CN114418180A - Ultra-short-term prediction method and device for wind power and storage medium - Google Patents
Ultra-short-term prediction method and device for wind power and storage medium Download PDFInfo
- Publication number
- CN114418180A CN114418180A CN202111543759.0A CN202111543759A CN114418180A CN 114418180 A CN114418180 A CN 114418180A CN 202111543759 A CN202111543759 A CN 202111543759A CN 114418180 A CN114418180 A CN 114418180A
- Authority
- CN
- China
- Prior art keywords
- data
- wind power
- weather
- wind
- operation index
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
Abstract
The invention discloses an ultra-short-term prediction method and device for wind power and a storage medium. The method comprises the steps of calculating a first correlation degree between a weather factor and wind power and a second correlation degree between an operation index factor and the wind power according to historical weather data of a wind power plant, operation index data of a wind turbine generator and the wind power data, and determining key factors influencing the wind power; screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and outputting the predicted wind power at the current moment by the model; the wind power ultra-short term prediction model is an extreme learning machine model. The technical scheme of the invention realizes accurate prediction of ultra-short-term wind power.
Description
Technical Field
The invention relates to the technical field of wind power prediction, in particular to a method and a device for ultra-short-term prediction of wind power and a storage medium.
Background
China is rich in sea wind resources, and the rapid development of wind power is an important means for realizing 'carbon peak reaching' in 2030 and 'carbon neutralization' in 2060. Wind power generation is one of the most mature and scaled development conditions of new energy power generation. However, the integration of large-scale wind farms into the grid also brings about corresponding problems. However, due to the intermittency, randomness and fluctuation of wind resources, the fluctuation of wind power output can cause the frequency and voltage of a large wind power plant to fluctuate correspondingly. Therefore, the wind power is accurately predicted and used in the scheduling plan, and the premise that the wind power grid-connection capacity is improved and the operation safety and economy of a power grid are improved is provided. On the basis, the accuracy of the wind power prediction technology is improved, economic optimization scheduling in the wind power plant can be achieved, power fluctuation of the wind power plant is reduced, the power reliability is improved, and reliable inertial response, frequency and voltage support are further provided.
For wind power prediction, the traditional wind power prediction can be generally divided into a physical model method and a statistical model method. The large wind power plant has a plurality of units and different states. The influence of the state of the unit and natural factors on the wind power prediction is not negligible. And the quality of weather forecast data influences the accuracy of wind power prediction to a great extent, however, a weather system is an unstable power system and has a certain deviation from a true value. Therefore, according to historical data and weather data of the wind turbine generator, a grey correlation model is used for screening factors influencing wind power, two extreme learning machine models are established, error correction is carried out on wind speed in weather forecast data, and ultra-short-term prediction is carried out on the power of the wind turbine generator by considering unit operation index data.
Disclosure of Invention
The invention provides a method and a device for ultra-short-term prediction of wind power and a storage medium, which realize accurate prediction of ultra-short-term wind power.
An embodiment of the present invention provides an ultra-short term prediction method for wind power, including the following steps:
calculating a first association degree between a weather factor and wind power and a second association degree between an operation index factor and the wind power according to historical weather data of a wind power plant, operation index data of a wind turbine generator and the wind power data, and determining key factors influencing the wind power according to the first association degree and the second association degree; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data;
screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data;
inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and outputting the predicted wind power at the current moment by the wind power ultra-short term prediction model; the wind power ultra-short term prediction model is an extreme learning machine model.
Further, inputting first weather prediction data of the wind power plant at the current moment and a weather prediction error of the wind power plant at the previous moment into the wind speed correction model, correcting the wind speed in the first weather prediction data by the wind speed correction model according to the prediction error of the previous moment, and outputting second weather prediction data of the corrected current moment; the wind speed correction model is an extreme learning machine model.
