CN117117858B - Wind turbine generator power prediction method, device and storage medium - Google Patents
Wind turbine generator power prediction method, device and storage medium Download PDFInfo
- Publication number
- CN117117858B CN117117858B CN202311346538.3A CN202311346538A CN117117858B CN 117117858 B CN117117858 B CN 117117858B CN 202311346538 A CN202311346538 A CN 202311346538A CN 117117858 B CN117117858 B CN 117117858B
- Authority
- CN
- China
- Prior art keywords
- wind speed
- power
- wind
- regression tree
- wind turbine
- 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 86
- 238000012549 training Methods 0.000 claims abstract description 60
- 238000004590 computer program Methods 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 5
- 230000009191 jumping Effects 0.000 claims description 4
- 230000004931 aggregating effect Effects 0.000 claims description 3
- 239000003570 air Substances 0.000 claims description 3
- 239000012080 ambient air Substances 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 4
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 4
- 238000010248 power generation Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- 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
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Power Engineering (AREA)
- Wind Motors (AREA)
Abstract
The invention relates to the technical field of fan unit power prediction, in particular to a method, a device and equipment for predicting the power of a wind turbine unit and a computer storage medium. According to the wind turbine generator power prediction method, interval division in a parameter method is combined with a gradient lifting regression tree in a non-parameter method, and data in each wind speed area is input as a training set of the gradient lifting regression tree to obtain reference power of each wind speed area and used for fitting a power curve; the invention also improves the gradient lifting regression tree by k neighbor weighted average so as to improve the contribution value of the sample which is closer to the predicted sample, thereby ensuring more accurate prediction effect.
Description
Technical Field
The invention relates to the technical field of fan unit power prediction, in particular to a method, a device and equipment for predicting the power of a wind turbine unit and a computer storage medium.
Background
In the modern society of lack of traditional fossil energy resources and serious pollution, wind energy is widely favored by the masses as a new energy source without pollution and renewable energy, and the wind power industry is one of novel renewable energy industries which are greatly developed at home and abroad. In China, the construction and related research work of wind power plants in recent decades are improved remarkably in quantity and quality, but a series of negative factors caused by continuous degradation of the wind turbine are accompanied while the wind power generation industry is developed greatly. In the use process of the wind driven generator, the wind speed has the characteristics of intermittence and high uncertainty, so that the performance evaluation of the wind driven generator is greatly influenced, and the correct evaluation and diagnosis of the performance and health condition of the wind driven generator are important points for reasonable planning in the aspect of wind power generation operation and maintenance.
In the actual running of the wind farm, the wind farm is affected by the environment, the state of the machine set, the operation mode, the power grid dispatching and the like, and the wind farm cannot keep higher efficiency level in each period, so that the generated energy is lost. The modeling of the power curve of the wind turbine generator is used for predicting the power of a fan, and is a necessary link for evaluating the power generation efficiency of a wind farm and searching an efficiency improvement path. The modeling method of the power curve of the wind turbine generator is divided into a parameter method and a non-parameter method. The parameter method mainly comprises a piecewise average method (IEC), a piecewise linear model method, polynomial fitting, multi-parameter logistic function regression and the like; the non-parametric method mainly comprises fuzzy logic regression, a neural network, a K nearest neighbor method and the like, however, the conventional parametric method is not accurate enough for a power estimated value and is greatly influenced by outliers, and the non-parametric method needs to perform a large amount of iterative computation, so that modeling needs to consume a large amount of time under large-scale data.
Disclosure of Invention
Therefore, the invention aims to solve the technical problem of lower power prediction precision in the prior art.
In order to solve the technical problems, the invention provides a wind turbine generator power prediction method, which comprises the following steps:
acquiring a wind power plant fan state data set, and dividing the wind power plant fan state data set according to a wind speed region;
taking the data in each wind speed area as training input of a gradient lifting regression tree to obtain reference power of each wind speed area;
fitting reference power of a plurality of wind speed areas by using a least square method to obtain a power curve of the wind turbine;
and predicting the power of the wind turbine according to the power curve of the wind turbine.
