CN115408860A - Abnormal value monitoring and correcting method for wind speed-power data of wind power plant - Google Patents
Abnormal value monitoring and correcting method for wind speed-power data of wind power plant Download PDFInfo
- Publication number
- CN115408860A CN115408860A CN202211050136.4A CN202211050136A CN115408860A CN 115408860 A CN115408860 A CN 115408860A CN 202211050136 A CN202211050136 A CN 202211050136A CN 115408860 A CN115408860 A CN 115408860A
- Authority
- CN
- China
- Prior art keywords
- data
- power
- wind speed
- wind
- margin
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/06—Wind turbines or wind farms
Abstract
The invention relates to a wind power plant wind speed-power abnormal data monitoring and correcting method, which divides historical data into a plurality of regions, discards farther outliers based on data point density before the application of the original maximum value principle method, namely discards points with smaller density, and fits the reserved data points to obtain a wind speed-power fitting curve, thereby improving the accuracy of fitting the wind speed-power curve and effectively avoiding the problem caused by overlarge error due to small data volume. Certain allowances are set in the constant torque operation stage and the constant power operation stage respectively, so that the applicability is improved, excessive normal data points are prevented from being lost due to the fitting of a curve, and abnormal data are marked; the identification and monitoring of the abnormal values of the wind speed-power data of the wind power plant are realized.
Description
Technical Field
The invention relates to the field of wind power generation, in particular to a method for monitoring and correcting abnormal values of wind speed-power data of a wind power plant.
Background
The safe operation and scheduling optimization of the wind power plant need accurate and effective wind speed-power data to provide support. Due to the influence of factors such as environmental influence, sensor failure, wind and electricity abandonment, unplanned shutdown and the like, abnormal data are difficult to avoid, can directly influence the precision of a prediction model, and have great influence on the evaluation of the operating condition of a wind power plant.
The establishment of the mathematical model of the wind power curve is an effective method for realizing the cleaning of the wind speed-power operation data of the wind turbine generator. The wind power curve is the basis for analysis by many wind generating sets and describes the relationship between wind speed and the output power of the set. The method is not only an important basis for designing a wind turbine generator control system, but also important indexes for checking the generating performance of the wind turbine generator and the operating condition of a wind power plant.
The current parameter methods for modeling the wind power curve include a piecewise linear method, a polynomial power curve method, a maximum value principle method, a dynamic power curve method, a probability model method and the like. The methods have respective advantages and disadvantages, for example, the maximum value principle method model is simple, but the value of the point is excessively estimated from the transition region of the fitting curve to the rated power, so that the accuracy is influenced, and the evaluation effect of the operating condition of the wind power plant is influenced.
Disclosure of Invention
Aiming at the defects of the prior art, the invention aims to solve the technical problem of providing the abnormal value monitoring and correcting method of the wind power plant wind speed-power data, and the method can improve the accuracy of monitoring the abnormal value and further improve the accuracy of the data.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a wind power plant wind speed-power abnormal data monitoring and correcting method comprises the following steps:
discretizing historical wind speed data and power data, dividing the discretized historical wind speed data and power data into m1 wind speed intervals and m2 power intervals, wherein one wind speed interval corresponds to m2 power intervals and forms m1xm2 areas together, and counting the number of wind speed-power data in each area respectively to obtain a wind speed-power joint frequency distribution histogram; obtaining the density of the wind speed-power data of each area according to the frequency distribution in each area, and discarding the wind speed-power data of the area with the density of the wind speed-power data below 0.00001; for the reserved wind speed-power data, the maximum power value in each area is obtained, meanwhile, the maximum wind speed value in each wind speed area is obtained, one maximum wind speed value corresponds to the maximum power values of m2 areas, the maximum power values in the areas are sequenced according to the maximum wind speed value, an array P _ v of the maximum wind speed value-the maximum power value is obtained, and an array interval is formed between the adjacent maximum wind speed values;