Further, according to historical weather data of the wind power plant, operation index data of the wind turbine generator and wind power data, a first degree of correlation between weather factors and wind power and a second degree of correlation between operation index factors and wind power are calculated, and key factors influencing the wind power are determined according to the first degree of correlation and the second degree of correlation, and the method specifically comprises the following steps:
inputting historical weather data of a wind power plant, operation index data of the wind generation set and wind power data of the wind generation set into a wind power association degree model, calculating a first association degree between a weather factor and wind power and a second association degree between the operation index factor and the wind power according to the historical weather data of the wind power plant, the operation index data of the wind generation set and the wind power data by the wind power association degree model, and determining key factors influencing the wind power according to the first association degree and the second association degree, wherein the key factors comprise the weather key factors and the operation index key factors; the wind power correlation degree model is a grey correlation analysis model.
Furthermore, screening the weather prediction data of the wind power plant at the current moment according to the key factors to obtain first weather prediction data.
Further, the wind power relevance model calculates a first relevance between a weather factor and wind power and a second relevance between the operation index factor and the wind power according to historical weather data of the wind power plant, operation index data of the wind power unit and the wind power data, and respectively performs normalization processing on the historical weather data of the wind power plant, the operation index data of the wind power unit and the wind power data before the first relevance between the weather factor and the wind power and the second relevance between the operation index factor and the wind power are calculated.
Further, the weather factors greater than 0.85 in the first degree of association are determined as weather key factors, and the operation index factors greater than 0.85 in the second degree of association are determined as operation index key factors.
Further, historical weather data input into the wind power relevancy model are screened according to the weather key factors to obtain screened historical weather data, operation index data input into the wind power relevancy model are screened according to the operation index key factors to obtain screened operation index data, and the screened historical weather data and the screened operation index data are used as a training set of the wind power ultra-short term prediction model.
The invention further provides an ultra-short-term prediction device of wind power, which comprises a key factor calculation module, a data screening module and a wind power prediction module.
The key factor calculation module is used for calculating a first correlation degree between the weather factors and the wind power and a second correlation degree between the operation index factors and the wind power according to historical weather data of the wind power plant, operation index data of the wind generation set and the wind power data, and determining key factors influencing the wind power according to the first correlation degree and the second correlation degree; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data;
the data screening module is used for screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data;
the wind power prediction module is used for inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and the wind power ultra-short term prediction model outputs the predicted wind power at the current moment; the wind power ultra-short term prediction model is an extreme learning machine model.
The wind speed correction module is used for inputting first weather prediction data of the wind power plant at the current moment and a weather prediction error at the previous moment into the wind speed correction model, correcting the wind speed in the first weather prediction data according to the prediction error at the previous moment by the wind speed correction model, and outputting second weather prediction data at the current moment after correction; the wind speed correction model is an extreme learning machine model.
Yet another embodiment of the present invention provides a readable storage medium, where the readable storage medium includes a stored computer program, and when the computer program is executed, the apparatus where the readable storage medium is located is controlled to execute the method for ultra-short term prediction of wind power according to any one of the method items in the present invention.
The embodiment of the invention has the following beneficial effects:
the invention provides a method and a device for ultra-short-term prediction of wind power, which are characterized in that a first correlation degree between a weather factor and wind power and a second correlation degree between an operation index factor and the wind power are calculated according to historical weather data of a wind power plant, operation index data of a wind turbine generator and the wind power data, and key factors influencing the wind power are determined according to the first correlation degree and the second correlation degree; screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; therefore, the invention can input the data of the model by determining the key factors in advance, namely, the training efficiency of the model and the accuracy of the model prediction are improved. Inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and outputting the predicted wind power at the current moment by the wind power ultra-short term prediction model; and the ultra-short-term prediction of the wind power is realized.
Drawings
Fig. 1 is a schematic flow chart of a method for ultra-short term prediction of wind power according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an ultra-short term wind power prediction apparatus according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a wind speed correction model according to an embodiment of the present invention.
Detailed Description
The technical solutions in the present invention will be described clearly and completely with reference to the accompanying drawings, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an ultra-short term prediction method for wind power provided in an embodiment of the present invention includes the following steps:
step S101, calculating a first correlation degree between a weather factor and wind power and a second correlation degree between an operation index factor and the wind power according to historical weather data of a wind power plant, operation index data of a wind turbine generator and the wind power data, and determining key factors influencing the wind power according to the first correlation degree and the second correlation degree; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data.
