CN112329344A - Fan wind speed soft measurement method based on principal component analysis method - Google Patents

Fan wind speed soft measurement method based on principal component analysis method Download PDF

Info

Publication number
CN112329344A
CN112329344A CN202011208128.9A CN202011208128A CN112329344A CN 112329344 A CN112329344 A CN 112329344A CN 202011208128 A CN202011208128 A CN 202011208128A CN 112329344 A CN112329344 A CN 112329344A
Authority
CN
China
Prior art keywords
wind speed
data set
svr
principal component
component analysis
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
Application number
CN202011208128.9A
Other languages
Chinese (zh)
Inventor
李荣丽
胡宏彬
周磊
张谦
霍红岩
杜荣华
张国斌
郭瑞君
赵炜
于海存
杨丽
党少佳
赵松
李旭
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Original Assignee
Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd filed Critical Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Priority to CN202011208128.9A priority Critical patent/CN112329344A/en
Publication of CN112329344A publication Critical patent/CN112329344A/en
Pending legal-status Critical Current

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
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
    • 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
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Abstract

The invention discloses a fan wind speed soft measurement method based on a principal component analysis method.A data collector collects original operation data and transmits the original operation data to a principal component analysis processor for data dimension reduction processing, a dimension reduction data set is divided into a training data set and a test data set, the training data set is transmitted to a wind speed GA-SVR trainer, and the test data set is transmitted to an optimal wind speed SVR predictor; the optimal wind speed SVR predictor takes an optimal wind speed SVR prediction model as a core and takes a test data set as input to realize optimal wind speed prediction. The invention provides a solution for solving the problem that the real wind speed borne by the fan cannot be directly measured through the measurable variable of the wind generating set. Analyzing the main effect by a main component analysis method, and eliminating the influence of dimension difference among different samples; the wind speed SVR model is trained based on the genetic algorithm, and the generated wind speed estimation model of the fan is high in accuracy. The application of the invention can realize the ideal wind speed estimation of the fan and provides a way for improving the wind energy utilization and the control quality of the unit.