interpolating an array interval of the array P _ v by adopting a Piecewise Cubic Hermite Interpolation Polynomial (PCHIP) mode, setting the interpolation quantity, and supplementing wind speed-power data; fitting the supplemented wind speed-power data by using a growth curve function to obtain a wind power curve upper boundary and a function expression thereof, wherein the wind power curve upper boundary and the function expression thereof are the wind speed-power curve model obtained according to a maximum value principle method;
primary screening: performing primary abnormal marking on data smaller than the cut-in wind speed in the supplemented wind speed-power data, marking the primary abnormal data as 1, and directly rejecting the data;
and (4) screening allowance: the remaining wind speed-power data with the abnormal data of the primary screen removed is divided into two stages for processing: in the constant torque operation stage and the constant power operation stage, corresponding allowances are respectively set in each stage to enable the allowances to be in line with the distribution characteristics of a standard wind speed-power curve of the corresponding stage, then a function expression of an upper boundary of a wind power curve determined by a growth curve function is used for obtaining corresponding power data in each stage by using a known wind speed, the upper boundary of the wind speed-power value of each stage is respectively obtained, and the upper boundary of the wind speed-power value is translated to the right to determine the lower boundary of the wind speed-power value; finally, adding corresponding margins to the upper boundary and the lower boundary of each stage respectively, wherein the data in the margin range are normal data, the data outside the margin range are margin abnormal data, the margin abnormal data are also marked as 1, all residual wind speed-power data with the removed primary screened abnormal data are divided into wind power normal data and margin abnormal data after primary screening of the margins, and the wind power normal data are marked as 0;
and (3) abnormal data correction: carrying out correlation analysis on normal wind power data of all fans of the wind power plant by using a Pearson correlation analysis method to obtain a correlation coefficient matrix between the fans; screening out the fan with the maximum correlation number in the matrix for the fan with the data to be supplemented, matching the screened fan data with the abnormal data corresponding to the fan with the data to be supplemented, and if the screened fan data is marked as 0, correcting and supplementing the margin abnormal data according to the normal data of the wind speed-wind power of the fan with the maximum correlation coefficient; and if the screened fan data is marked as 1, correcting and supplementing the corresponding data of the fan with the data to be supplemented.
A wind power plant wind speed-power abnormal data monitoring and correcting method specifically comprises the following steps:
the method comprises the following steps: generating arrays of wind speed-power maximums
1-1, arranging historical wind speed data and power data from small to large, performing discretization, and dividing the data into m1 wind speed intervals with the interval size of 1m/s and m2 power intervals with the interval size of 100 kw; counting the frequency of data points in each interval by taking the interval divided by the wind speed as a standard;
respectively counting the number N of wind speed-power data in each area of m1xm2 areas, wherein N = { N1, N2, N3 · · · · · } generates a wind speed-power joint frequency distribution histogram;
1-2, obtaining the wind speed-power density rho of the corresponding area according to the formula (1);
in the formula (1), N is the total number of the historical wind speed data, and N is the number of the wind speed-power data in each area, namely the area frequency;
1-3, testing the density of the wind speed-power in each area, and if the density rho of the wind speed-power is less than 0.00001, discarding the area; otherwise, the area is reserved;
1-4, respectively finding the maximum power value and the maximum wind speed value of each area in the data which are tested to meet the density requirement, combining the maximum wind speed values and the maximum power values of all the areas to obtain an array P _ v of the maximum wind speed value-the maximum power value, wherein an array interval is formed between the adjacent maximum wind speed values;
step two: growth curve function fitting wind speed-power curve
2-1, carrying out interpolation on each array interval of the array P _ v by using a piecewise cubic Hermite interpolation polynomial to supplement wind speed-power data;
formula (2) is an expression of a cubic Hermite interpolation polynomial, where x 0 ，x 1 The positions of two adjacent points of the point to be interpolated, namely the endpoint wind speed value of each array interval; y is 0 ，y 1 Corresponding to the argument x 0 ，x 1 The dependent variable of (a), namely the maximum power value corresponding to the endpoint wind speed; y' 0 And y' 1 Is the corresponding derivative, x is the wind speed;