Preferably, the obtaining a wind farm fan state data set, and dividing the wind farm fan state data set according to a wind speed region includes:
acquiring a fan state data set of a wind power plantWherein, the fan state at the ith moment +.>,/>Indicating wind speed, & lt & gt>Representing active power, +.>Indicating ambient air pressure +.>Representing ambient temperature;
initializing a wind speed zone setWherein the number of wind speed areas,/>And->The cut-in wind speed and the cut-out wind speed of the wind power plant fan are respectively the kth wind speed zone,/>Representing a set of fan status data contained in a kth wind speed zone,/>Center wind speed, which represents the kth wind speed zone, is->Representing a reference power for a kth wind speed zone;
traversing the wind farm fan state data setAccording to the status of the fan at the i-th moment->Wind speed of (2)Calculating wind speed interval number +.>And let the fan status at the i-th moment +.>Added to the kth wind speed zone->Fan status data set->Is a kind of medium.
Preferably, the acquiring the wind farm fan state data set further includes:
the active power is setCorrected to the value of power at standard atmospheric pressure and standard ambient temperature +.>:
Wherein,for air density->Atmospheric pressure>For ambient temperature->Represents the standard atmospheric pressure, +.>Indicating a standard ambient temperature.
Preferably, the gradient lifting regression tree is a gradient lifting regression tree improved by k-nearest neighbor weighted average.
Preferably, the step of obtaining the reference power of each wind speed region by using the data in each wind speed region as the training input of the gradient lifting regression tree includes:
step a, wind speed in the t-th wind speed zoneAs training set related variable +.>Active power +.>Target variable +.>Obtaining a training set and initializing a weak learner;
step b, calculating the residual error of the current regression tree model;
Step c, taking the residual error as a training set target variable to obtain a new training set, and fitting by using a cart algorithm to obtain an mth regression tree;
step d, calculating leaf node areas of the mth regression treeThe distance between all training samples and the sample mean value is set up;
step e, screening K training samples with the minimum distance, and calculating the weight of each training sample;
f, calculating a predicted value of a leaf node of the mth regression tree, and updating a strong learner;
step g, when the current training times are not less than the maximum training times, obtaining a final regression tree;
step h, calculating the current wind speed areaIs>And inputting the final regression tree to obtain the reference power of the current wind speed region.
Preferably, the calculation of the current wind speed zoneIs>And inputting the final regression tree, and obtaining the reference power of the current wind speed region, wherein the method further comprises the following steps:
updating a wind speed-power counter t=t+1;
when (when)And (c) jumping to the step a.
Preferably, the fitting the reference powers of the wind speed areas by using a least square method to obtain a power curve of the wind turbine includes:
aggregating the wind speed regionsIn kth wind speed zone->Is>As abscissa, reference power +.>And performing power curve fitting by using a least square method as an ordinate to obtain the power curve of the wind turbine generator.
The invention also provides a wind turbine generator power prediction device, which comprises:
the data set dividing module is used for acquiring a wind power plant fan state data set and dividing the wind power plant fan state data set according to a wind speed region;
the reference power calculation module is used for taking the data in each wind speed area as the training input of the gradient lifting regression tree to obtain the reference power of each wind speed area;
the power curve fitting module is used for fitting reference power of a plurality of wind speed areas by using a least square method to obtain a power curve of the wind turbine;
and the power prediction module is used for predicting the power of the wind turbine according to the power curve of the wind turbine.
The invention also provides a wind turbine generator power prediction device, which comprises:
a memory for storing a computer program;
and the processor is used for realizing the steps of the wind turbine generator power prediction method when executing the computer program.
The invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program realizes the steps of the wind turbine generator power prediction method when being executed by a processor.
Compared with the prior art, the technical scheme of the invention has the following advantages:
according to the wind turbine generator power prediction method, interval division in a parameter method is combined with a gradient lifting regression tree in a non-parameter method, and data in each wind speed area is input as a training set of the gradient lifting regression tree to obtain reference power of each wind speed area and used for fitting a power curve; the invention also improves the gradient lifting regression tree by k neighbor weighted average so as to improve the contribution value of the sample which is closer to the predicted sample, thereby ensuring more accurate prediction effect.
