CN110365053B - Short-term wind power prediction method based on delay optimization strategy - Google Patents
Short-term wind power prediction method based on delay optimization strategy Download PDFInfo
- Publication number
- CN110365053B CN110365053B CN201910732490.7A CN201910732490A CN110365053B CN 110365053 B CN110365053 B CN 110365053B CN 201910732490 A CN201910732490 A CN 201910732490A CN 110365053 B CN110365053 B CN 110365053B
- Authority
- CN
- China
- Prior art keywords
- power
- data
- wind speed
- wind
- prediction
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000005457 optimization Methods 0.000 title claims abstract description 8
- 230000000630 rising effect Effects 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 27
- 238000005070 sampling Methods 0.000 claims abstract description 23
- 238000012937 correction Methods 0.000 claims abstract description 22
- 230000002159 abnormal effect Effects 0.000 claims abstract description 7
- 238000006243 chemical reaction Methods 0.000 claims abstract description 6
- 230000001174 ascending effect Effects 0.000 claims description 19
- 238000010606 normalization Methods 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 description 13
- 238000011160 research Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000010248 power generation Methods 0.000 description 4
- 238000009499 grossing Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000004940 physical analysis method Methods 0.000 description 1
- 238000000053 physical method Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- H02J3/386—
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
-
- 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
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Wind Motors (AREA)
Abstract
The invention discloses a short-term wind power prediction method based on a delay optimization strategy, which comprises the following steps: acquiring wind speed-power data; sampling the acquired data; removing abnormal data in the sampled data; selecting continuous sampling data; training a prediction model; predicting power by using the trained prediction model; a delay function is defined for the continuously rising wind speed and power correction is performed. The method quantitatively describes the magnitude and the speed of the wind speed fluctuation by defining the wind speed fluctuation rate; the method is characterized in that the method is used for analyzing the correlation between the amplitude and the included angle of the wind speed fluctuation rate and the prediction delay, and a delay correction function is provided by combining the wind energy conversion theory to correct the prediction power, so that the wind power prediction accuracy of the section with severe wind speed fluctuation can be improved.
Description
Technical Field
The invention relates to wind power prediction, in particular to a short-term wind power prediction method based on a delay optimization strategy.
Background
With the development of wind power generation industry, the wind power integration quantity is increased year by year, and the large-scale wind power integration brings higher requirements for wind power prediction. However, due to the instability of wind, and the turbine power generation is proportional to the cube of the wind speed, the fluctuation of the wind speed in a specific time period will tend to affect the energy conversion rate of the wind driven generator. Wind power prediction methods are divided into physical analysis methods, statistical learning methods and combinations of the two. The physical method is suitable for the physical process of wind change, and the wind power plant geographic information and the wind power plant characteristics are added for analysis, modeling and prediction to form a wind power plant or wind power plant power curve to predict power. The statistical learning method is based on historical data, such as NWP data, collected data of a wind driven generator and a wind measuring tower, and the like, and a model of wind-to-power conversion is obtained for prediction through a trend analysis method, but the method has higher requirements on the historical data, and the data with gentle fluctuation and certain regularity can obtain a prediction result with higher precision. Therefore, the problems of unstable wind power generation, low wind energy utilization rate and the like caused by factors such as sudden changes and fluctuation of wind still need to be improved, and high-precision short-term wind power prediction is still the focus of research.
Regarding the research on the influence of wind speed characteristics on the power generated by a fan in the prior art, mo Shuting et al, in the analysis of fan power fluctuation characteristics based on equivalent wind speed, research on the power fluctuation characteristics of the fan under an equivalent wind speed model by establishing an equivalent wind speed model containing wind shearing and tower shadow effects, but only show that the influence difference of components of the equivalent wind speed on the power characteristics of a unit is obvious by comparing simulation data with actual operation data of a wind field. Yang Mao et al qualitatively analyze the association between wind speed and power fluctuation series in a real-time wind power prediction study based on probability distribution quantization index and gray association decision, and simultaneously select the power fluctuation sequence and gray association of the wind speed fluctuation sequence as decision variables to enter the decision, and finally predict the power by combining the power obtained by the standard wind speed power curve of the fan with the power obtained by searching the same wind speed section by gray association decision analysis. However, the above research does not quantitatively analyze the influence of wind speed fluctuation on power prediction, and because inertia exists in the process of converting wind energy into electric energy from wind energy received by a fan, a delay phenomenon exists between predicted power and actual power when wind speed changes, particularly continuously changes, so that the accuracy of power prediction is further reduced.