2-2, fitting the supplemented wind speed-power data by using a growth curve function to obtain an upper boundary of a wind power curve and a functional expression thereof, wherein the upper boundary of the wind power curve and the functional expression thereof are shown in a formula (3);
wherein a, b and K are parameters of a pierce model; y represents power, x represents wind speed;
step three: abnormal data marking
3-1 primary screening: the cut-in wind speed is 3m/s, data smaller than the cut-in wind speed are subjected to primary abnormal marking in supplemented wind speed-power data, the primary abnormal data are marked as 1, and the data are directly removed;
3-2 screening allowance: the remaining wind speed-power data with the abnormal data of the primary screen removed is divided into two stages for processing: in the constant torque operation stage and the constant power operation stage, corresponding allowances are respectively set in each stage to enable the allowances to be in line with the distribution characteristics of a standard wind speed-power curve of the corresponding stage, then a function expression of an upper boundary of a wind power curve determined by a growth curve function is used for obtaining corresponding power data in each stage by using a known wind speed, the upper boundary of the wind speed-power value of each stage is respectively obtained, and the upper boundary of the wind speed-power value is translated to the right to determine the lower boundary of the wind speed-power value; finally, adding corresponding margins to the upper boundary and the lower boundary of each stage respectively, wherein the data in the margin range are normal data, the data outside the margin range are margin abnormal data, the margin abnormal data are also marked as 1, all residual wind speed-power data with the removed primary screened abnormal data are divided into wind power normal data and margin abnormal data after primary screening of the margins, and the wind power normal data are marked as 0;
step four: correcting and supplementing abnormal data to wind speed data
Establishing a wind speed correlation coefficient matrix of all fans of the wind power plant by using normal wind power data according to a Pearson correlation analysis method, screening out the fan with the maximum correlation number in the matrix for the fan with data to be supplemented, matching the screened fan data with abnormal data corresponding to the fan with the data to be supplemented, and correcting and supplementing the margin abnormal data according to the normal wind speed-wind power data of the fan with the maximum correlation coefficient if the screened fan data is marked as 0; and if the screened fan data is marked as 1, correcting and supplementing the corresponding data of the fan with the data to be supplemented.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, historical data are divided into a plurality of regions, far outliers are abandoned based on the density of data points before the application of the original maximum value principle method, namely, points with low density are abandoned, the reserved data points are fitted to obtain the wind speed-power fitting curve, the accuracy of the fitting wind speed-power curve is improved, and the problem caused by overlarge errors caused by small data volume is effectively avoided.
The method of the invention respectively sets certain margins at the constant torque operation stage and the constant power operation stage, thereby increasing the applicability, avoiding losing too many normal data points due to fitting curves, and marking abnormal data; the identification and monitoring of the abnormal values of the wind speed-power data of the wind power plant are realized.
In the method, the distribution characteristics of the standard wind speed-power curve are similar to those of the growth curve, and the method creatively adopts a growth curve function fitting mode to carry out wind speed power fitting to describe the relation between the wind speed and the power, thereby improving the fitting precision. Meanwhile, a Pearson correlation analysis method is adopted to carry out correlation analysis on the wind speed data of different fans, and the data of the abnormal fan is corrected by the data of the normal fan with high correlation, so that the high-precision correction of the data is realized.
Drawings
Fig. 1 is a discretization of historical wind speed data and power data, which are divided into n wind speed intervals and m power intervals, and the number of wind speed-power data in each area is respectively counted to obtain a wind speed-power combined frequency distribution histogram.
FIG. 2 is a top and bottom boundary graph of a wind speed-power curve with all historical data obtained according to the maximum principle after discarding historical data points with a density below 0.00001.
FIG. 3 is a scatter plot of wind speed-power data using cubic polynomial interpolation for each region.
FIG. 4 is a graph of the upper boundary of a normal wind speed-power curve obtained by fitting a growth curve function.