Description

Fan wind speed soft measurement method based on principal component analysis method
Technical Field
The invention relates to a method for measuring wind speed of a fan, in particular to a method for soft measuring the wind speed of the fan based on a principal component analysis method.
Background
Compared with the traditional energy, the wind energy is clean and renewable. The available wind energy is widely distributed worldwide. The wind power generation has no fuel cost and no environmental governance cost. By virtue of unique advantages, wind power generation is more and more emphasized, and the installed capacity of wind power is rapidly increased. The large-capacity wind power integration inevitably brings challenges to the safety and stability of a power grid and the quality of electric energy, and the prediction of the wind speed of the fan is an effective means for realizing the accurate control of the fan and reducing the adverse effect of wind power on the power grid. However, due to the randomness of wind speed in time and space, the wind speed acting on the wind-sweeping surface of the whole blade is not uniformly distributed, and in addition, an anemometer is usually arranged at the tail of a fan, and the measured wind speed is not consistent with the wind speed acting on the blade. Therefore, the anemometer cannot directly measure the actual wind speed acting on the blade, and the wind speed measured by the anemometer is not enough to be used as the basis for controlling the rotating speed of the fan.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a fan wind speed soft measurement method based on a principal component analysis method.
In order to solve the technical problems, the invention adopts the technical scheme that: a fan wind speed soft measurement method based on a principal component analysis method comprises the following steps:
step one, 7 items of original operation data of the rotating speed, the power of a generator, the pitch angle, the yaw angle, the power of an adjacent fan, the wind direction and the ambient temperature of a wind turbine are collected by a data collector; synthesizing an original operation data set D, and transmitting the original operation data set D to a principal component analysis processor;
analyzing the main effect by a principal component analysis processor, and performing data dimension reduction processing on the original operation data set D by using a principal component analysis method to generate a dimension reduction data set S;
step three, randomly selecting 2/3 behavior training data set S from the dimension reduction data set StrainThe remaining 1/3 rows constitute test data set StestTraining data set StrainTransmitting to a wind speed GA-SVR trainer to obtain a test data set StestTransmitting to an optimal wind speed SVR predictor;
fourthly, the wind speed GA-SVR trainer utilizes a GA algorithm to conduct optimization training on the wind speed SVR model until an optimal wind speed SVR estimation model is generated;
step five, the optimal wind speed SVR predictor tests a data set S by taking the optimal wind speed SVR prediction model obtained in the step four as a coretestAnd the optimal wind speed estimation value is input, and is transmitted to the variable speed wind driven generator controller, so that a basis is provided for rotating speed control.
Storing 7 items of collected original operation data of rotating speed, generator power, pitch angle, yaw angle, adjacent fan power, wind direction and ambient temperature into column vectors D1, D2, D3, D4, D5, D6 and D7 respectively, and synthesizing an original operation data set D ═ D [ D4 ]1,D2,D3,D4,D5,D6,D7]；
Order to
Figure BDA0002757725380000021
The normalized matrix Z is calculated using the original running data set,
Figure BDA0002757725380000022
Figure BDA0002757725380000023
wherein z isijRepresenting the value of the ith row and jth column in the matrix Z, dijRepresents the value of the ith row and the jth column in the matrix D,
Figure BDA0002757725380000024
represents the average value of j columns in the D matrix, sjRepresents the standard deviation;
Figure BDA0002757725380000025
and sjThe calculation method of (a) is shown as follows:
Figure BDA0002757725380000026
further utilizing the standardized matrix Z to perform dimension reduction processing on the original operation data set D, specifically comprising the following steps:
s1, calculating a correlation coefficient matrix rho of the standardized matrix Z; calculating characteristic value lambda of correlation coefficient matrix rho1＞λ2＞λ3＞…＞λ7Corresponding unit feature vector
Figure BDA0002757725380000031
S2, calculating principal component contribution rate pi(ii) a Rate of contribution
Figure BDA0002757725380000032
Wherein lambda is a characteristic value of the correlation coefficient array rho;
s3, cumulative contribution rate of the first q principal components
Figure BDA0002757725380000033
Determining a q value according to the accumulated contribution rate eta value, wherein eta is more than 0.5 and less than 1;
s4, according to the formula
Figure BDA0002757725380000034
Determining a transformation matrix
Figure BDA0002757725380000035
In the formula, ZTRepresents the transpose of the normalized matrix Z;
s5, according to the formula
Figure BDA0002757725380000036
Carrying out dimensionality reduction processing on an original operation data set D to generate a dimensionality reduction data set S ═ S1,S2,…,Sq]。
Further, the method for generating the optimal wind speed SVR estimation model by the wind speed GA-SVR trainer comprises the following steps:
the wind speed SVR model regression function is: y ═ f (x) ═ ω · K (x · x)i) + b, where y is a wind speed measurement; x is the rotation speed, generator power, pitch angle, yaw angle, adjacent fan power, wind direction and ring of wind turbineAmbient temperature 7 factor; x is the number ofiIs the ith sample variable which has an influence relation with y; ω, b are coefficients that determine the relationship of x to y.
Further, the operation of kernel function K in the above formula is obtained by using gaussian radial basis function:
Figure BDA0002757725380000037
wherein sigma is a kernel width parameter and is a parameter to be optimized;
the process of solving the optimal regression function f (x) is as follows:
Figure BDA0002757725380000038
Figure BDA0002757725380000039
wherein ξi，ξi *Is a relaxation variable; c is a factor for balancing risk experience and learning machine complexity, epsilon is an insensitive loss parameter, and epsilon and C are parameters to be optimized;
the GA algorithm fitness function is as follows:
Figure BDA0002757725380000041
wherein the content of the first and second substances,
Figure BDA0002757725380000042
is the predicted value of the wind speed of the SVR model, yiIs a wind speed measurement; n is the iteration number;
setting three parameters of GA algorithm population quantity, evolution algebra and ditch generation;
and training the wind speed SVR model by using a GA algorithm, continuously optimizing sigma, C and epsilon parameters in the wind speed SVR model by using the GA algorithm until an iteration stopping condition is met by using the minimum fitness function of the GA algorithm as a target, and generating an optimal wind speed SVR estimation model.
The invention has the following beneficial effects: the invention establishes the mathematical relationship between the measurable variable and the variable to be measured, obtains the variable to be measured by utilizing the measurable variable, and provides possibility for solving the detection problem that the variable cannot be directly measured. The application of the invention provides a solution for solving the problem that the real wind speed of the fan cannot be directly measured through the measurable variable of the wind generating set. Analyzing the main effect by a main component analysis method, and eliminating the influence of dimension difference among different samples; the support vector machine (SVR) has strong generalization capability and can adapt to the large-range change of the wind speed; the Genetic Algorithm (GA) has outstanding global search advantages, and the wind speed SVR model is trained based on the genetic algorithm, so that the generated wind speed estimation model of the fan has high accuracy. The application of the invention can realize the ideal wind speed estimation of the fan and provides a way for improving the wind energy utilization and the control quality of the unit.
Drawings
FIG. 1 is an overall flow chart of the present invention.
FIG. 2 is a schematic diagram of the wind speed GA-SVR trainer in the present invention.
