CN112949201B - Wind speed prediction method and device, electronic equipment and storage medium - Google Patents
Wind speed prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112949201B CN112949201B CN202110287049.XA CN202110287049A CN112949201B CN 112949201 B CN112949201 B CN 112949201B CN 202110287049 A CN202110287049 A CN 202110287049A CN 112949201 B CN112949201 B CN 112949201B
- Authority
- CN
- China
- Prior art keywords
- wind speed
- prediction
- interpretable
- data set
- error
- 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 67
- 238000012549 training Methods 0.000 claims abstract description 37
- 230000006870 function Effects 0.000 claims description 78
- 238000005315 distribution function Methods 0.000 claims description 30
- 230000001186 cumulative effect Effects 0.000 claims description 29
- 238000004364 calculation method Methods 0.000 claims description 24
- 238000012360 testing method Methods 0.000 claims description 17
- 238000012795 verification Methods 0.000 claims description 15
- 238000007619 statistical method Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 238000010200 validation analysis Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 10
- 108090000672 Annexin A5 Proteins 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/06—Wind turbines or wind farms
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a wind speed prediction method, a wind speed prediction device, electronic equipment and a storage medium, wherein the method comprises the following steps: acquiring a wind speed data set of a wind power plant, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector; extracting M guide pairs from the wind speed data set; respectively training M interpretable networks according to the M guide pairs to obtain interpretable network models, wherein one guide pair is used for training one interpretable network which comprises interpretable parameters influencing wind speed, and the interpretable network models comprise M trained interpretable networks; and predicting the wind speed interval of the wind power plant through the interpretable network model. The method and the device can predict the wind speed interval and can realize the explanation of the final prediction result. The characteristics of the wind speed data can be theoretically extracted, and the relationship between the characteristic data and the prediction intervals with different nominal confidence degrees can be clearly explained.
Description
Technical Field
The present invention relates to the field of power information processing, and in particular, to a wind speed prediction method, apparatus, electronic device, and storage medium.
Background
With the popularization of wind power in modern power grids, accurate and reliable wind speed prediction plays an increasingly important role in the planning and operation of power systems. The high-precision wind speed prediction has important significance for saving power generation cost, spare capacity and maintenance plan of a power system, and can furthest exert the advantages of wind energy in modern renewable energy sources. In fact, the chaos characteristic of the weather system brings uncertainty to wind power generation, and greatly hinders the planning, management and operation of the electric energy system. The traditional point prediction is difficult to accurately predict, and the probability and fluctuation range of the predicted value cannot be obtained.
The wind power prediction method of the probability interval is mainly divided into three categories, including a physical modeling method, a statistical model and a hybrid artificial intelligence method. The core of the physical modeling method is to establish an accurate mathematical model by using various meteorological collections so as to obtain forecast distribution and estimate the uncertainty of forecast. Statistical methods attempt to establish a relationship between future winds and historical samples by minimizing errors. And researching the uncertainty of wind power prediction according to the statistical analysis of the wind speed prediction and the error nonlinear power curve. Compared with the first two common methods, artificial intelligence has been developed vigorously due to its potential capabilities in data mining and feature extraction. However, because the artificial neural networks have the black box characteristic, the networks are not interpretable, and accurate prediction result analysis cannot be provided for a data analyst of a factory.
Disclosure of Invention
The embodiment of the invention provides a wind speed prediction method which can predict a wind speed interval and can explain a final prediction result. The characteristics of the wind speed data can be theoretically extracted, and the relationship between the characteristic data and the prediction intervals with different nominal confidence degrees can be clearly explained.
In a first aspect, an embodiment of the present invention provides a wind speed prediction method, which is used for wind speed interval prediction of a wind farm, and includes:
acquiring a wind speed data set of a wind power plant, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector;
extracting M guide pairs from the wind speed data set;
respectively training M interpretable networks (xNN) according to the M guide pairs to obtain interpretable Network models, wherein one guide pair is used for training one interpretable Network which comprises interpretable parameters influencing wind speed, and the interpretable Network models comprise M trained interpretable networks;
and predicting the wind speed interval of the wind power plant through the interpretable network model.
Optionally, the wind speed data set includes a training set, and the step of extracting M guide pairs from the wind speed data set includes:
m guide pairs are extracted from the training set by a Bootstrap method.