Fig. 5 is a wind speed-power scatter diagram with marks obtained after certain margins are set in the constant torque operation stage and the constant power operation stage, wherein black scatter points are original data, and gray scatter points are normal wind speed-power data.
Fig. 6 is a wind speed broken line graph obtained after the wind speed data of the normal fan with high correlation degree replaces the data of the abnormal fan at the same time, the broken line is the wind speed data before supplement, and the circle is marked as the supplement wind speed data.
Detailed Description
The technical solution of the present invention is further specifically described below by examples and drawings, but the scope of the present invention is not limited thereto.
Abnormal data in the wind power plant wind speed-power abnormal data monitoring and correcting method can be divided into three types, wherein the first type is data with high wind speed and zero power caused by sensor failure or shutdown; the second type is data caused by abandoned wind or faults; the third category is data with very low wind speed and high power caused by sensor failure or communication error. And the second type of abnormal data is abundant. The third type of abnormal data can be primarily screened and removed through cut-in wind speed, and the first type of abnormal data and the second type of abnormal data are screened and marked in a margin setting mode.
The embodiment is a method for monitoring and correcting wind speed-power abnormal data of a wind power plant, which comprises the following steps:
the method comprises the following steps: generating arrays of wind speed-power maximums
1-1, arranging historical wind speed data and power data from small to large, performing discretization, and dividing the data into m1 wind speed intervals with the interval size of 1m/s and m2 power intervals with the interval size of 100 kw; the two intervals are divided respectively and are not related to each other; counting the frequency of data points in each interval by taking the interval divided by the wind speed as a standard;
respectively counting the number N of wind speed-power data in each area of m1xm2 areas, wherein N = { N1, N2, N3 · · · · · } generates a wind speed-power joint frequency distribution histogram;
1-2, further obtaining the density distribution of the corresponding wind speed-power;
in the formula (1), N is the total number of the historical wind speed data, and N is the number of the wind speed-power data in each region, namely the region frequency.
1-3, detecting the density of wind speed-power in each area, and if the density of wind speed-power scattered points is less than 0.00001, discarding the points; otherwise, the point is retained.
1-4 respectively finding the maximum power value and the maximum wind speed value of each area in the data which are tested to meet the density requirement, combining the maximum wind speed values and the maximum power values of all the areas to obtain an array P _ v of the maximum wind speed value-the maximum power value, and forming an array interval between the adjacent maximum wind speed values.
Step two: growth curve function fitting wind speed-power curve
2-1, carrying out interpolation in each array interval of the arrays P _ v by using a segmented cubic Hermite interpolation polynomial (PCHIP) to supplement wind speed-power data;
formula (2) is an expression of a cubic Hermite interpolation polynomial, where x 0 ，x 1 The positions of two adjacent points of the point to be interpolated, namely the endpoint wind speed value of each array interval; y is 0 ，y 1 Corresponding to the argument x 0 ，x 1 The dependent variable of (a), namely the maximum power value corresponding to the endpoint wind speed; y' 0 And y' 1 Is the corresponding derivative, x is the wind speed; in different regions y within the same array interval 0 ，y 1 The value of (c) is different.