As one embodiment, a first degree of association between a weather factor and wind power and a second degree of association between an operation index factor and the wind power are calculated according to historical weather data of a wind farm, operation index data of a wind turbine generator and the wind power data, and a key factor influencing the wind power is determined according to the first degree of association and the second degree of association, specifically:
inputting historical weather data of a wind power plant, operation index data of the wind generation set and wind power data of the wind generation set into a wind power association degree model, calculating a first association degree between a weather factor and wind power and a second association degree between the operation index factor and the wind power according to the historical weather data of the wind power plant, the operation index data of the wind generation set and the wind power data by the wind power association degree model, and determining key factors influencing the wind power according to the first association degree and the second association degree, wherein the key factors comprise the weather key factors and the operation index key factors; the wind power correlation degree model is a grey correlation analysis model.
As one embodiment, before the wind power relevance model calculates a first relevance between a weather factor and wind power and a second relevance between an operation index factor and the wind power according to historical weather data of a wind farm, operation index data of a wind turbine generator and the wind power data, the historical weather data of the wind farm, the operation index data of the wind turbine generator and the wind power data are respectively normalized.
Preferably, the weather factor greater than 0.85 in the first degree of association is determined as a weather key factor, and the operation index factor greater than 0.85 in the second degree of association is determined as an operation index key factor.
As one embodiment, the historical weather data input into the wind power relevancy model is screened according to the weather key factors to obtain screened historical weather data, the operation index data input into the wind power relevancy model is screened according to the operation index key factors to obtain screened operation index data, and the screened historical weather data and the screened operation index data are used as a training set of the wind power ultra-short term prediction model.
Step S102, screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data.
As one embodiment, in step S102, screening weather data of the wind farm at the current time according to the weather key factor to obtain first weather prediction data, inputting the first weather prediction data of the wind farm at the current time and a weather prediction error of a previous time into the wind speed correction model, correcting the wind speed in the first weather prediction data according to the prediction error of the previous time by the wind speed correction model, and outputting second weather prediction data of the corrected current time; the wind speed correction model is an extreme learning machine model.
Step S103, inputting the third weather forecast data and the third operation index data into a wind power ultra-short term forecast model, and outputting the forecast wind power at the current moment by the wind power ultra-short term forecast model; the wind power ultra-short term prediction model is an extreme learning machine model.
As one example of more details therein, step a 01: the method comprises the steps of constructing a wind power association degree model according to a grey association analysis method, collecting historical weather data of a wind power plant within a period of time and operation index data of a wind turbine generator set corresponding to the historical weather data as first sample data, wherein the time interval of the first sample data is 10 minutes. Corresponding weather factors can be determined according to the weather data, and corresponding operation index factors can be determined according to the operation index data. Establishing a sample data matrix of formula (1) according to the collected sample data:
wherein m is the number of various data sequences, npThe weather factor type corresponding to the historical weather data; t isi＝[Ti(1),Ti(2),…,Ti(mp),Ti(mp+1),…,Ti(m)]T，TiAnd the influence factors are the ith influence factors, and the influence factors comprise weather factors and operation index factors.
Determining sequence data to be associated, wherein the sequence data to be associated is historical wind power data to be associated, and the following formula is shown:
P0＝[P0(1) P0(2) P0(m)]T (2)；
due to the different physical meanings of the various factors in the obtained data, the dimensions of the data are different, the comparison is inconvenient, or a correct conclusion is difficult to obtain in the comparison. Therefore, when performing the gray correlation analysis, the data is subjected to non-dimensionalization processing, that is, normalization processing is performed on each of equations (1) and (2), where the normalization processing is performed by the following equation:
wherein, T*The normalized data is obtained; t is data to be normalized; t ismin，TmaxRespectively the maximum and minimum of the data to be normalized.