Optionally, the wind speed dataset further comprises a validation set, the validation set comprising real values, the method further comprising:
inputting the verification set into the interpretable network model for calculation to obtain a first prediction result;
comparing the first prediction result with the real value to obtain a prediction error data set;
an error probability density function and an error cumulative distribution function are calculated from the prediction error data set.
Optionally, the step of calculating the error probability density function and the error cumulative distribution function according to the prediction error data set specifically includes:
and (3) performing statistical analysis and calculation on the prediction error data set by a nonparametric Kernel Density Estimation (KDE), so as to obtain an error probability density function and an error cumulative distribution function.
Optionally, the step of performing statistical analysis and calculation on the prediction error data set by using a nonparametric kernel density estimation method to obtain an error probability density function includes:
determining a kernel function;
constructing an initial probability density function according to the kernel function;
calculating the mean squared integral variance of the initial probability density function;
calculating an asymptotic mean square integral variance of the initial probability density function;
calculating the optimal bandwidth of the initial probability density function;
and obtaining an error probability density function based on the average integral square difference of the initial probability density function, the asymptotic average integral square difference of the initial probability density function and the optimal bandwidth of the initial probability density function.
Optionally, the wind speed data set further includes a test set, and the method further includes:
inputting the test set into the interpretable network model for calculation to obtain a second prediction result;
and evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of probability interval prediction.
Optionally, the step of evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of the probability interval prediction specifically includes:
evaluating the second prediction result by the error cumulative distribution function;
the upper and lower bounds of the probability interval prediction are calculated with a confidence level of 100 (1- α)% where α is greater than or equal to 0 and α is less than or equal to 1.
In a second aspect, an embodiment of the present invention further provides a wind speed prediction apparatus, configured to predict a wind speed interval of a wind farm, where the apparatus includes:
the wind speed data set comprises a wind speed scalar and a wind speed vector;
an extraction module for extracting M guide pairs from the wind speed data set;
a training module, configured to train M interpretable networks respectively according to the M guide pairs to obtain an interpretable network model, where one guide pair is used to train one interpretable network, the interpretable network includes interpretable parameters affecting wind speed, and the interpretable network model includes M trained interpretable networks;
and the prediction module is used for predicting the wind speed interval of the wind power plant through the interpretable network model.
In a third aspect, an embodiment of the present invention provides an electronic device, including: the wind speed prediction method comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the wind speed prediction method provided by the embodiment of the invention.
In the embodiment of the invention, a wind speed data set of a wind power plant is obtained, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector; extracting M guide pairs from the wind speed data set; respectively training M interpretable networks according to the M guide pairs to obtain interpretable network models, wherein one guide pair is used for training one interpretable network which comprises interpretable parameters influencing wind speed, and the interpretable network models comprise M trained interpretable networks; and predicting the wind speed interval of the wind power plant through the interpretable network model. The method and the device can predict the wind speed interval and can realize the explanation of the final prediction result. The characteristics of the wind speed data can be theoretically extracted, and the relationship between the characteristic data and the prediction intervals with different nominal confidence degrees can be clearly explained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of a wind speed prediction method provided by an embodiment of the invention;
FIG. 2 is a block diagram of an example of a framework for interpreting a network model;
FIG. 3 is a diagram illustrating a prediction interval according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating another prediction interval provided by an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of a wind speed prediction device according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the 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.
Referring to fig. 1, fig. 1 is a flowchart of a wind speed prediction method provided by an embodiment of the present invention, the method is used for wind speed prediction of a wind farm, as shown in fig. 1, the method includes the following steps:
s1, acquiring a wind speed data set of a wind power plant.
In an embodiment of the invention, the wind speed data set comprises a wind speed scalar and a wind speed vector. The wind speed data set described above may be written as
Wherein t is i Is the wind speed at time i (i.e., the wind speed scalar in the embodiment of the present invention), X i Is an input vector (i.e., a wind speed vector in an embodiment of the present invention) including a historical wind speed, a wind direction, radiation, and the like. Specifically, the wind power plant raw data is collected to form a raw data set, and the raw data set is processed to obtain the wind speed data set.
And S2, extracting M guide pairs from the wind speed data set.
In the embodiment of the present invention, the above-mentioned bootstrap pair may be understood as a data set, and the data of one bootstrap pair may be utilized to train the network.