2-2, fitting the supplemented wind speed-power data by using a growth curve function to obtain an upper boundary of a wind power curve and a function expression thereof, wherein the upper boundary of the wind power curve and the function expression thereof are shown in a formula (3), and the upper boundary and the function expression thereof are a wind speed-power curve model obtained according to a maximum value principle method;
the formula (3) is a Pear growth curve model, wherein a, b and K are parameters of the Pear growth curve model; y represents power, x represents wind speed;
step three: abnormal data marking
3-1 primary screening: the cut-in wind speed is 3m/s, the cut-in wind speeds of different wind power plants may be different, the 3m/s is taken as an example, data smaller than the cut-in wind speed are subjected to primary abnormal marking in supplemented wind speed-power data, the primary abnormal data are marked as 1, and the data are directly removed;
3-2 screening of allowance: the remaining wind speed-power data with the abnormal data of the primary screen removed is divided into two stages for processing: in the constant torque operation stage and the constant power operation stage, corresponding allowances are respectively set in each stage to enable the allowances to be in accordance with the distribution characteristics of a standard wind speed-power curve of the corresponding stage, then a function expression of the upper boundary of a wind power curve determined by a growth curve function is used for obtaining corresponding power data in each stage by using the known wind speed, the upper boundary of the wind speed-power value of each stage is respectively obtained, and the upper boundary of the wind speed-power value is translated to the right to determine the lower boundary of the wind speed-power value; finally, adding corresponding margins to the upper and lower boundaries of each stage respectively, wherein the data in the margin range is normal data, the data outside the margin range is margin abnormal data, the margin abnormal data is also marked as 1, all residual wind speed-power data with the primary screening abnormal data removed are divided into wind power normal data and margin abnormal data after the primary screening of the margins, and the wind power normal data is marked as 0;
the wind speed range of the constant torque operation stage is larger than the cut-in wind speed and smaller than the rated wind speed; the wind speed range in the constant power operation stage is larger than the rated wind speed; setting an initial allowance a left in a constant torque operation stage and an initial allowance b left in a constant power operation stage; after the upper boundary of the wind power curve is determined in the step 2-2, the parameter of the growth curve function is a known value at the moment, the known wind speed data of the corresponding stage is directly substituted to obtain the corresponding power data, the upper boundary of the exact wind speed-power value is obtained, the lower boundary is obtained by translation, and the left end point of the horizontal line of the lower boundary is close to the sparse point at the moment; adding the left margin to the upper and lower boundaries of each stage to obtain a new boundary, wherein the two new boundaries form a margin range; and judging whether the power data is in a margin range or not, and if the power data is beyond the margin range, further performing margin abnormity marking on the power data, wherein the margin abnormity marking is also marked as 1.
The setting of the margin value finally needs to meet the distribution characteristics of a standard wind speed-power curve, the initial margin value can be set in practical use and then the standard can be achieved through adjustment, if the overall shape of the obtained normal data scatter diagram does not meet the distribution characteristics of the standard wind speed-power curve, the magnitudes of the margin values a and b need to be adjusted again until the standard is met, the final margin value is further determined, and the wind speed-power normal data scatter diagram with a certain margin is obtained.
Step four: correcting and supplementing abnormal data to wind speed data
4-1, establishing a wind speed correlation coefficient matrix of all fans of the wind power plant according to a pearson correlation analysis method, and finding normal fan data with high correlation degree with the missing data by taking 50 fans in the wind power plant as an example in the embodiment. The pearson correlation coefficient, also called pearson product-moment correlation coefficient, is used to measure the linear correlation between two sets of data X, Y, and its value is between-1 and 1.
The calculation formula is
In the formula: cov (X, Y) is the covariance of X and Y; sigma X 、σ Y Standard deviations for X and Y, respectively; x and Y are respectively the wind speed values of two different fans. Rho X,Y The closer to 1 the value of (a) indicates that the correlation between the two sets of data is stronger.
4-2, sorting the correlation coefficients of the fan with the data to be supplemented and other fans from large to small, correspondingly replacing the abnormal wind speed-power data of the fan with the highest correlation coefficient at the moment with the normal fan wind speed-power data with the highest correlation coefficient to obtain a corrected value v eq 、P eq 。
Nothing in this specification is said to apply to the prior art.