Disclosure of Invention
The invention aims to: the invention aims to provide a short-term wind power prediction method based on a delay optimization strategy, which solves the problem that in the prior art, when wind speed changes, the prediction accuracy is reduced due to the fact that delay exists between predicted power and actual power.
The technical scheme is as follows: the invention provides a short-term wind power prediction method based on a delay optimization strategy, which comprises the following steps:
(1) Acquiring wind speed-power data: collecting wind speed and power data in a specific time period according to a set time resolution by using a wind measuring tower and a wind generator of a wind power plant to obtain sampling data;
(2) Removing abnormal data in the sampling data to obtain normal sampling data, wherein the abnormal data comprises data of missing measurement at a specific moment and error data;
(3) Selecting continuous sampling data: selecting continuous N groups of sampling data without abnormality from normal sampling data as continuous sampling data, taking x groups of continuous data as training data and y groups of continuous data as prediction data, wherein x+y=N;
(4) Training a prediction model: training a model by using training data, using wind speed data as input and power data as output, and training a prediction model to adapt to a wind speed-to-power conversion rule;
(5) Predicting power by using the trained prediction model: taking wind speed data in the prediction data as input, and predicting through a prediction model to obtain predicted power;
(6) Defining a delay function for continuously ascending wind speed and carrying out power correction, wherein the delay function f (v) is related to the wind speed, v represents the wind speed, and v (t) represents the wind speed at the moment t; according toCorrecting the power to obtain corrected power P X, wherein />Representing the predicted power.
Further, the prediction model can adopt a bp neural network or an Xgboost algorithm model.
Further, the method step (6) comprises:
(61) Taking the data for prediction as a test set wind speed sequence, and carrying out EMD reconstruction to realize smooth processing of the data;
(62) Selecting a continuous wind speed ascending section from the wind speed sequence reconstructed by the EDM, searching an extreme point of the wind speed sequence in the ascending section, and quantitatively describing the magnitude and the speed of the wind speed change through the amplitude and the included angle;
(63) Data normalization: normalizing the amplitude value and the included angle of the wind speed of the continuous ascending section;
(64) Constructing a delay function f (v), wherein the expression is as follows:
f(v)=ωv 3
ω=-(αU+βθ)
wherein v is the wind speed of the continuous rising section, ω is a correction coefficient, U and θ are the amplitude and the included angle of the wind speed of the rising section after normalization, α and β are the proportionality factors between U and θ and the correction coefficient ω, and λ and μ are the correlation coefficients between U and θ and the power prediction error respectively;
Further, the step (63) includes:
calculating the difference between the predicted and actual power, i.e wherein ,/>For predicted power, P is actual power;
amplitude set A= { U of n wind speed rising sections to be selected 1 ,U 1 ,...,U n Included angle set b= { θ for n wind speed rising sections 1 ,θ 1 ,...,θ n Power error set z= { p for n wind speed rising sections } 1 ,p 2 ,...,p n Normalizing to set a, b and z, and calculating correlation coefficients lambda and mu between the set a, b and z respectively;
wherein ,ΔP i for the difference between the predicted power and the actual power of the ith point, m is the number of data points of the nth rising segment, p n Is the power error mean value of the nth rising segment.
Further, in step (4), the number of training data is larger than the number of prediction data, that is, x > y.
The beneficial effects are that: compared with the prior art, the invention quantitatively defines the continuous fluctuation of the wind speed through two variables of the amplitude and the included angle, and analyzes the relation between the wind speed rising sections with different degrees and the power delay; by defining a delay function to correct the prediction error caused by the continuous rise of wind speed, the accuracy of power prediction is improved.
Drawings
FIG. 1 upwind-downwind versus fitted power curve;
FIG. 2 is a power prediction and correction flow chart;
FIG. 3 is a schematic diagram of wind speed fluctuation rate;
FIG. 4 is a graph of the result of the EMD data smoothing process;
the normalization index of the rising segments of different intervals in FIG. 5;
fig. 6 is a graph comparing the correction results.