Further, the wind speed data set includes a training set, and M guide pairs may be extracted from the training set by a boottrap method. In one possible embodiment, the training set may be a part of the wind speed data set, for example, the sample size of the training set is 20% of the sample size of the wind speed data set, and the sample size of the wind speed data set may also be referred to as the total sample size. Specifically, the guide pairs can be generated by replacing training data in the wind speed data set (or training set) by uniform sampling
Estimating the ith leadThe output of the data set at the ith interpretable network. Using i = i +1 as a cycle when i>And when M is needed, the loop is terminated, and M guide pairs are obtained.
Further, the training set includes sample data and label data, wherein one sample in the sample data corresponds to one label in the label data. The tag data described above may also be referred to as a true value.
And S3, respectively training M interpretable networks according to the M guide pairs to obtain interpretable network models.
In an embodiment of the invention, a bootstrap pair is used for training an interpretable network, the interpretable network comprises interpretable parameters influencing wind speed, and the interpretable network model comprises M trained interpretable networks.
During the training process, the interpretable network may be trained using a training set. M Bootstrap pairs were generated using the Bootstrap method. An interpretable network is trained using data from a bootstrap pair to obtain optimal network structure and parameters. And finally obtaining M trained interpretable networks. M interpretable network models with optimal parameters and structure are stored.
Referring to fig. 2, fig. 2 is a block diagram of an interpretable network model according to an embodiment of the present invention, and as can be seen from fig. 2, a whole block of the interpretable network model includes M interpretable networks, which can be trained by M pilots, respectively, and a part of the M interpretable networks, which can be called B-xNNs, where B is a pilot pair, xNN is an interpretable network, and s represents a plurality of neural networks, and the part is trained by M pilots, respectively. It should be noted that the above-mentioned framework capable of interpreting the network model may also be referred to as a prediction framework. The interpretable network may be a neural network, and the interpretable network includes an input layer, a hidden layer, and an output layer.
Furthermore, each interpretable network can be trained by using a gradient descent method, when the error loss of one interpretable network is less than a set threshold value, the interpretable network can be determined to be trained, and model parameters and structures of the trained interpretable network are stored. The hidden layer in the interpretable network includes a first hidden layer and a second hidden layer, and specifically, the input and output relationship of each interpretable network can be represented by the following expression:
in the above expression, i is the number of subnets in the interpretable network, K is the number of neurons in the first hidden layer, β is the input weight, γ is the weight of the second hidden layer, and μ is the deviation of the output layer. The overall output expression of the interpretable network (B-xNNs) is as follows:
in the above expression, M is the number of interpretable networks, f i Is the output of the ith interpretable network.
Further, the wind speed data set further comprises a verification set, the verification set comprises real values, and the verification set can be input into the interpretable network model for calculation to obtain a first prediction result; comparing the first prediction result with the real value to obtain a prediction error data set; an error probability density function and an error cumulative distribution function are calculated from the prediction error data set. In one possible embodiment, the verification set may be a part of the wind speed data set, for example, the sample size of the verification set is 70% of the sample size of the wind speed data set, and the sample size of the wind speed data set may also be referred to as the total sample size.
Specifically, the prediction error data set may be statistically analyzed and calculated by a nonparametric kernel density estimation method to obtain an error probability density function and an error cumulative distribution function. More specifically, a kernel function may be determined first; constructing an initial probability density function according to the kernel function; calculating the mean integral square error of the initial probability density function; calculating the asymptotic mean integral-square difference of the initial probability density function; calculating the optimal bandwidth of the initial probability density function; and obtaining an error probability density function based on the average integral-square difference of the initial probability density function, the asymptotic average integral-square difference of the initial probability density function and the optimal bandwidth of the initial probability density function. The expression of the probability density function adopted by the embodiment of the invention is as follows:
in the above expression, s i For a sample point, h is the bandwidth and K is the kernel function.
In an embodiment of the present invention, the expression of the kernel function is determined as follows:
combining the probability density function and the kernel function to obtain an initial probability density function expression as follows:
the mean-squared error expression can be found as follows:
the expression of asymptotic mean integral-squared error can be found as follows:
wherein, in the above expression, E IQR Is a quartile range. The optimal broadband is as follows:
in the embodiment of the present invention, the error cumulative distribution function may be understood as an integral formula of the error function, and therefore, the wind speed data set further includes a test set, and the test set may be input into the interpretable network model for calculation to obtain a second prediction result; and evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of probability interval prediction. More specifically, the second prediction result may be evaluated by the error cumulative distribution function; the upper and lower bounds of the probability interval prediction are calculated with a confidence level of 100 (1- α)% where α is greater than or equal to 0 and α is less than or equal to 1. In one possible embodiment, the test set may be a part of the wind speed data set, for example, the sample size of the test set is 10% of the sample size of the wind speed data set, and the sample size of the wind speed data set may also be referred to as the total sample size. In the embodiment of the invention, the wind speed data set can be divided into a test set, a training set and a verification set according to the data volume proportion of 7. The actual values are included in the test set.