Drawings
In order that the invention may be more readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings, in which:
FIG. 1 is a flowchart of a method for predicting power of a wind turbine provided by the invention;
fig. 2 is a flowchart of a wind turbine power prediction implementation provided in an embodiment of the present invention.
Detailed Description
The core of the invention is to provide a method, a device, equipment and a computer storage medium for predicting the power of a wind turbine, so that the accuracy of power prediction is effectively improved.
In order to better understand the aspects of the present invention, the present invention will be described in further detail with reference to the accompanying drawings and detailed description. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention mainly aims to improve the accuracy of fan unit power prediction. The interval division-based method is to divide wind speeds into different wind speed intervals, and obtain an estimated value of power in the different wind speed intervals by a certain means. In the conventional power curve modeling method, a piecewise mean method (IEC), a maximum probability point method and a piecewise linear model belong to the same class of methods, but the methods are not accurate enough for power estimation values and are greatly influenced by outliers. The method provided by the invention is used for power prediction, and the interval division is combined with the machine learning method so as to improve the accuracy of power prediction.
The invention provides a wind turbine generator power prediction method based on an improved gradient lifting regression tree, which comprises the main ideas: after a wind power plant fan data set is acquired, correcting power into power under standard atmospheric pressure and standard environmental temperature, partitioning the data set according to wind speed regions, obtaining a regression tree function by using an improved gradient lifting regression tree for wind power data of each wind speed region, inputting the central wind speed of the wind speed region to obtain a reference power, and fitting the points by using a least square method after obtaining the reference power of a plurality of wind speed regions to obtain a power curve for power prediction of a wind turbine generator:
referring to fig. 1, fig. 1 is a flowchart illustrating an implementation method of a wind turbine power prediction method according to the present invention; the specific operation steps are as follows:
s101, acquiring a wind power plant fan state data set, and dividing the wind power plant fan state data set according to a wind speed region;
s102, taking data in each wind speed area as training input of a gradient lifting regression tree to obtain reference power of each wind speed area;
s103, fitting reference power of a plurality of wind speed areas by using a least square method to obtain a power curve of the wind turbine generator;
s104, predicting the power of the wind turbine according to the power curve of the wind turbine.
Based on the above embodiments, the present embodiment describes step S101 in detail:
acquiring a fan state data set of a wind power plantWherein, the fan state at the ith moment +.>,/>Indicating wind speed, & lt & gt>Representing active power, +.>Indicating ambient air pressure +.>Representing ambient temperature;
initializing a wind speed zone setWherein the number of wind speed areas,/>And->The cut-in wind speed and the cut-out wind speed of the wind power plant fan are respectively the kth wind speed zone,/>Representing a fan status data set contained in the kth wind speed zone, initialized to null, ++>Center wind speed, which represents the kth wind speed zone, is->A reference power (initially 0) representing a kth wind speed zone, an initial wind speed power counter t=1;
traversing the wind farm fan state data setAccording to the status of the fan at the i-th moment->Wind speed of (2)Calculating wind speed interval number +.>And let the fan status at the i-th moment +.>Added to the kth wind speed zone->Fan status data set->Is a kind of medium.
Based on the above embodiment, the obtaining the wind farm fan state data set further includes:
the active power is setCorrected to the value of power at standard atmospheric pressure and standard ambient temperature +.>:
Wherein,for air density->Atmospheric pressure>For ambient temperature->Represents a standard atmospheric pressure of 101.325kPa, & lt/EN & gt>Represents a standard ambient temperature of 20 ℃.
Based on the above embodiments, the present embodiment describes step S102 in detail:
the gradient lifting regression tree is a gradient lifting regression tree improved by k neighbor weighted average.
The step of obtaining the reference power of each wind speed zone by taking the data in each wind speed zone as the training input of the gradient lifting regression tree comprises the following steps:
step a, wind speed in the t-th wind speed zoneAs training set related variable +.>Active power +.>Target variable +.>Obtaining a training set and initializing a weak learner;
from a wind speed zone collectionSelecting the t-th wind speed zone +.>Wind power data set->Wind speed->As training set related variable +.>The corrected power value is +.>Target variable +.>Obtaining input training set +.>The number of training data set samples N, the maximum training times is M, and the loss function L. And calculates the initialization weak learner +.>。
Step b, calculating the current regression treeModel residual->Where m=1, 2, … …, M, i=1, 2, … … N.