Detailed Description
The invention is further described below with reference to the accompanying drawings and examples:
according to the method of the difference value before and after the wind speed, the wind speed is roughly divided into ascending wind and descending wind, and compared with a wind speed power curve fitted by the Bian method, as shown in FIG. 1, the actual power of the ascending wind is almost completely below the fitted power curve, and the deviation degree is increased along with the ascending of the wind speed; although part of the actual power of the downwind is also above the fitted power curve, most of the data is clung to the power curve, and as the wind speed increases, the more obvious the phenomenon is, even below the curve. The correction method adopted by the invention is different because the wind speed is increased and decreased to have different effects on wind power generation, and the invention only considers the influence of the continuous increase of the wind speed on the power prediction accuracy and corrects the power prediction accuracy.
The short-term wind power prediction method based on the delay optimization strategy provided by the invention has a flow shown in a figure 2, and comprises the following steps:
(1) Acquiring wind speed-power data: collecting wind speed and power data in a specific time period according to a set time resolution by using a wind measuring tower and a wind generator of a wind power plant to obtain sampling data; wind speed and power data include cut-in wind speed, rated wind speed, cut-out wind speed, rated power, time resolution of the wind turbine.
(2) And removing abnormal data in the sampling data to obtain normal sampling data, wherein the abnormal data comprise data which are not measured at a certain moment caused by faults of a wind measuring tower and a wind driven generator acquisition system or external factors and error data such as wind speed and power less than 0 or other data which do not accord with common sense.
(3) Selecting continuous sampling data: selecting continuous N groups of sampling data without abnormality from normal sampling data as continuous sampling data, taking x groups of continuous data as training data and y groups of continuous data as prediction data, wherein x+y=N; in an embodiment of the invention, the number of training data is greater than the number of prediction data, i.e. x > y.
(4) Training a prediction model: training a model by using training data, using wind speed data as input and power data as output, and training a prediction model to adapt to a wind speed-to-power conversion rule; the prediction model can adopt a bp neural network or an Xgboost algorithm model.
(5) Predicting power by using the trained prediction model: the wind speed data in the prediction data is taken as input, and the prediction model is used for predicting to obtain the predicted power
(6) Defining a delay function for continuously ascending wind speed and carrying out power correction, wherein the delay function f (v) is related to the wind speed, v represents the wind speed, and v (t) represents the wind speed at the moment t;
(61) Taking the data for prediction as a test set wind speed sequence, and carrying out EMD reconstruction to realize smooth processing of the data;
(62) And selecting a continuous wind speed ascending section from the wind speed sequence reconstructed by the EDM, searching an extreme point of the wind speed sequence in the ascending section, and quantitatively describing the magnitude and the speed of the wind speed change through the amplitude and the included angle, as shown in figure 3. Starting from the first data point, searching for an extreme point of the wind speed sequence: if v (t) satisfies v (t-1) < v (t) and v (t) > v (t+1), then v (t) is the maximum point; if v (t) satisfies v (t-1) > v (t) and v (t) < v (t+1), v (t) is a minimum point. The amplitude U is the difference between wind speeds corresponding to the maximum value max and the minimum value min, and the included angle theta is the included angle between the hypotenuse of the triangle marked on the way and the horizontal axis. In order to better embody and analyze the influence of the continuous rise of the wind speed on the power, after the extreme points of the wind speed sequence are identified, the maximum value and the minimum value points with reasonable intervals are selected to describe the continuous rise of the wind speed.
(63) Data normalization: the amplitude and the included angle of the wind speed of the continuous ascending section are normalized by the following method:
calculating the difference between the predicted and actual power, i.e wherein ,/>For predicted power, P is actual power;
amplitude set A= { U of n wind speed rising sections to be selected 1 ,U 1 ,...,U n Included angle set b= { θ for n wind speed rising sections 1 ,θ 1 ,...,θ n Power error set z= { p for n wind speed rising sections } 1 ,p 2 ,...,p n Normalizing to set a, b and z, and calculating correlation coefficients lambda and mu between the set a, b and z respectively;
wherein ,ΔP i pre-determined for the ith pointThe difference between the measured power and the actual power, m is the number of data points of the nth rising segment, p n Is the power error mean value of the nth rising segment.