Specifically, the upper and lower bounds of the probability interval prediction may be output using a test set. Inputting the test set into the trained interpretable network model, and evaluating the prediction error by using an error accumulation distribution function. Taking the case of a confidence level of 100 (1- α)% as an example, the upper and lower boundaries of the output probability interval for a confidence level of 100 (1- α)% are obtained from the corresponding error cumulative distribution function. Specifically, the following expression is shown:
in the above expression, C1, C2, C3, and μ are constant parameters. P1, P2 and P3 are interpretable parameters and are influenced by different interpretation indexes.
The probability section is a probability section corresponding to the wind speed section.
And S4, predicting the wind speed interval of the wind power plant through the interpretable network model.
In the embodiment of the invention, the historical grid speed data of the wind power plant is obtained, and is input into the interpretable network model to predict the wind speed interval in a certain time in the future.
In the embodiment of the invention, a wind speed data set of a wind power plant is obtained, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector; extracting M guide pairs from the wind speed data set; respectively training M interpretable networks according to the M guide pairs to obtain an interpretable network model, wherein one guide pair is used for training one interpretable network which comprises interpretable parameters influencing wind speed, and the interpretable network model comprises M trained interpretable networks; and predicting the wind speed interval of the wind power plant through the interpretable network model. The method and the device can predict the wind speed interval and can realize the explanation of the final prediction result. The characteristics of the wind speed data can be theoretically extracted, and the relationship between the characteristic data and the prediction intervals with different nominal confidence degrees can be clearly explained.
Optionally, in the embodiment of the present invention, in order to evaluate the performance of the probability interval, indexes of an Average Coverage Error (ACE) and an interval score (IS, also called interval sharpness, and the quality of the prediction interval IS evaluated by rewarding a narrower prediction interval and penalizing a wider prediction interval) are introduced. Specifically, the following expression is shown:
in the above equation, PICP (PI coverage, abbreviated as PICP) is a probability interval coverage, and the PINC is a set nominal confidence (a name set for the user). Average overlay error
In the embodiment of the present invention, the prediction is performed by taking the season of different seasons as an example, and the parameters P1, P2, and P3 are parameters related to seasons and are affected by different seasons. In order to fully evaluate the effectiveness of the wind speed prediction method provided by the embodiment of the invention, the wind speed prediction method provided by the embodiment of the invention was tested by using data of a transmission company named Elia in belgium. The wind speed data set is divided into 7:2:1, dividing a verification set, a training set and a test set. The interpretable network is trained using a training set. M Bootstrap pairs were generated using the Bootstrap method. An interpretable network is trained using data from a bootstrap pair to obtain optimal network structure and parameters. And finally obtaining M trained interpretable networks. M models with optimal parameters and structures are saved as interpretable network models. The obtained input and output mathematical expression is as follows:
taking the autumn wind speed prediction as an example, the following parameter sizes are parameters of one of the networks used for the autumn prediction. In the above expression, C 1 ＝0.6894438,C 2 ＝0.13415974,C 3 =0.05075899, μ = 0.54264. The P1, P2, P3 parameters are shown in table 1 below:
TABLE 1
Table 1 is a parameter of the interpretable network. The predicted result can be expressed by a mathematical formula according to table 1, and the whole predicted result can be explained. The wind speed prediction method provided by the embodiment of the invention can learn interpretable characteristics of wind speed, and express the influence of input variables on results through clear formulas, thereby providing an effective tool for operation and planning of a power system.