Claims (5)
1. A wind power plant wind speed-power abnormal data monitoring and correcting method comprises the following steps: discretizing historical wind speed data and power data, dividing the discretized historical wind speed data and power data into m1 wind speed intervals and m2 power intervals, wherein one wind speed interval corresponds to m2 power intervals and forms m1xm2 areas together, and counting the number of wind speed-power data in each area respectively to obtain a wind speed-power joint frequency distribution histogram; obtaining the density of the wind speed-power data of each area according to the frequency distribution in each area, and discarding the wind speed-power data of the area with the density of the wind speed-power data below 0.00001; for the reserved wind speed-power data, the maximum power value in each area is obtained, the maximum wind speed value in each wind speed interval is obtained, one maximum wind speed value corresponds to the maximum power values of m2 areas, the maximum power values in the areas are sequenced according to the maximum wind speed value, an array P _ v of the maximum wind speed values-the maximum power values is obtained, and an array interval is formed between the adjacent maximum wind speed values;
interpolating the array interval of the array P _ v by adopting a piecewise cubic Hermite interpolation polynomial mode, setting the interpolation number, and supplementing wind speed-power data; fitting the supplemented wind speed-power data by using a growth curve function to obtain an upper boundary of a wind power curve and a function expression thereof;
primary screening: performing primary abnormal marking on data smaller than the cut-in wind speed in the supplemented wind speed-power data, marking the primary abnormal data as 1, and directly removing the data;
screening the margin: the remaining wind speed-power data with the abnormal data of the primary screening removed is divided into two stages for processing: in the constant torque operation stage and the constant power operation stage, corresponding allowances are respectively set in each stage to enable the allowances to be in line with the distribution characteristics of a standard wind speed-power curve of the corresponding stage, then a function expression of an upper boundary of a wind power curve determined by a growth curve function is used for obtaining corresponding power data in each stage by using a known wind speed, the upper boundary of the wind speed-power value of each stage is respectively obtained, and the upper boundary of the wind speed-power value is translated to the right to determine the lower boundary of the wind speed-power value; finally, adding corresponding margins to the upper boundary and the lower boundary of each stage respectively, wherein the data in the margin range are normal data, the data outside the margin range are margin abnormal data, the margin abnormal data are also marked as 1, all residual wind speed-power data with the removed primary screened abnormal data are divided into wind power normal data and margin abnormal data after primary screening of the margins, and the wind power normal data are marked as 0;
and (3) abnormal data correction: carrying out correlation analysis on normal wind power data of all fans of the wind power plant by using a Pearson correlation analysis method to obtain a correlation coefficient matrix among the fans; screening out the fan with the maximum relation number in the matrix for the fan with the data to be supplemented, matching the screened fan data with corresponding abnormal data of the fan with the data to be supplemented, and if the screened fan data is marked as 0, correcting and supplementing margin abnormal data according to the normal data of the wind speed-wind power of the fan with the maximum correlation coefficient; and if the screened fan data is marked as 1, correcting and supplementing the corresponding data of the fan with the data to be supplemented.
2. The wind farm wind speed-power anomaly data monitoring and correcting method according to claim 1, characterized in that the anomaly data are divided into three categories, the first category is data with high wind speed and zero power due to sensor failure or shutdown; the second type is data caused by wind curtailment or faults; the third category is data with very low wind speed and high power caused by sensor failure or communication error.
3. A wind power plant wind speed-power abnormal data monitoring and correcting method comprises the following steps:
the method comprises the following steps: generating arrays of wind speed-power maximums
1-1, arranging historical wind speed data and power data from small to large, performing discretization processing, and dividing the historical wind speed data and the power data into m1 wind speed intervals with the interval size of 1m/s and m2 power intervals with the interval size of 100 kw; counting the frequency of data points in each interval by taking the interval divided by the wind speed as a standard;
respectively counting the number N of wind speed-power data in each area of m1xm2 areas, wherein N = { N1, N2, N3 · · · · · } generates a wind speed-power joint frequency distribution histogram;
1-2, obtaining the wind speed-power density rho of the corresponding area according to the formula (1);