The normalized data sequence is shown below:
calculating the correlation coefficient xi of each factor and wind power according to the expressions (1) and (2)iThe factors comprise weather factors and operation index factors, and the calculation method is shown as the formula (5):
wherein i is historical weather data and an index of the type of an operation data sequence of the wind turbine generator; k is a sequence index of historical weather data and operation index data; ρ is a resolution coefficient, 0< ρ < 1. If rho is smaller, the difference between the correlation coefficients is larger, and the distinguishing capability is stronger; preferably, ρ is taken to be 0.5;
calculating the association degree epsilon of each factor according to the formula (6), namely calculating the mean value of the association coefficient of each factor and the wind power;
wherein m is the number of various data sequences, and k is the sequence index of historical weather data and operation index data.
As shown in the following table, wind speed, wind direction, ambient temperature, air pressure, air density and humidity are weather factors, and the rotating speed of the generator, the temperature of the cabin and the angle of the blade are operation index factors of the wind turbine generator.
Preferably, the weather factor with the degree of association greater than 0.85 is determined as a weather key factor, and the operation index factor with the degree of association greater than 0.85 is determined as an operation index key factor.
And screening historical weather data input into the wind power relevancy model according to the weather key factors to obtain screened historical weather data, screening operation index data input into the wind power relevancy model according to the operation index key factors to obtain screened operation index data, and taking the screened historical weather data and the screened operation index data as a training set of the wind power ultra-short term prediction model.
Step A02: constructing a wind speed correction model, which is an Extreme Learning Machine (ELM) model, as shown in FIG. 3; considering that the quality of weather forecast data greatly influences the accuracy of wind power prediction, a weather system is an unstable dynamic system and has a certain deviation from a true value. Therefore, according to the weather data of the historical wind power plant, a wind speed correction model is established to correct errors of wind speed in the weather prediction data.
Determining a sample set, establishing a wind speed correction model between a wind speed prediction result and an actually measured wind speed result, inputting sample variables mainly comprising weather prediction data and a weather prediction error at the previous moment, and outputting an output variable comprising corrected wind speed, wherein the input sample variables comprise the following formula:
in the formula, v* tFor correcting the corrected wind speed, Tt,iIs weather forecast data corresponding to the ith weather factor at time t, Et-1The weather prediction error at the time t-1.
The wind speed correction model of the embodiment can analyze the output weight of the network by one-step calculation, greatly improves the generalization capability and the learning speed of the network, has strong nonlinear fitting capability, and greatly reduces the calculated amount and the search space.
The training set of the wind speed correction model is { xp，up|xp∈RD，tp∈RGP ═ 1,2, …, S }; wherein xpInput layer data representing the p-th data sample, upAnd representing the output layer data of the p-th data sample, wherein the number of hidden layer nodes of the wind speed correction model is L. As shown in fig. 3, from left to right, the input layer, the hidden layer, the output layer are respectively and fully connected, and the calculation formula of the hidden layer output matrix H is as follows:
H＝[h1(x)，…，hL(x)]T (8)；
hj(x)＝g(wj,bj,xp)＝g(wjxp+bj) (9)；
wherein, in the formula (9), g (w)jxp+bj) To activate a function, wjAnd bjThe variables are unknown quantities, and the unknown quantities satisfy a nonlinear piecewise continuous function of a general approximation capability theorem in an Extreme Learning Machine (ELM) model, preferably, the activation function is a Sigmoid function, and the Sigmoid function specifically is as follows:
and when the data enters the output layer through the hidden layer, processing the data of the output layer according to the formulas (11) and (12):
β＝[β1，...，βL]T (12)；
wherein β is the weight of the output layer.
Determining w using a randomly generated methodj、bjValue of (a), betajThe weights of the output layers are obtained by training according to the training set. For example, the training set has S training samples, and the relation between the target matrix U of the S training samples and the output layer weight β is obtained after the wind speed correction model calculation is performed on the S training samples. And obtaining the trained output layer weight beta by further solving:
β*＝H+U (13)；
wherein H+Is the generalized inverse of matrix H and U is the target matrix.