FIG. 3 is a schematic diagram of the wind speed estimation effect of the GA-SVR fan wind speed soft measurement system based on the principal component analysis method in the invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.
Fig. 1 shows a fan wind speed soft measurement method based on a principal component analysis method, which includes the following steps:
step one, 7 items of original operation data of the rotating speed, the power of a generator, the pitch angle, the yaw angle, the power of an adjacent fan, the wind direction and the ambient temperature of a wind turbine are collected by a data collector; and synthesizing an original operating data set D: storing 7 items of collected original operation data of rotating speed, generator power, pitch angle, yaw angle, adjacent fan power, wind direction and ambient temperature into column vectors D1, D2, D3, D4, D5, D6 and D7 respectively, and synthesizing an original operation data set D ═ D [ D4 ]1,D2,D3,D4,D5,D6,D7]；
Order to
Figure BDA0002757725380000051
And transmits the raw operational data set D to the principal component analysis processor.
Step two, a principal component analysis processor analyzes the principal effect, and performs data dimension reduction processing on an original operation data set D by using a principal component analysis method to generate a dimension reduction data set S:
the normalized matrix Z is calculated using the original running data set,
Figure BDA0002757725380000052
Figure BDA0002757725380000053
wherein z isijRepresenting the value of the ith row and jth column in the matrix Z, dijRepresents the value of the ith row and the jth column in the matrix D,
Figure BDA0002757725380000054
represents the average value of j columns in the D matrix, sjRepresents the standard deviation;
Figure BDA0002757725380000055
and sjThe calculation method of (a) is shown as follows:
Figure BDA0002757725380000056
further utilizing the standardized matrix Z to perform dimension reduction processing on the original operation data set D, specifically comprising the following steps:
s1, calculating a correlation coefficient matrix rho of the standardized matrix Z; calculating characteristic value lambda of correlation coefficient matrix rho1＞λ2＞λ3＞…＞λ7Corresponding unit feature vector
Figure BDA0002757725380000057
S2、Calculating principal component contribution rate pi(ii) a Rate of contribution
Figure BDA0002757725380000061
Wherein lambda is a characteristic value of the correlation coefficient array rho;
s3, cumulative contribution rate of the first q principal components
Figure BDA0002757725380000062
Determining a q value according to the accumulated contribution rate eta value, wherein eta is more than 0.5 and less than 1;
s4, according to the formula
Figure BDA0002757725380000063
Determining a transformation matrix
Figure BDA0002757725380000064
In the formula, ZTRepresents the transpose of the normalized matrix Z;
s5, according to the formula
Figure BDA0002757725380000065
Carrying out dimensionality reduction processing on an original operation data set D to generate a dimensionality reduction data set S ═ S1,S2,…,Sq]。
And step three, randomly selecting 2/3 behavior training data sets Strain from the dimension reduction data set S, and forming a test data set Stest by the rest 1/3 rows. Transmitting the training data set Strain to a wind speed GA-SVR trainer, and transmitting the test data set Stest to an optimal wind speed SVR predictor:
as shown in FIG. 2, the method for generating the optimal wind speed SVR estimation model by the wind speed GA-SVR trainer comprises the following steps:
the wind speed SVR model regression function is: y ═ f (x) ═ ω · K (x · x)i) + b, where y is a wind speed measurement; x is a sample variable consisting of 7 factors of the rotating speed of the wind turbine, the power of the generator, the pitch angle, the yaw angle, the power of an adjacent fan, the wind direction and the ambient temperature; x is the number ofiIs the ith sample variable which has an influence relation with y; ω, b are coefficients that determine the relationship of x to y.
The operation of the kernel function K in the above formula adopts a Gaussian radial basisThe function finds:
Figure BDA0002757725380000066
wherein sigma is a kernel width parameter and is a parameter to be optimized;
the process of solving the optimal regression function f (x) is as follows:
Figure BDA0002757725380000067
Figure BDA0002757725380000071
wherein ξi，ξi *Is a relaxation variable; c is a factor for balancing risk experience and learning machine complexity, epsilon is an insensitive loss parameter, and epsilon and C are parameters to be optimized;
the GA algorithm fitness function is as follows:
Figure BDA0002757725380000072
wherein the content of the first and second substances,
Figure BDA0002757725380000073
is the predicted value of the wind speed of the SVR model, yiIs a wind speed measurement; n is the iteration number;
and setting three parameters of GA algorithm population quantity, evolution algebra and generation channels.
Fourthly, the wind speed GA-SVR trainer utilizes a GA algorithm to conduct optimization training on the wind speed SVR model until an optimal wind speed SVR estimation model is generated;
and training the wind speed SVR model by using a GA algorithm, continuously optimizing sigma, C and epsilon parameters in the wind speed SVR model by using the GA algorithm until an iteration stopping condition is met by using the minimum fitness function of the GA algorithm as a target, and generating an optimal wind speed SVR estimation model.
And step five, the optimal wind speed SVR predictor takes the optimal wind speed SVR prediction model obtained in the step four as a core, the test data set Stest is used as input, the optimal wind speed prediction is realized, and the optimal wind speed estimation value is transmitted to the variable speed wind driven generator controller so as to provide a basis for controlling the rotating speed.
The first embodiment,
Taking a certain wind turbine as an example, 1800 groups of operation data of the wind turbine, such as rotating speed, generator power, pitch angle, yaw angle, adjacent fan power, wind direction and ambient temperature, are acquired and stored as column vectors D1, D2, D3, D4, D5, D6 and D7 respectively to form a matrix D;
namely, it is
Figure BDA0002757725380000074
Computing matrix D standardized matrix
Figure BDA0002757725380000075
Computing matrix Z correlation coefficient array
Figure BDA0002757725380000081
Calculating characteristic value lambda of correlation coefficient matrix rho1＞λ2＞λ3＞…＞λ7Corresponding unit feature vector
Figure BDA0002757725380000082
Calculating the contribution rate of the principal component: rate of contribution
Figure BDA0002757725380000083
Cumulative contribution rate of the first q principal components
Figure BDA0002757725380000084
Taking eta equal to 0.85; determining q to be 4 according to the eta value;
according to
Figure BDA0002757725380000085
Determining a transformation matrix
Figure BDA0002757725380000086
According to
Figure BDA0002757725380000087
Performing dimensionality reduction on the original operation data set D to generate a dimensionality reduction data set S1800×4＝[S1,S2,S3,S4]；
Wherein S is1＝u11*D1+u21*D2+…+u71*D7，S2＝u12*D1+u22*D2+…+u72*D7，S3＝u13*D1+u23*D2+…+u73*D7，S4＝u14*D1+u24*D2+…+u74*D7。
From a reduced dimension dataset S1800x4In-line random selection of 1200 rows to form a data set StrainThe wind speed is input as a GA-SVR trainer; initializing sigma, C and epsilon parameters in the wind speed SVR model, setting the population quantity of GA algorithm to be 20, the evolution algebra to be 100 and the gully to be 0.9, and using a fitness function
Figure BDA0002757725380000088
And training the wind speed SVR model until the minimum value is a target, and generating the optimal wind speed SVR model until an iteration stop condition is met.
The remaining 600 rows constitute the data set StestAnd the wind speed is estimated as the input of an optimal wind speed SVR estimator by utilizing the optimal wind speed SVR model. The obtained result of the estimated wind speed is shown in fig. 3. As can be seen from FIG. 3, the difference between the wind speed estimated by the method of the present invention and the actual wind speed is not large, which shows that the method of the present invention can simulate the actual wind speed well, thereby improving the accuracy of the wind speed measurement model.
The above embodiments are not intended to limit the present invention, and the present invention is not limited to the above examples, and those skilled in the art may make variations, modifications, additions or substitutions within the technical scope of the present invention.