Referring to fig. 3 and 4, fig. 3 is a schematic diagram of a prediction interval provided by an embodiment of the present invention, and fig. 4 is a schematic diagram of another prediction interval provided by an embodiment of the present invention, as shown in fig. 3 and 4, the prediction interval has a smaller width and a larger interval coverage rate. As the nominal confidence level increases, more points actually fall within the interval, and the width of the prediction interval increases, but not by a large amount. It can be seen that the wind speed prediction method provided by the embodiment of the invention achieves or even exceeds the interval confidence level in the interval coverage (PICP) index of the prediction interval. Specific values may be as shown in table 2 below:
TABLE 2
As can be seen from the table 2, the obtained wind speed is always close to the corresponding nominal confidence level, the interval score is also smaller, the wind speed prediction method provided by the embodiment of the invention has high stability and robustness, the prediction result can effectively grasp uncertain information in wind power, and reference can be provided for a power system decision maker.
In the embodiment of the invention, the kernel density estimation avoids inaccurate prediction caused by unreasonable assumed error distribution under the condition that the wind power error distribution is not assumed in advance. And according to the prediction result of the B-xNNs, counting the uncertainty of the prediction to obtain a probability prediction result with high reliability. The embodiment of the invention provides a wind speed prediction method (a new interpretable wind speed prediction framework), which can learn interpretable characteristics of wind speed, express the influence of input variables on results through clear formulas and provide an effective tool for operation and planning of a power system.
Referring to fig. 5, fig. 5 is a schematic structural diagram of a wind speed prediction apparatus according to an embodiment of the present invention, and as shown in fig. 5, the apparatus is used for wind speed interval prediction of a wind farm, and the apparatus includes:
an obtaining module 501, configured to obtain a wind speed dataset of a wind farm, where the wind speed dataset includes a wind speed scalar and a wind speed vector;
an extraction module 502, configured to extract M guide pairs from the wind speed data set;
a training module 503, configured to train M interpretable networks respectively according to the M guide pairs to obtain an interpretable network model, where one guide pair is used to train one interpretable network, the interpretable network includes interpretable parameters affecting wind speed, and the interpretable network model includes M trained interpretable networks;
and the prediction module 504 is used for predicting the wind speed interval of the wind power plant through the interpretable network model.
Optionally, the wind speed data set includes a training set, and the step of extracting M guide pairs from the wind speed data set includes:
m guide pairs are extracted from the training set by a Bootstrap method.
Optionally, the wind speed dataset further includes a validation set, the validation set includes real values, and the apparatus further includes:
the first calculation module is used for inputting the verification set into the interpretable network model for calculation to obtain a first prediction result;
the comparison module is used for comparing the first prediction result with the real value to obtain a prediction error data set;
and the second calculation module is used for calculating an error probability density function and an error cumulative distribution function according to the prediction error data set.
Optionally, the step of calculating the error probability density function and the error cumulative distribution function according to the prediction error data set specifically includes:
and carrying out statistical analysis calculation on the prediction error data set by a nonparametric kernel density estimation method to obtain an error probability density function and an error cumulative distribution function.
Optionally, the step of performing statistical analysis and calculation on the prediction error data set by using a non-parametric kernel density estimation method to obtain an error probability density function specifically includes:
determining a kernel function;
constructing an initial probability density function according to the kernel function;
calculating the mean squared integral variance of the initial probability density function;
calculating an asymptotic mean squared integral variance of the initial probability density function;
calculating the optimal bandwidth of the initial probability density function;
and obtaining an error probability density function based on the average integral square difference of the initial probability density function, the asymptotic average integral square difference of the initial probability density function and the optimal bandwidth of the initial probability density function.
Optionally, the wind speed data set further includes a test set, and the apparatus further includes:
the third calculation module is used for inputting the test set into the interpretable network model for calculation to obtain a second prediction result;
and the fourth calculation module is used for evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of probability interval prediction.
Optionally, the step of evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of the probability interval prediction specifically includes:
evaluating the second prediction result by the error cumulative distribution function;
the upper and lower bounds of the probability interval prediction are calculated with a confidence level of 100 (1- α)% where α is greater than or equal to 0 and α is less than or equal to 1.
The wind speed prediction device provided by the embodiment of the invention can be applied to devices such as mobile phones, monitors, computers, servers and the like which can perform wind speed prediction.
The wind speed prediction device provided by the embodiment of the invention can realize each process realized by the wind speed prediction method in the method embodiment, and can achieve the same beneficial effect. To avoid repetition, further description is omitted here.