Obtaining a trained wind speed correction model through the steps, and processing the trained wind speed correction model as follows:
and introducing a kernel function into the wind speed correction model to construct a new wind speed correction model, namely a kernel limit learning model, wherein the generalization capability and the stability of the model can be enhanced by introducing the kernel function. The kernel matrix is defined using Mercer's conditions:
replacement of random matrix HH in ELM with kernel matrix omegaTAll input samples are mapped from the n-dimensional input space to the high-dimensional hidden layer feature space using a kernel function. The type of kernel function adopted in this embodiment is a gaussian kernel function, and its expression is:
to further enhance the stability and generalization ability of the network, based on the ridge regression concept, in HHTThe small positive number I/C is added to the diagonal of (b), the trained output layer weight β becomes:
β*＝HT(I/C+HHT)-1U (16)；
wherein, I is a unit diagonal matrix, and C is a penalty coefficient.
From equations (13) to (15), the final predicted sample output of the wind speed correction model is:
step A03: an extreme learning machine model (ELM) is adopted to construct a wind power ultra-short term prediction model, and the wind power ultra-short term prediction model is as follows:
wherein the content of the first and second substances,
for ultra-short-term wind power prediction results, T*t,gFor the category g weather forecast data at time t,
is the h-type wind turbine generator operation index data v at t-1 moment* tIs the corrected wind speed.
The power of the wind turbine generator is determined by natural factors such as wind speed and wind direction, and the state and health degree of the wind turbine generator. Meanwhile, when wind power prediction is carried out, a plurality of units are arranged in a field group, and the states of the units are different, so that the active power output of the wind turbine generator is subjected to ultra-short-term prediction by using an extreme learning machine based on corrected weather prediction data and historical data and by considering operation index data of the wind turbine generator.
Screening historical weather data according to the key factors obtained in the step A01 to obtain first historical weather data, inputting the first historical weather data and the corresponding prediction error into the wind speed correction model in the step A02 to obtain second historical weather data after the wind speed is corrected; or inputting the historical weather data and the corresponding prediction error into the wind speed correction model in the step A02 to obtain third historical weather data after the wind speed is corrected, and screening the third historical weather data according to the key factors obtained in the step A01 to obtain fourth historical weather data.
Establishing a training sample set according to the second historical weather data and the operation index data of the corresponding wind turbine generator, or establishing a training sample set according to the fourth historical weather data and the operation index data of the corresponding wind turbine generator:
wherein, npIs the number of the key factors of weather, and n' is the number of the key factors.
And training the wind power ultra-short term prediction model according to the sample training set, and preferably updating the wind power ultra-short term prediction model once a day.
Obtaining the weight of an output layer of the wind power ultra-short term prediction model according to the sample training set as follows:
obtaining the prediction data output by the wind power ultra-short term prediction model according to the output layer weight as follows:
wherein, Ttrain,t-1Is the training sample at time t-1.
When the wind power ultra-short term prediction model is used for predicting wind power, inputting first weather prediction data of a wind power plant at the current moment and a weather prediction error at the previous moment into the wind speed correction model, correcting the wind speed in the first weather prediction data by the wind speed correction model according to the prediction error at the previous moment, and outputting second weather prediction data at the current moment after correction;
screening second weather prediction data at the current moment and operation index data of the wind turbine generator corresponding to the second weather prediction data according to the key factors to obtain third weather prediction data and third operation index data; the second operation index data is operation index data of the wind turbine generator at the last moment;
constructing a training sample set according to the corrected third weather forecast data and the third operation index number at the current moment:
and inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and outputting the predicted wind power at the current moment by the wind power ultra-short term prediction model.
The embodiment of the invention has the advantages that key factors can be selected for prediction, the dimension of model variables is reduced, the training efficiency is improved, error correction is carried out on the wind speed, and the prediction precision is improved. Meanwhile, ultra-short-term prediction is carried out on the power of the wind turbine generator by utilizing an extreme learning machine model (ELM) and considering unit operation index data, the comprehensiveness of wind power prediction is improved, and prediction errors caused by unit state abnormity are reduced.