Step c, taking the residual error as a training set target variable to obtain a new training setFitting by using a cart algorithm to obtain an mth regression tree +.>;
Step d, calculating leaf node areas of the mth regression treeAll training samplesMean value of sample->Distance between:
where j=1, 2, … …, J is the number of leaf nodes of the regression tree m, S isThe number of samples;
step e, screening K training samples with the minimum distance, and calculating the weight of each training sample of the K training samples:
find distanceMinimum ofK training samples, and recording the output variable values of the K training samples asThey are +.>Distance of +.>The weight of each training sample is +.>Weight->Calculated according to the following formula:
f, calculating a predicted value of a leaf node of the mth regression tree, and updating a strong learner:
and g, when the current training times are not less than the maximum training times, obtaining a final regression tree:
and (c) judging whether the current training times M is smaller than the maximum training times M, if so, jumping to the step a, otherwise, continuing to execute.
Step h, calculating the current wind speed areaIs>And input the final regressionTree->Obtaining the reference power of the current wind speed area>=/>:
Based on the above embodiment, the current wind speed zone is calculatedIs>And inputting the final regression tree, and obtaining the reference power of the current wind speed region, wherein the method further comprises the following steps:
updating a wind speed-power counter t=t+1;
when (when)And (c) jumping to the step a.
Based on the above embodiments, the present embodiment describes in detail step S103:
aggregating wind speed zonesIs->Is>As abscissa, reference power +.>And (3) as an ordinate, performing power curve fitting by using a least square method, and finally obtaining a power curve of the wind turbine, wherein the power curve is used for predicting the power of the wind turbine.
Step a adopts a method of combining wind speed partitions and gradient lifting regression trees, and data in each wind speed region is input as a training set of the gradient lifting regression tree to obtain reference power of each wind speed region, and the reference power is used for fitting a power curve and is a file in related work similar documents, so that the method has originality.
And d, e, f and h improve a gradient lifting regression tree algorithm, introduce a k neighbor weighted average method to replace simple average, calculate the distance between a training sample and a predicted sample on a leaf node, select the smallest k predicted sample, calculate the weight of the predicted sample, improve the contribution value of the training sample close to the predicted sample to the reference power of a predicted wind speed region, and have originality in related work similar documents.
The invention provides a prediction method based on an improved gradient lifting regression tree all the time, and a current mainstream wind turbine generator power curve modeling method is divided into a parameter method such as interval division and polynomial fitting and a non-parameter method such as machine learning. The main advantages of this strategy are: and combining interval division in the parameter method with a gradient lifting regression tree in the non-parameter method, and improving the gradient lifting regression tree by using k neighbor weighted average to improve the contribution value of a sample closer to a predicted sample, so that the predicted effect is closer.
For most wind power prediction methods, the main points to be considered are: screening and optimizing the data set and improving the modeling efficiency and accuracy. The present invention essentially differs from other prediction methods while taking these two points into account. The main differences are as follows:
1. the method combines the Bien method in the parameter method with the gradient lifting regression tree in the non-parameter method, the conventional parameter method is not accurate enough for the power estimated value and is greatly influenced by the outlier, and the non-parameter method needs to perform a large amount of iterative computation, so that modeling needs to consume a large amount of time under large-scale data. The invention combines a parameter method with a non-parameter method, and gives consideration to the efficiency and modeling precision of wind power curve modeling;
2. and improving the gradient lifting regression tree by using k neighbor weighted average to improve the contribution value of samples closer to the predicted samples, so that the prediction effect is more similar, and the influence of outliers is further reduced to improve the prediction accuracy.
Referring to fig. 2, fig. 2 is a flowchart of a wind turbine power prediction implementation according to an embodiment of the present invention.