Claims (5)

1. A fan wind speed soft measurement method based on a principal component analysis method is characterized by comprising the following steps: the method comprises the following steps:
step one, 7 items of original operation data of the rotating speed, the power of a generator, the pitch angle, the yaw angle, the power of an adjacent fan, the wind direction and the ambient temperature of a wind turbine are collected by a data collector; synthesizing an original operation data set D, and transmitting the original operation data set D to a principal component analysis processor;
analyzing the main effect by a principal component analysis processor, and performing data dimension reduction processing on the original operation data set D by using a principal component analysis method to generate a dimension reduction data set S;
step three, randomly selecting 2/3 behavior training data set S from the dimension reduction data set StrainThe remaining 1/3 rows constitute test data set StestTraining data set StrainTransmitting to a wind speed GA-SVR trainer to obtain a test data set StestTransmitting to an optimal wind speed SVR predictor;
fourthly, the wind speed GA-SVR trainer utilizes a GA algorithm to conduct optimization training on the wind speed SVR model until an optimal wind speed SVR estimation model is generated;
step five, the optimal wind speed SVR predictor tests a data set S by taking the optimal wind speed SVR prediction model obtained in the step four as a coretestAnd the optimal wind speed estimation value is input, and is transmitted to the variable speed wind driven generator controller, so that a basis is provided for rotating speed control.
2. The fan wind speed soft measurement method based on the principal component analysis method according to claim 1, characterized in that: storing 7 items of collected original operation data of rotating speed, generator power, pitch angle, yaw angle, adjacent fan power, wind direction and ambient temperature into column vectors D1, D2, D3, D4, D5, D6 and D7 respectively, and synthesizing an original operation data set D ═ D [ D4 ]1,D2,D3,D4,D5,D6,D7]；
Order to
Figure FDA0002757725370000011
The normalized matrix Z is calculated using the original running data set,
Figure FDA0002757725370000021
Figure FDA0002757725370000022
wherein z isijRepresenting the value of the ith row and jth column in the matrix Z, dijRepresents the value of the ith row and the jth column in the matrix D,
Figure FDA0002757725370000023
represents the average value of j columns in the D matrix, sjRepresents the standard deviation;
Figure FDA0002757725370000024
and sjThe calculation method of (a) is shown as follows:
Figure FDA0002757725370000025
3. the fan wind speed soft measurement method based on the principal component analysis method according to claim 2, characterized in that: further utilizing the standardized matrix Z to perform dimension reduction processing on the original operation data set D, specifically comprising the following steps:
s1, calculating a correlation coefficient matrix rho of the standardized matrix Z; calculating characteristic value lambda of correlation coefficient matrix rho1＞λ2＞λ3＞…＞λ7Corresponding unit feature vector
Figure FDA0002757725370000026
S2, calculating principal component contribution rate pi(ii) a Rate of contribution
Figure FDA0002757725370000027
Wherein lambda is a characteristic value of the correlation coefficient array rho;
s3, cumulative contribution rate of the first q principal components
Figure FDA0002757725370000028
Determining a q value according to the accumulated contribution rate eta value, wherein eta is more than 0.5 and less than 1;
s4, according to the formula
Figure FDA0002757725370000029
Determining a transformation matrix
Figure FDA00027577253700000210
In the formula, ZTRepresents the transpose of the normalized matrix Z;
s5, according to the formula
Figure FDA00027577253700000211
Carrying out dimensionality reduction processing on an original operation data set D to generate a dimensionality reduction data set S ═ S1,S2,…,Sq]。
4. The fan wind speed soft measurement method based on the principal component analysis method according to claim 3, characterized in that: the method for generating the optimal wind speed SVR estimation model by the wind speed GA-SVR trainer comprises the following steps:
the wind speed SVR model regression function is: y ═ f (x) ═ ω · K (x · x)i) + b, where y is a wind speed measurement; x is a sample variable consisting of 7 factors of the rotating speed of the wind turbine, the power of the generator, the pitch angle, the yaw angle, the power of an adjacent fan, the wind direction and the ambient temperature; x is the number ofiIs the ith sample variable which has an influence relation with y; ω, b are coefficients that determine the relationship of x to y.
5. Principal component analysis-based according to claim 4The method for soft measurement of the wind speed of the fan is characterized by comprising the following steps: the operation of the kernel function K in the above formula is obtained by using a gaussian radial basis function:
Figure FDA0002757725370000031
wherein sigma is a kernel width parameter and is a parameter to be optimized;
the process of solving the optimal regression function f (x) is as follows:
Figure FDA0002757725370000032
Figure FDA0002757725370000033
wherein ξi，ξi *Is a relaxation variable; c is a factor for balancing risk experience and learning machine complexity, epsilon is an insensitive loss parameter, and epsilon and C are parameters to be optimized;
the GA algorithm fitness function is as follows:
Figure FDA0002757725370000034
wherein the content of the first and second substances,
Figure FDA0002757725370000035
is the predicted value of the wind speed of the SVR model, yiIs a wind speed measurement; n is the iteration number;
setting three parameters of GA algorithm population quantity, evolution algebra and ditch generation;
and training the wind speed SVR model by using a GA algorithm, continuously optimizing sigma, C and epsilon parameters in the wind speed SVR model by using the GA algorithm until an iteration stopping condition is met by using the minimum fitness function of the GA algorithm as a target, and generating an optimal wind speed SVR estimation model.
CN202011208128.9A 2020-11-03 2020-11-03 Fan wind speed soft measurement method based on principal component analysis method Pending CN112329344A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011208128.9A CN112329344A (en) 2020-11-03 2020-11-03 Fan wind speed soft measurement method based on principal component analysis method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011208128.9A CN112329344A (en) 2020-11-03 2020-11-03 Fan wind speed soft measurement method based on principal component analysis method