As shown in fig. 2, another embodiment of the present invention provides an ultra-short term prediction apparatus for wind power, where the key factor calculation module is configured to calculate a first degree of association between a weather factor and wind power and a second degree of association between the operation index factor and the wind power according to historical weather data of a wind farm, operation index data of wind turbine generators, and wind power data, and determine a key factor affecting the wind power according to the first degree of association and the second degree of association; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data;
the data screening module is used for screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data;
the wind power prediction module is used for inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and the wind power ultra-short term prediction model outputs the predicted wind power at the current moment; the wind power ultra-short term prediction model is an extreme learning machine model.
As one embodiment, the wind speed correction system further comprises a wind speed correction module, wherein the wind speed correction module is used for inputting first weather prediction data of a wind power plant at the current moment and a weather prediction error of the previous moment into the wind speed correction model, correcting the wind speed in the first weather prediction data according to the prediction error of the previous moment by the wind speed correction model, and outputting second weather prediction data of the corrected current moment; the wind speed correction model is an extreme learning machine model.
On the basis of the embodiment of the invention, the invention correspondingly provides an embodiment of a readable storage medium;
another embodiment of the present invention provides a readable storage medium, which includes a stored computer program, and when the computer program is executed, the readable storage medium controls a device on which the readable storage medium is located to perform the method for ultra-short term prediction of wind power according to any one of the method embodiments of the present invention.
Illustratively, the computer program may be partitioned into one or more modules that are stored in the memory and executed by the processor to implement the invention. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the terminal device.
The terminal device can be a desktop computer, a notebook, a palm computer, a cloud server and other computing devices. The terminal device may include, but is not limited to, a processor, a memory.
The Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may be any conventional processor or the like, which is the control center of the terminal device and connects the various parts of the whole terminal device using various interfaces and lines.
The memory may be used for storing the computer programs and/or modules, and the processor may implement various functions of the terminal device by executing or executing the computer programs and/or modules stored in the memory and calling data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the cellular phone, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
Wherein, the terminal device integrated module/unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, can be stored in a computer readable storage medium (i.e. the above readable storage medium). Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like.
It should be noted that the above-described device embodiments are merely illustrative, where the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. In addition, in the drawings of the embodiment of the apparatus provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, and may be specifically implemented as one or more communication buses or signal lines.
One of ordinary skill in the art can understand and implement it without inventive effort. While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention. It will be understood by those skilled in the art that all or part of the processes of the above embodiments may be implemented by hardware related to instructions of a computer program, and the computer program may be stored in a computer readable storage medium, and when executed, may include the processes of the above embodiments. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
Claims (10)
1. An ultra-short-term prediction method for wind power is characterized by comprising the following steps:
calculating a first association degree between a weather factor and wind power and a second association degree between an operation index factor and the wind power according to historical weather data of a wind power plant, operation index data of a wind turbine generator and the wind power data, and determining key factors influencing the wind power according to the first association degree and the second association degree; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data;
screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data;
inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and outputting the predicted wind power at the current moment by the wind power ultra-short term prediction model; the wind power ultra-short term prediction model is an extreme learning machine model.
2. The ultra-short-term wind power prediction method according to claim 1, wherein a first weather prediction data of a wind farm at a current time and a weather prediction error at a previous time are input to a wind speed correction model, the wind speed correction model corrects a wind speed in the first weather prediction data according to the prediction error at the previous time, and outputs a second weather prediction data at the current time after correction; the wind speed correction model is an extreme learning machine model.
3. The ultra-short term prediction method for wind power according to claim 2, wherein a first degree of correlation between weather factors and wind power and a second degree of correlation between operation index factors and wind power are calculated according to historical weather data of a wind farm, operation index data of a wind turbine generator and wind power data, and key factors affecting the wind power are determined according to the first degree of correlation and the second degree of correlation, specifically:
inputting historical weather data of a wind power plant, operation index data of the wind generation set and wind power data of the wind generation set into a wind power association degree model, calculating a first association degree between a weather factor and wind power and a second association degree between the operation index factor and the wind power according to the historical weather data of the wind power plant, the operation index data of the wind generation set and the wind power data by the wind power association degree model, and determining key factors influencing the wind power according to the first association degree and the second association degree, wherein the key factors comprise the weather key factors and the operation index key factors; the wind power correlation degree model is a grey correlation analysis model.