(64) Constructing a delay function f (v), wherein the expression is as follows:
f(v)=ωv 3 (1)
ω=-(αU+βθ) (2)
wherein v is the wind speed of the continuous rising section, ω is a correction coefficient, U and θ are the amplitude and the included angle of the wind speed of the rising section after normalization, α and β are the proportionality factors between U and θ and the correction coefficient ω, and λ and μ are the correlation coefficients between U and θ and the power prediction error respectively;
(65) Based on predicted powerAnd a delay function f (v) to obtain a corrected power P X The expression is:
in the embodiment of the invention, in order to explore the relation among the amplitude, the included angle and the delay phenomenon, an Xgboost algorithm and a bp neural network algorithm are respectively used for preliminary prediction.
Currently, the Xgboost algorithm is often used in classification or regression problems, and in the training process, a plurality of classifiers are learned by changing the weight of a training sample, so that an optimal classifier is finally obtained. After each round of training is finished, the weight of the training samples which are correctly classified is reduced, the weight of the samples which are incorrectly classified is increased, after multiple times of training, some training samples which are incorrectly classified can get more attention, the weight of the correct training samples approaches 0, a plurality of simple classifiers are obtained, and a final model is obtained by combining the classifiers. The objective function is:
where i represents the i-th sample,representing the prediction error of the ith sample, Σ k Ω(f k ) A function representing the complexity of the tree.
The bp neural network is most widely applied to short-term wind power prediction, is also most representative, and has strong nonlinear mapping capability and higher flexibility, and is characterized by error back propagation.
A certain wind farm in Shanghai Chongming is taken as a research object, and is positioned in the east Asia monsoon area, and 30 fans are shared. The cut-in wind speed of the fan is 3m/s, the rated wind speed is 12m/s, the cut-out wind speed is 25m/s, the rated power is 2MW, the time resolution is 5min, and the data are acquired from the wind tower and the wind driven generator in 2014, 9 months and 2015. 2000 sets of continuous data were extracted from fans 2014, 9, and 2015, 9, one year, with 70% used for training and 30% used for testing.
For the fluctuating wind speed sequence, for better identification of extreme points, the data is smoothed by using EMD, and the smoothing result is shown in figure 4. It can be seen that the mutation points of the wind speed sequence after EMD reconstruction are better captured, namely the extreme points of the continuous change of the wind speed are better identified, the reconstructed wind speed sequence maintains the characteristics of the original sequence, and the identification of the extreme points can be carried out by adopting the reconstructed sequence.
After the extreme points of the reconstructed wind speed sequence are identified, in order to select a continuous ascending sequence with reasonable interval length, the number epsilon of interval data among the extreme points is more than or equal to 1, epsilon is more than or equal to 2, epsilon is more than or equal to 3, epsilon is more than or equal to 4, and the ascending segments are selected according to the correlation. Fig. 5 is a line graph of angles, magnitudes and powers of 4 rising segments, and for more clearly explaining the correlation of angles, magnitudes and powers of rising segments at different intervals, the corresponding correlation coefficients are shown in table 1.
TABLE 1 correlation coefficient of different interval rising segments
As can be seen from table 1, as the data interval of the rising segments increases, the correlation coefficient also increases, but there is a problem that the interval increases and the number of rising segments meeting the requirement also decreases. Therefore, comprehensively considering the number of data intervals between extreme points of the rising sections and the number of the rising sections, herein, a continuously rising wind speed section with 5 or more data points (i.e. a rising section with epsilon being larger than or equal to 3) is selected as a continuously rising wind speed section, and the amplitude U and the included angle theta of 5 rising sections meeting the conditions in fig. 4 are shown in table 2:
TABLE 2 wind speed fluctuation quantity at the ascending section
Normalizing the set A, B, Z to obtain sets a, b and z, and obtaining correlation coefficients between the set a and the set z and between the set b and the set z, wherein the obtained results are shown in table 3:
table 3 variable normalization
Since both the amplitude and the angle have a certain correlation with the power delay error, it is feasible to define the correction coefficient ω with these two variables. Wherein alpha and beta in the formula (2) are set according to the correlation coefficient of the amplitude and the included angle, and the alpha and beta values are calculated by the formula (3). In this example, α is 0.63 and β is 0.37.
Aiming at the delay phenomenon of the wind speed sequence prediction result of the ascending segment, a delay function is provided for correcting the prediction result, and in order to avoid accidental, the prediction results of 2 prediction methods, namely Xgboost and bp neural network, are respectively corrected, and the result is shown in figure 6.
