine generator set power abnormal value detection method based on iterative statistics and hypothesis test | |
| CN111461564A (en) | Wind turbine generator power characteristic evaluation method based on cloud model and optimal combined weighting | |
| CN113626767A (en) | Fan power limit identification method, system, equipment and storage medium | |
| CN114968999A (en) | Abnormal data cleaning method based on time sequence matching and bidirectional quartile algorithm | |
| CN111950131B (en) | Wind power plant output equivalent aggregation model construction method considering electricity limiting factors | |
| CN117290666A (en) | Photovoltaic abnormal power data cleaning method | |
| CN116301652A (en) | Wind turbine generator online monitoring historical data cold-hot separation system | |
| CN116050072A (en) | Wind turbine theoretical power curve identification method and device based on random sampling consistency | |
| CN115271423A (en) | Wind generating set running state evaluation method based on hybrid similarity | |
| Bao et al. | Iterative modeling of wind turbine power curve based on least‐square B‐spline approximation | |
| CN114912807A (en) | Method and system for evaluating generated energy improving effect of technically improved wind turbine generator | |
| CN114880790A (en) | Variable working condition mechanism modeling method and system for steam-driven induced draft fan and storage medium | |
| CN111738556B (en) | Method for evaluating power generation capacity of fan based on machine head microclimate | |
| CN114065618A (en) | Method for fitting fan power curve based on parameter model of differential evolution | |
| CN111242411B (en) | Wind turbine generator power characteristic evaluation method and system | |
| CN114510815A (en) | Wind turbine generator power characteristic evaluation method and system | |
| CN113848347A (en) | Health state detection method for wind meter of wind driven generator | |
| CN113378102A (en) | Data missing preprocessing method, medium and application for short-term load prediction | |
| CN114547829A (en) | Theoretical power calculation method and system for wind power plant |
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 |