4. The ultra-short-term prediction method of wind power according to claim 3, characterized in that the first weather prediction data is obtained after screening the weather prediction data of the wind farm at the current moment according to the key factors.
5. The ultra-short term prediction method of wind power as claimed in claim 4, wherein the wind power correlation model respectively normalizes the historical weather data of the wind farm, the operation index data of the wind turbine generator and the wind power data before calculating the first correlation between the weather factor and the wind power and the second correlation between the operation index factor and the wind power according to the historical weather data of the wind farm, the operation index data of the wind turbine generator and the wind power data.
6. The ultra-short term prediction method of wind power according to claim 5, wherein the weather factors of the first degree of correlation greater than 0.85 are determined as weather key factors, and the operation index factors of the second degree of correlation greater than 0.85 are determined as operation index key factors.
7. The ultra-short term prediction method of wind power according to any one of claims 1 to 6, characterized in that the historical weather data input to the wind power relevancy model is screened according to the weather key factors to obtain screened historical weather data, the operation index data input to the wind power relevancy model is screened according to the operation index key factors to obtain screened operation index data, and the screened historical weather data and the screened operation index data are used as a training set of the wind power ultra-short term prediction model.
8. An ultra-short-term prediction device for wind power is characterized by comprising a key factor calculation module, a data screening module and a wind power prediction module;
the key factor calculation module is used for calculating a first correlation degree between the weather factors and the wind power and a second correlation degree between the operation index factors and the wind power according to historical weather data of the wind power plant, operation index data of the wind generation set and the wind power data, and determining key factors influencing the wind power according to the first correlation degree and the second correlation degree; the operation index data is the operation index data of the wind turbine generator corresponding to the historical weather data, and the wind power data is the wind power data of the wind turbine generator corresponding to the historical weather data;
the data screening module is used for screening second weather forecast data at the current moment according to the key factors to obtain third weather forecast data, and screening second operation index data at the previous moment according to the key factors to obtain third operation index data; the second operation index data is the operation index data of the wind turbine generator corresponding to the second weather prediction data;
the wind power prediction module is used for inputting the third weather prediction data and the third operation index data into a wind power ultra-short term prediction model, and the wind power ultra-short term prediction model outputs the predicted wind power at the current moment; the wind power ultra-short term prediction model is an extreme learning machine model.
9. The ultra-short-term wind power prediction device according to claim 8, comprising a wind speed correction module, wherein the wind speed correction module is configured to input a first weather prediction data of a wind farm at a current time and a weather prediction error of a previous time into a wind speed correction model, and the wind speed correction model corrects a wind speed in the first weather prediction data according to the prediction error of the previous time and outputs a second weather prediction data of the current time after correction; the wind speed correction model is an extreme learning machine model.
10. A readable storage medium, characterized in that the readable storage medium comprises a stored computer program, which when executed controls a device on which the readable storage medium is located to perform the ultra-short term prediction method of wind power as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111543759.0A CN114418180B (en) | 2021-12-16 | 2021-12-16 | Ultra-short-term prediction method and device for wind power and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111543759.0A CN114418180B (en) | 2021-12-16 | 2021-12-16 | Ultra-short-term prediction method and device for wind power and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114418180A true CN114418180A (en) | 2022-04-29 |
| CN114418180B CN114418180B (en) | 2023-01-20 |
Family