As can be seen from fig. 6, the power correction for the rising section has a certain effect, most of the data points are more attached to the actual curve than before correction, but there is a problem that the power corresponding to the extreme point of the wind speed change does not reach a good correction effect, reasonable guess of the problem is the influence of the regulation strategy of the fan when the fan is in response to the sudden change of the wind speed, and later researches can take the influence as the cut-in point. To more clearly illustrate the effect after the delay correction, MAE and RMSE were used as evaluation indexes for the correction results, and the results were compared with those before the correction, and are shown in Table 4.
Table 4 error comparison
It can be seen that, whatever the method and the error index, the power corrected by the delay correction function is greatly improved in accuracy compared with the original predicted power, wherein the Mean Absolute Error (MAE) reflects the amplitude of the predicted error, the Root Mean Square Error (RMSE) reflects the degree of dispersion of the error, and the method provided herein can better correct the predicted power of the continuously rising wind speed section.
Claims (3)
1. A short-term wind power prediction method based on a delay optimization strategy is characterized by comprising the following steps:
(1) Acquiring wind speed-power data: collecting wind speed and power data in a specific time period according to a set time resolution by using a wind measuring tower and a wind generator of a wind power plant to obtain sampling data;
(2) Removing abnormal data in the sampling data to obtain normal sampling data, wherein the abnormal data comprises data of missing measurement at a specific moment and error data;
(3) Selecting continuous sampling data: selecting continuous N groups of sampling data without abnormality from normal sampling data as continuous sampling data, taking x groups of continuous data as training data and y groups of continuous data as prediction data, wherein x+y=N;
(4) Training a prediction model: training the model by using the training data, using wind speed data as input and power data as output, and training the prediction model to adapt to the wind speed-to-power conversion rule;
(5) Predicting power by using the trained prediction model: taking wind speed data in the prediction data as input, and predicting through the prediction model to obtain predicted power;
(6) Defining a delay function for continuously ascending wind speed and carrying out power correction, wherein the delay function f (v) is related to the wind speed, v represents the wind speed, and v (t) represents the wind speed at the moment t; according toCorrecting the power to obtain corrected power P X, wherein />Representing the predicted power; the method specifically comprises the following steps:
(61) Taking the data for prediction as a test set wind speed sequence, and carrying out EMD reconstruction to realize smooth processing of the data;
(62) Selecting a continuous wind speed ascending section from the wind speed sequence reconstructed by the EDM, searching an extreme point of the wind speed sequence in the ascending section, and quantitatively describing the magnitude and the speed of the wind speed change through the amplitude and the included angle;
(63) Data normalization: normalizing the amplitude value and the included angle of the wind speed of the continuous ascending section;
calculating the difference between the predicted and actual power, i.e wherein ,/>For predicted power, P is actual power;
amplitude set A= { U of n wind speed rising sections to be selected 1 ,U 1 ,...,U n Included angle of n wind speed rising sectionsSet b= { θ 1 ,θ 1 ,...,θ n Power error set z= { p for n wind speed rising sections } 1 ,p 2 ,...,p n Normalizing to set a, b and z, and calculating correlation coefficients lambda and mu between the set a, b and z respectively;
wherein ,ΔP i for the difference between the predicted power and the actual power of the ith point, m is the number of data points of the nth rising segment, p n The power error mean value of the nth rising section;
(64) Constructing a delay function f (v), wherein the expression is as follows:
f(v)=ωv 3
ω=-(αU+βθ)
wherein v is the wind speed of the continuous rising section, ω is a correction coefficient, U and θ are the amplitude and the included angle of the wind speed of the rising section after normalization, α and β are the proportionality factors between U and θ and the correction coefficient ω, and λ and μ are the correlation coefficients between U and θ and the power prediction error respectively;
2. The short-term wind power prediction method according to claim 1, wherein the prediction model can adopt a bp neural network or an Xgboost algorithm model.