The embodiment of the invention also provides a wind turbine generator power prediction device; the specific apparatus may include:
the data set dividing module 100 is used for acquiring a wind power plant fan state data set and dividing the wind power plant fan state data set according to a wind speed region;
the reference power calculation module 200 is configured to use the data in each wind speed area as a training input of a gradient lifting regression tree to obtain a reference power of each wind speed area;
the power curve fitting module 300 is configured to fit reference powers of a plurality of wind speed regions by using a least square method, so as to obtain a power curve of the wind turbine;
and the power prediction module 400 is configured to predict the power of the wind turbine according to the power curve of the wind turbine.
The wind turbine power prediction device of the present embodiment is used to implement the foregoing wind turbine power prediction method, so that the specific implementation of the wind turbine power prediction device may be referred to the foregoing example portions of the wind turbine power prediction method, for example, the data set dividing module 100, the reference power calculating module 200, the power curve fitting module 300, and the power prediction module 400 are respectively used to implement steps S101, S102, S103, and S104 in the foregoing wind turbine power prediction method, so that the specific implementation thereof may refer to the description of the corresponding examples of each portion and will not be repeated herein.
The specific embodiment of the invention also provides a wind turbine generator power prediction device, which comprises: a memory for storing a computer program; and the processor is used for realizing the steps of the wind turbine generator power prediction method when executing the computer program.
The specific embodiment of the invention also provides a computer readable storage medium, wherein the computer readable storage medium is stored with a computer program, and the computer program realizes the steps of the wind turbine generator power prediction method when being executed by a processor.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations and modifications of the present invention will be apparent to those of ordinary skill in the art in light of the foregoing description. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (9)
1. The wind turbine generator power prediction method is characterized by comprising the following steps of:
acquiring a wind power plant fan state data set, and dividing the wind power plant fan state data set according to a wind speed region;
taking the data in each wind speed area as training input of a gradient lifting regression tree to obtain reference power of each wind speed area;
fitting reference power of a plurality of wind speed areas by using a least square method to obtain a power curve of the wind turbine;
predicting the power of the wind turbine according to the power curve of the wind turbine;
the dividing the wind farm fan state data set according to the wind speed zone comprises the following steps:
acquiring a wind power plant fan state data set;
initializing a wind speed zone set;
traversing the fan state data set of the wind power plant, calculating a wind speed interval number according to the wind speed of the fan state at the ith moment, and adding the fan state at the ith moment into the fan state data set of the kth wind speed interval;
the step of obtaining the reference power of each wind speed zone by taking the data in each wind speed zone as the training input of the gradient lifting regression tree comprises the following steps:
wind speed in the t-th wind speed zoneAs training set related variable +.>Active power +.>Target variable +.>Obtaining a training set and initializing a weak learner;
calculating the residual error of the current regression tree model;
Taking the residual error as a training set target variable to obtain a new training set, and fitting by using a cart algorithm to obtain an mth regression tree;
calculating leaf node areas of the mth regression treeThe distance between all training samples and the sample mean value is set up;
screening K training samples with the minimum distance, and calculating the weight of each training sample;
calculating a predicted value of a leaf node of the mth regression tree, and updating the strong learner;
when the current training times are not less than the maximum training times, obtaining a final regression tree;
calculating a current wind speed zoneIs>And inputting the final regression tree to obtain the reference power of the current wind speed region.
2. The method for predicting power of a wind turbine according to claim 1, wherein the obtaining a wind farm fan state data set and dividing the wind farm fan state data set according to a wind speed region comprises:
the wind farm fan state data set is thatWherein the fan state at the ith moment is +.>,/>Indicating wind speed, & lt & gt>Representing active power, +.>Indicating ambient air pressure +.>Representing ambient temperature;
the initialization wind speed zone set isWherein the number of wind speed areas,/>And->The wind speed is respectively the cut-in wind speed and the cut-out wind speed of a wind turbine of the wind power plant, and the kth wind speed area is +.>,/>Representing a set of fan status data contained within a kth wind speed zone,a center wind speed indicative of said kth wind speed zone, a>A reference power representing the kth wind speed zone;
the state of the fan at the ith momentIs +.>The wind speed interval number is +.>。
3. The method for predicting power of a wind turbine according to claim 2, wherein the step of obtaining the wind farm fan state data set further comprises:
the active power is setCorrected to the value of power at standard atmospheric pressure and standard ambient temperature +.>:
Wherein,for air density->Atmospheric pressure>For ambient temperature->Represents the standard atmospheric pressure, +.>Indicating a standard ambient temperature.