in the formula (1), N is the total number of the historical wind speed data, and N is the number of the wind speed-power data in each area, namely the area frequency;
1-3, testing the density of the wind speed-power in each area, and if the density rho of the wind speed-power is less than 0.00001, discarding the area; otherwise, reserving the region;
1-4, respectively finding the maximum power value and the maximum wind speed value of each area in the data which are tested to meet the density requirement, combining the maximum wind speed values and the maximum power values of all the areas to obtain an array P _ v of the maximum wind speed value-the maximum power value, wherein an array interval is formed between the adjacent maximum wind speed values;
step two: growth curve function fitting wind speed-power curve
2-1, performing interpolation in each array interval of the arrays P _ v by using a piecewise cubic Hermite interpolation polynomial to supplement wind speed-power data;
formula (2) is an expression of a cubic Hermite interpolation polynomial, where x 0 ，x 1 The positions of two adjacent points of the point to be interpolated, namely the endpoint wind speed value of each array interval; y is 0 ，y 1 Corresponding to the argument x 0 ，x 1 The dependent variable of (a), namely the maximum power value corresponding to the endpoint wind speed; y' 0 And y' 1 Is the corresponding derivative, x is the wind speed;
2-2, fitting the supplemented wind speed-power data by using a growth curve function to obtain an upper boundary of a wind power curve and a functional expression thereof, wherein the upper boundary of the wind power curve and the functional expression thereof are shown in a formula (3);
wherein a, b and K are parameters of a pierce model; y represents power, x represents wind speed;
step three: abnormal data marking
3-1 primary screening: the cut-in wind speed is 3m/s, data smaller than the cut-in wind speed are subjected to preliminary abnormal marking in the supplemented wind speed-power data, the preliminary abnormal data are marked as 1, and the data are directly eliminated;
3-2 screening allowance: the remaining wind speed-power data with the abnormal data of the primary screening removed is divided into two stages for processing: in the constant torque operation stage and the constant power operation stage, corresponding allowances are respectively set in each stage to enable the allowances to be in line with the distribution characteristics of a standard wind speed-power curve of the corresponding stage, then a function expression of an upper boundary of a wind power curve determined by a growth curve function is used for obtaining corresponding power data in each stage by using a known wind speed, the upper boundary of the wind speed-power value of each stage is respectively obtained, and the upper boundary of the wind speed-power value is translated to the right to determine the lower boundary of the wind speed-power value; finally, adding corresponding margins to the upper boundary and the lower boundary of each stage respectively, wherein the data in the margin range are normal data, the data outside the margin range are margin abnormal data, the margin abnormal data are also marked as 1, all residual wind speed-power data with the removed primary screened abnormal data are divided into wind power normal data and margin abnormal data after primary screening of the margins, and the wind power normal data are marked as 0;
step four: correcting and supplementing abnormal data to wind speed data
Establishing a wind speed correlation coefficient matrix of all fans of the wind power plant by using wind power normal data according to a Pearson correlation analysis method, screening the fan with the largest correlation number in the matrix for the fan with data to be supplemented, matching the screened fan data with corresponding abnormal data of the fan with the data to be supplemented, and if the screened fan data is marked as 0, correcting and supplementing margin abnormal data according to the wind speed-wind power normal data of the fan with the largest correlation coefficient; and if the screened fan data is marked as 1, correcting and supplementing the corresponding data of the fan with the data to be supplemented.
4. The wind farm wind speed-power anomaly data monitoring and correcting method according to claim 3, wherein the wind speed range in the constant torque operation stage is greater than a cut-in wind speed and less than a rated wind speed; the wind speed range of the constant power operation stage is larger than the rated wind speed; setting an initial allowance a left in a constant torque operation stage and an initial allowance b left in a constant power operation stage; after the upper boundary of the wind power curve is determined in the step 2-2, the parameter of the growth curve function at the moment is a known value, the known wind speed data of the corresponding stage is directly substituted to obtain corresponding power data, the upper boundary of the exact wind speed-power value is obtained, and the lower boundary is determined by translation; adding the left margin to the upper and lower boundaries of each stage to obtain a new boundary, wherein the two new boundaries form a margin range; and judging whether the power data is in a margin range or not, and if the power data is beyond the margin range, further performing margin abnormity marking on the power data, wherein the margin abnormity marking is also marked as 1.