3. The short-term wind power prediction method according to claim 1, wherein in the step (2), the number of training data is larger than the number of prediction data, that is, x > y.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910732490.7A CN110365053B (en) | 2019-08-09 | 2019-08-09 | Short-term wind power prediction method based on delay optimization strategy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910732490.7A CN110365053B (en) | 2019-08-09 | 2019-08-09 | Short-term wind power prediction method based on delay optimization strategy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110365053A CN110365053A (en) | 2019-10-22 |
CN110365053B true CN110365053B (en) | 2023-04-25 |
Family
ID=68223504
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910732490.7A Active CN110365053B (en) | 2019-08-09 | 2019-08-09 | Short-term wind power prediction method based on delay optimization strategy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110365053B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111414354A (en) * | 2020-03-12 | 2020-07-14 | 广西电网有限责任公司 | Wind power plant data processing method and device and computer readable storage medium |
CN111476402A (en) * | 2020-03-16 | 2020-07-31 | 云南电网有限责任公司 | Wind power generation capacity prediction method coupling meteorological information and EMD technology |
CN116317169B (en) * | 2023-05-17 | 2023-08-04 | 三峡智控科技有限公司 | Remote intelligent comparison platform and method for wind power prediction system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102496927A (en) * | 2011-12-16 | 2012-06-13 | 中国电力科学研究院 | Wind power station power projection method based on error statistics modification |
CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
CN108092319A (en) * | 2017-12-13 | 2018-05-29 | 国家电网公司 | A kind of Uncertainty Analysis Method and device of short-term wind-electricity power prediction |
CN109190845A (en) * | 2018-09-29 | 2019-01-11 | 南京信息工程大学 | A kind of two stages dynamic optimization short-term wind power forecast method |
-
2019
- 2019-08-09 CN CN201910732490.7A patent/CN110365053B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102496927A (en) * | 2011-12-16 | 2012-06-13 | 中国电力科学研究院 | Wind power station power projection method based on error statistics modification |
CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
CN108092319A (en) * | 2017-12-13 | 2018-05-29 | 国家电网公司 | A kind of Uncertainty Analysis Method and device of short-term wind-electricity power prediction |
CN109190845A (en) * | 2018-09-29 | 2019-01-11 | 南京信息工程大学 | A kind of two stages dynamic optimization short-term wind power forecast method |
Non-Patent Citations (2)
Title |
---|
基于小波分析和PSO优化神经网络的短期风电功率预测;叶小岭等;《可再生能源》;20141031;第32卷(第10期);第1486-1492页 * |
基于风速升降特征的短期风电功率预测;叶小岭等;《电力系统保护与控制》;20161001;第44卷(第19期);第56-62页 * |
Also Published As
Publication number | Publication date |
---|---|
CN110365053A (en) | 2019-10-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110365053B (en) | Short-term wind power prediction method based on delay optimization strategy | |
CN106529719B (en) | Wind power prediction method based on particle swarm optimization algorithm wind speed fusion | |
CN110685868A (en) | Wind turbine generator fault detection method and device based on improved gradient elevator | |
CN110594107B (en) | Wind turbine generator fault detection method and device based on rapid gradient elevator | |
CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
CN110766200A (en) | Method for predicting generating power of wind turbine generator based on K-means mean clustering | |
CN108022019B (en) | Wind power plant scheduling method and system based on wind turbine generator classification | |
CN112186761B (en) | Wind power scene generation method and system based on probability distribution | |
CN113379142B (en) | Short-term wind power prediction method based on wind speed correction and fusion model | |
CN111626473A (en) | Two-stage photovoltaic power prediction method considering error correction | |
CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
CN112580874A (en) | Short-term wind power prediction method based on random forest algorithm and TCN | |
CN113505938A (en) | Ultra-short-term wind power combined prediction method and system | |
CN110991701A (en) | Wind power plant fan wind speed prediction method and system based on data fusion | |
CN112994101A (en) | Neural network-based wind power plant generated power post-evaluation and monitoring method | |
CN112651576A (en) | Long-term wind power prediction method and device | |
CN113821931B (en) | Fan output power prediction method and system | |
CN114583710A (en) | Wind power plant reactive voltage optimization control method based on data-driven modeling | |
CN114429233A (en) | Cluster algorithm-based station-level optimal power control method and system | |
CN111931967B (en) | Short-term power prediction method for wind power plant | |
CN112598148A (en) | Fan variable pitch motor temperature fault early warning method based on collaborative expression and LightGBM algorithm | |
CN116123029A (en) | Pitch angle control method and device, electronic equipment and wind turbine generator | |
CN111178601B (en) | Wind turbine generator power prediction method based on meteorological data post-processing | |
CN113610285A (en) | Power prediction method for distributed wind power |
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 |