4. The wind turbine power prediction method according to claim 2, wherein the gradient lifting regression tree is a gradient lifting regression tree improved by k-nearest neighbor weighted average.
5. A wind turbine power prediction method according to claim 1, wherein the calculation of the current wind speed zoneIs>And inputting the final regression tree, and obtaining the reference power of the current wind speed region, wherein the method further comprises the following steps:
updating wind speed-power counter;
When (when)And (c) jumping to the step a.
6. The method for predicting power of a wind turbine according to claim 2, wherein the fitting the reference powers of the plurality of wind speed regions by using a least square method to obtain a power curve of the wind turbine comprises:
aggregating the wind speed regionsIn kth wind speed zone->Is>As abscissa, reference power +.>And performing power curve fitting by using a least square method as an ordinate to obtain the power curve of the wind turbine generator.
7. The utility model provides a wind turbine generator system power prediction device which characterized in that includes:
the data set dividing module is used for acquiring a wind power plant fan state data set and dividing the wind power plant fan state data set according to a wind speed region;
the reference power calculation module is used for taking the data in each wind speed area as the training input of the gradient lifting regression tree to obtain the reference power of each wind speed area;
the power curve fitting module is used for fitting reference power of a plurality of wind speed areas by using a least square method to obtain a power curve of the wind turbine;
the power prediction module is used for predicting the power of the wind turbine according to the power curve of the wind turbine;
wherein the data set partitioning module is further configured to: acquiring a wind power plant fan state data set, initializing a wind speed zone set, traversing the wind power plant fan state data set, calculating a wind speed zone number according to the wind speed of the fan state at the ith moment, and adding the fan state at the ith moment into the fan state data set of the kth wind speed zone;
the reference power calculation module is further configured to: wind speed in the t-th wind speed zoneAs training set related variable +.>Active power +.>Target variable +.>Obtaining a training set, initializing a weak learner, and calculating the residual error of the current regression tree modelObtaining a new training set by taking the residual error as a training set target variable, fitting by using a cart algorithm to obtain an mth regression tree, and calculating the leaf node area +.>Screening out K training samples with minimum distance from all training samples to sample mean values, calculating weights of the K training samples, calculating a predicted value of leaf nodes of an mth regression tree, updating a strong learner, obtaining a final regression tree when the current training times are not less than the maximum training times, and calculating a current wind speed zone->Is>And inputting the final regression tree to obtain the reference power of the current wind speed region.
8. A wind turbine power prediction apparatus, comprising:
a memory for storing a computer program;
a processor for implementing the steps of a method for predicting the power of a wind turbine as claimed in any one of claims 1 to 6 when executing said computer program.
9. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of a wind turbine power prediction method according to any of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311346538.3A CN117117858B (en) | 2023-10-18 | 2023-10-18 | Wind turbine generator power prediction method, device and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311346538.3A CN117117858B (en) | 2023-10-18 | 2023-10-18 | Wind turbine generator power prediction method, device and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117117858A CN117117858A (en) | 2023-11-24 |
CN117117858B true CN117117858B (en) | 2024-01-30 |
Family
ID=88805761
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311346538.3A Active CN117117858B (en) | 2023-10-18 | 2023-10-18 | Wind turbine generator power prediction method, device and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117117858B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011143531A2 (en) * | 2010-05-13 | 2011-11-17 | University Of Cincinnati | Turbine-to-turbine prognostics technique for wind farms |
CN111396266A (en) * | 2020-03-05 | 2020-07-10 | 浙江运达风电股份有限公司 | GBRT-based wind turbine generator bearing fault early warning method |
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 |
CN112818604A (en) * | 2021-02-05 | 2021-05-18 | 河海大学 | Wind turbine generator risk degree assessment method based on wind power prediction |
CN113313139A (en) * | 2021-04-13 | 2021-08-27 | 华北水利水电大学 | Wind power prediction uncertainty quantification method based on dynamic characteristics of unit |
CN114298140A (en) * | 2021-11-18 | 2022-04-08 | 华能新能源股份有限公司 | Wind power short-term power prediction correction method considering unit classification |
CN114330094A (en) * | 2021-11-18 | 2022-04-12 | 国家电网有限公司 | Wind power short-term prediction method based on TCN-GRU combined model |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11436395B2 (en) * | 2018-06-27 | 2022-09-06 | Dalian University Of Technology | Method for prediction of key performance parameter of an aero-engine transition state acceleration process based on space reconstruction |
-
2023
- 2023-10-18 CN CN202311346538.3A patent/CN117117858B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011143531A2 (en) * | 2010-05-13 | 2011-11-17 | University Of Cincinnati | Turbine-to-turbine prognostics technique for wind farms |
CN111396266A (en) * | 2020-03-05 | 2020-07-10 | 浙江运达风电股份有限公司 | GBRT-based wind turbine generator bearing fault early warning method |
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 |
CN112818604A (en) * | 2021-02-05 | 2021-05-18 | 河海大学 | Wind turbine generator risk degree assessment method based on wind power prediction |
CN113313139A (en) * | 2021-04-13 | 2021-08-27 | 华北水利水电大学 | Wind power prediction uncertainty quantification method based on dynamic characteristics of unit |
CN114298140A (en) * | 2021-11-18 | 2022-04-08 | 华能新能源股份有限公司 | Wind power short-term power prediction correction method considering unit classification |
CN114330094A (en) * | 2021-11-18 | 2022-04-12 | 国家电网有限公司 | Wind power short-term prediction method based on TCN-GRU combined model |
Also Published As
Publication number | Publication date |
---|---|
CN117117858A (en) | 2023-11-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113051769B (en) | Power curve modeling method based on asymmetric loss and hybrid intelligent optimization algorithm | |
CN111915079B (en) | Hybrid KNN wind power prediction method and system | |
CN114021483A (en) | Ultra-short-term wind power prediction method based on time domain characteristics and XGboost | |
CN115085202A (en) | Power grid multi-region intelligent power collaborative optimization method, device, equipment and medium | |
CN117195449A (en) | Wind farm arrangement optimization method, device, equipment and medium | |
CN116245019A (en) | Load prediction method, system, device and storage medium based on Bagging sampling and improved random forest algorithm | |
CN116599050A (en) | Photovoltaic prediction method and related device based on self-attention mechanism | |
CN113759708B (en) | Optimization control method and device of system and electronic equipment | |
CN110598947A (en) | Load prediction method based on improved cuckoo-neural network algorithm | |
CN111178601B (en) | Wind turbine generator power prediction method based on meteorological data post-processing | |
CN110210657B (en) | Fan power prediction method and system based on single machine model and computer storage medium | |
CN117117858B (en) | Wind turbine generator power prediction method, device and storage medium | |
CN117371581A (en) | New energy generated power prediction method, device and storage medium | |
CN104269886A (en) | Method used for calculating and distributing capacity benefit margins of wind power grid-connected system | |
CN116896093A (en) | Online analysis and optimization method for grid-connected oscillation stability of wind farm | |
CN116542385A (en) | Integrated learning method and system for wind and light power prediction | |
CN116154875A (en) | Photovoltaic power station active power optimization distribution method based on TCN and error function | |
CN104504466A (en) | Wind power plant power prediction method considering atmospheric disturbance effect | |
CN113807019A (en) | MCMC wind power simulation method based on improved scene classification and coarse grain removal | |
CN111178593A (en) | Photovoltaic system output power prediction method and device | |
CN112215383A (en) | Distributed photovoltaic power generation power prediction method and system | |
CN117893030B (en) | Power system risk early warning method based on big data | |
CN117057255B (en) | Pre-training model acquisition method for online synchronization of digital twin model | |
CN118195542A (en) | Electric power system project construction success prediction system and method | |
Ai et al. | Wind Power Prediction Based on Gray Wolf Algorithm and Long Short-Term Memory Neural Networks |
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 |