5. The wind power plant wind speed-power abnormal data monitoring and correcting method according to claim 4, characterized in that the margin value is set to meet the distribution characteristics of a standard wind speed-power curve, the standard is reached by adjustment at the later stage of setting an initial margin value in actual use, if the obtained normal data scatter diagram is not in accordance with the distribution characteristics of the standard wind speed-power curve, the magnitudes of the margin values a and b need to be readjusted until the margin values are in accordance with the standard, and then the final margin value is determined, so that the wind speed-power normal data scatter diagram with a certain margin is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211050136.4A CN115408860A (en) | 2022-08-31 | 2022-08-31 | Abnormal value monitoring and correcting method for wind speed-power data of wind power plant |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211050136.4A CN115408860A (en) | 2022-08-31 | 2022-08-31 | Abnormal value monitoring and correcting method for wind speed-power data of wind power plant |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115408860A true CN115408860A (en) | 2022-11-29 |
Family
ID=84163468
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211050136.4A Pending CN115408860A (en) | 2022-08-31 | 2022-08-31 | Abnormal value monitoring and correcting method for wind speed-power data of wind power plant |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115408860A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115980504A (en) * | 2023-03-21 | 2023-04-18 | 中车山东风电有限公司 | Online detection method and detection terminal for power generation performance of wind generating set |
-
2022
- 2022-08-31 CN CN202211050136.4A patent/CN115408860A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115980504A (en) * | 2023-03-21 | 2023-04-18 | 中车山东风电有限公司 | Online detection method and detection terminal for power generation performance of wind generating set |
| CN115980504B (en) * | 2023-03-21 | 2023-08-11 | 中车山东风电有限公司 | Online detection method and detection terminal for generating performance of wind generating set |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11401913B2 (en) | Yaw calibration method and system for wind turbine | |
| CN106014858B (en) | A kind of Wind turbines are to wind error automatic calibrating method and device | |
| CN107038292B (en) | Multi-wind-field output correlation modeling method based on self-adaptive multivariate nonparametric kernel density estimation | |
| CN107846014B (en) | Power grid probability vulnerability assessment method based on random features and series calculation | |
| CN109002650A (en) | A kind of Wind turbines power curve modeling method | |
| CN109779848B (en) | Method and device for obtaining full-field wind speed correction function and wind power plant | |
| CN111275570A (en) | Wind turbine generator set power abnormal value detection method based on iterative statistics and hypothesis test | |
| CN115408860A (en) | Abnormal value monitoring and correcting method for wind speed-power data of wind power plant | |
| CN111260503A (en) | Wind turbine generator power curve outlier detection method based on cluster center optimization | |
| CN108869174A (en) | A kind of blade of wind-driven generator intrinsic frequency operating condition compensation method of Nonlinear Modeling | |
| CN110705099B (en) | Method for verifying output correlation of wind power plant | |
| CN115115090A (en) | Wind power short-term prediction method based on improved LSTM-CNN | |
| CN110365053B (en) | Short-term wind power prediction method based on delay optimization strategy | |
| CN113359435B (en) | Correction method for dynamic working condition data of thermal power generating unit | |
| CN110879927A (en) | Sea clutter amplitude statistical distribution field modeling method for sea target detection | |
| CN107808209B (en) | Wind power plant abnormal data identification method based on weighted kNN distance | |
| Pandit et al. | Comparison of binned and Gaussian Process based wind turbine power curves for condition monitoring purposes | |
| CN115438312A (en) | Probability distribution model establishment and application method and medium for new energy output | |
| CN113048012B (en) | Wind turbine generator yaw angle identification method and device based on Gaussian mixture model | |
| Bao et al. | Iterative modeling of wind turbine power curve based on least‐square B‐spline approximation | |
| CN110083804B (en) | Wind power plant SCADA data missing intelligent repairing method based on condition distribution regression | |
| CN117473854A (en) | Marine wind turbine gearbox health assessment method, system and equipment | |
| CN112459970B (en) | Wind turbine generator anomaly detection method using data counterstudy | |
| CN113309661B (en) | Method for improving generating capacity of wind generating set | |
| Pandit et al. | Comparative study of binning and gaussian process based rotor curves of a wind turbine for the purpose of condition monitoring |
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 |