ID=81267497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111543759.0A Active CN114418180B (en) | 2021-12-16 | 2021-12-16 | Ultra-short-term prediction method and device for wind power and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114418180B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115936924A (en) * | 2022-12-14 | 2023-04-07 | 广西电网有限责任公司 | Wind power plant wind energy prediction method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108229732A (en) * | 2017-12-20 | 2018-06-29 | 上海电机学院 | ExtremeLearningMachine wind speed ultra-short term prediction method based on error correction |
| CN111461444A (en) * | 2020-04-07 | 2020-07-28 | 上海电气风电集团股份有限公司 | Prediction method, system, medium and electronic device for unit power of wind power plant |
| CN113326969A (en) * | 2021-04-29 | 2021-08-31 | 淮阴工学院 | Short-term wind speed prediction method and system based on improved whale algorithm for optimizing ELM |
| CN113642784A (en) * | 2021-07-27 | 2021-11-12 | 西安理工大学 | Wind power ultra-short term prediction method considering fan state |
| CN113657662A (en) * | 2021-08-13 | 2021-11-16 | 浙江大学 | Downscaling wind power prediction method based on data fusion |
-
2021
- 2021-12-16 CN CN202111543759.0A patent/CN114418180B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108229732A (en) * | 2017-12-20 | 2018-06-29 | 上海电机学院 | ExtremeLearningMachine wind speed ultra-short term prediction method based on error correction |
| CN111461444A (en) * | 2020-04-07 | 2020-07-28 | 上海电气风电集团股份有限公司 | Prediction method, system, medium and electronic device for unit power of wind power plant |
| CN113326969A (en) * | 2021-04-29 | 2021-08-31 | 淮阴工学院 | Short-term wind speed prediction method and system based on improved whale algorithm for optimizing ELM |
| CN113642784A (en) * | 2021-07-27 | 2021-11-12 | 西安理工大学 | Wind power ultra-short term prediction method considering fan state |
| CN113657662A (en) * | 2021-08-13 | 2021-11-16 | 浙江大学 | Downscaling wind power prediction method based on data fusion |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115936924A (en) * | 2022-12-14 | 2023-04-07 | 广西电网有限责任公司 | Wind power plant wind energy prediction method and system |
| CN115936924B (en) * | 2022-12-14 | 2023-08-25 | 广西电网有限责任公司 | Wind energy prediction method and system for wind power plant |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114418180B (en) | 2023-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Smart wind speed forecasting approach using various boosting algorithms, big multi-step forecasting strategy | |
| He et al. | A combined model for short-term wind power forecasting based on the analysis of numerical weather prediction data | |
| Yao et al. | Short-term load forecasting method based on feature preference strategy and LightGBM-XGboost | |
| CN111144644B (en) | Short-term wind speed prediction method based on variation variance Gaussian process regression | |
| Zhang et al. | Wind speed prediction research considering wind speed ramp and residual distribution | |
| CN112508299A (en) | Power load prediction method and device, terminal equipment and storage medium | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN115034430A (en) | Carbon emission prediction method, device, terminal and storage medium | |
| CN114418180B (en) | Ultra-short-term prediction method and device for wind power and storage medium | |
| CN116799796A (en) | Photovoltaic power generation power prediction method, device, equipment and medium | |
| CN115016966A (en) | Measurement automation system fault prediction method and device based on Transformer and storage medium | |
| Sang et al. | Ensembles of gradient boosting recurrent neural network for time series data prediction | |
| CN112819246A (en) | Energy demand prediction method for optimizing neural network based on cuckoo algorithm | |
| CN116595895A (en) | Training method of short-time electric quantity prediction model and short-time electric quantity prediction method | |
| CN115545164A (en) | Photovoltaic power generation power prediction method, system, equipment and medium | |
| CN115146822A (en) | Photovoltaic power generation prediction method and device and terminal equipment | |
| CN115619449A (en) | Deep learning model-based fertilizer price prediction method, device and storage medium | |
| CN114547552A (en) | Method and device for generating analog data, intelligent terminal and storage medium | |
| CN112906757A (en) | Model training method, optical power prediction method, device, equipment and storage medium | |
| Zhang et al. | A Hybrid Neural Network-Based Intelligent Forecasting Approach for Capacity of Photovoltaic Electricity Generation | |
| Yang et al. | LRG-Net: Lightweight Residual Grid Network for Modeling Electrical Induction Motor Dynamics | |
| CN114154679B (en) | Spark-based PCFOA-KELM wind power prediction method and device | |
| CN117688367B (en) | Wind power generation ultra-short term power prediction method and device based on instant learning | |
| Jiang | A combined monthly precipitation prediction method based on CEEMD and improved LSTM | |
| CN111461418B (en) | Wind speed prediction method, system, electronic equipment and storage medium |
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 |