Publications (1)

Publication Number Publication Date
CN112329344A true CN112329344A (en) 2021-02-05

Family

ID=74324526

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011208128.9A Pending CN112329344A (en) 2020-11-03 2020-11-03 Fan wind speed soft measurement method based on principal component analysis method

Country Status (1)

Country Link
CN (1) CN112329344A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102005135A (en) * 2010-12-09 2011-04-06 上海海事大学 Genetic algorithm-based support vector regression shipping traffic flow prediction method
CN106979126A (en) * 2017-04-12 2017-07-25 浙江大学 Wind power generating set high wind speed section effective wind speed method of estimation based on SVR
CN107045574A (en) * 2017-04-12 2017-08-15 浙江大学 The low wind speed section effective wind speed method of estimation of wind power generating set based on SVR
US20180046924A1 (en) * 2015-08-31 2018-02-15 Guangzhou Institute Of Energy Conversion, Chinese Academy Of Sciences Whole-life-cycle power output classification prediction system for photovoltaic systems

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102005135A (en) * 2010-12-09 2011-04-06 上海海事大学 Genetic algorithm-based support vector regression shipping traffic flow prediction method
US20180046924A1 (en) * 2015-08-31 2018-02-15 Guangzhou Institute Of Energy Conversion, Chinese Academy Of Sciences Whole-life-cycle power output classification prediction system for photovoltaic systems
CN106979126A (en) * 2017-04-12 2017-07-25 浙江大学 Wind power generating set high wind speed section effective wind speed method of estimation based on SVR
CN107045574A (en) * 2017-04-12 2017-08-15 浙江大学 The low wind speed section effective wind speed method of estimation of wind power generating set based on SVR

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
任化准: "基于主成分分析的GA-SVM模型在中长期径流预报中的应用", 《水利科技与经济》 *
余金等: "基于GA-SVR的永磁同步风电机组多工况噪声预测", 《可再生能源》 *
张佼等: "基于遗传算法优化支持向量回归机参数的供热负荷预测", 《暖通空调》 *
张泽龙等: "基于主成分分析与遗传优化BP神经网络的风电场短期功率预测研究", 《宁夏电力》 *

Similar Documents

Publication Publication Date Title
CN110807554A (en) Generation method and system based on wind power/photovoltaic classical scene set
CN106295908A (en) A kind of SVM wind power forecasting method
Ehsan et al. Wind speed prediction and visualization using long short-term memory networks (LSTM)
CN113344288A (en) Method and device for predicting water level of cascade hydropower station group and computer readable storage medium
CN116014722A (en) Sub-solar photovoltaic power generation prediction method and system based on seasonal decomposition and convolution network
Osman et al. Soft sensor modeling of key effluent parameters in wastewater treatment process based on SAE-NN
CN116454863A (en) Optimal weight determining method of wind power combination prediction model based on improved hawk optimization algorithm
CN108052963A (en) The data screening method, apparatus and wind power generating set of wind power prediction modeling
CN116599050A (en) Photovoltaic prediction method and related device based on self-attention mechanism
CN116865343B (en) Model-free self-adaptive control method, device and medium for distributed photovoltaic power distribution network
CN112329344A (en) Fan wind speed soft measurement method based on principal component analysis method
CN110414734B (en) Method for forecasting and evaluating wind resource utilization rate
CN116292057A (en) Control method, device, equipment and medium for variable-speed pumped storage unit
CN116054240A (en) Wind power grid-connected operation control optimization method and system based on power prediction
CN115977874A (en) Wind turbine generator yaw self-adaptive calibration method and system based on laser wind finding radar
CN113642784B (en) Wind power ultra-short-term prediction method considering fan state
CN113151842B (en) Method and device for determining conversion efficiency of wind-solar complementary water electrolysis hydrogen production
CN114565186A (en) Method and system for correcting predicted wind speed based on generalized regression neural network
Bouzem et al. Probabilistic and Reliability Analysis of an Intelligent Power Control for a Doubly Fed Induction Generator-Based Wind Turbine System
CN112232598A (en) Output power combination prediction method and system for progressive photovoltaic power generation system
CN112581311A (en) Method and system for predicting long-term output fluctuation characteristics of aggregated multiple wind power plants
CN111382895A (en) Seawater desalination load prediction method and system
He et al. A Novel Soft Sensor Model Based on Stacking Ensemble Learning Framework
Guo et al. Modelling for multi-phase batch processes using steady state identification and deep recurrent neural network
CN116613736B (en) New energy power system equivalent inertia prediction method containing virtual inertia

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20210205