Referring to fig. 6, fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 6, including: a memory 602, a processor 601, and a computer program stored on the memory 602 and executable on the processor 601, wherein:
the processor 601 is configured to call the computer program stored in the memory 602 for wind speed interval prediction of the wind farm, and the processor 601 performs the following steps:
acquiring a wind speed data set of a wind power plant, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector;
extracting M guide pairs from the wind speed data set;
respectively training M interpretable networks according to the M guide pairs to obtain interpretable network models, wherein one guide pair is used for training one interpretable network which comprises interpretable parameters influencing wind speed, and the interpretable network models comprise M trained interpretable networks;
and predicting the wind speed interval of the wind power plant through the interpretable network model.
Optionally, the wind speed data set includes a training set, and the step of extracting M guide pairs from the wind speed data set by the processor 601 specifically includes:
m guide pairs are extracted from the training set by a Bootstrap method.
Optionally, the wind speed dataset further comprises a verification set, the verification set comprises real values, and the method executed by the processor 601 further comprises:
inputting the verification set into the interpretable network model for calculation to obtain a first prediction result;
comparing the first prediction result with the real value to obtain a prediction error data set;
an error probability density function and an error cumulative distribution function are calculated from the prediction error data set.
Optionally, the specific steps executed by the processor 601 to calculate the error probability density function and the error cumulative distribution function according to the prediction error data set include:
and carrying out statistical analysis calculation on the prediction error data set by a nonparametric kernel density estimation method to obtain an error probability density function and an error cumulative distribution function.
Optionally, the specific step of the processor 601, executed to perform statistical analysis and calculation on the prediction error data set by using a non-parametric kernel density estimation method, to obtain an error probability density function includes:
determining a kernel function;
constructing an initial probability density function according to the kernel function;
calculating the mean squared integral variance of the initial probability density function;
calculating an asymptotic mean squared integral variance of the initial probability density function;
calculating the optimal bandwidth of the initial probability density function;
and obtaining an error probability density function based on the average integral square difference of the initial probability density function, the asymptotic average integral square difference of the initial probability density function and the optimal bandwidth of the initial probability density function.
Optionally, the wind speed data set further includes a test set, and the method executed by the processor 601 further includes:
inputting the test set into the interpretable network model for calculation to obtain a second prediction result;
and evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of probability interval prediction.
Optionally, the step, executed by the processor 601, of evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of the probability interval prediction specifically includes:
evaluating the second prediction result by the error cumulative distribution function;
the upper and lower bounds of the probability interval prediction are calculated with a confidence level of 100 (1- α)% where α is greater than or equal to 0 and α is less than or equal to 1.
The electronic device may be a device that can be applied to a mobile phone, a monitor, a computer, a server, or the like that can predict wind speed.
The electronic device provided by the embodiment of the invention can realize each process realized by the wind speed prediction method in the method embodiment, can achieve the same beneficial effects, and is not repeated here for avoiding repetition.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above disclosure is only for the purpose of illustrating the preferred embodiments of the present invention, and it is therefore to be understood that the invention is not limited by the scope of the appended claims.
Claims (8)
1. A wind speed prediction method is used for wind speed interval prediction of a wind power plant and is characterized by comprising the following steps:
acquiring a wind speed data set of a wind power plant, wherein the wind speed data set comprises a wind speed scalar and a wind speed vector; the wind speed dataset further comprises a validation set comprising real values;
extracting M guide pairs from the wind speed data set;
respectively training M interpretable networks according to the M guide pairs to obtain interpretable network models, wherein one guide pair is used for training one interpretable network which comprises interpretable parameters influencing wind speed, and the interpretable network models comprise M trained interpretable networks;
predicting a wind speed interval of the wind power plant through the interpretable network model; wherein the method further comprises:
inputting the verification set into the interpretable network model for calculation to obtain a first prediction result;
comparing the first prediction result with the real value to obtain a prediction error data set;
performing statistical analysis calculation on the prediction error data set by a nonparametric kernel density estimation method to obtain an error probability density function and an error cumulative distribution function;
the expression of the probability density function is as follows:
in the above expression, s i For a sample point, h is the bandwidth and K is the kernel function.
2. The wind speed prediction method according to claim 1, wherein the wind speed dataset comprises a training set, and the step of extracting M guide pairs from the wind speed dataset comprises:
m guide pairs are extracted from the training set by a Bootstrap method.
3. The method for wind speed prediction according to claim 1, wherein the step of obtaining the error probability density function by performing statistical analysis calculation on the prediction error data set through a non-parametric kernel density estimation method comprises:
determining a kernel function;
constructing an initial probability density function according to the kernel function;
calculating the mean squared integral variance of the initial probability density function;
calculating an asymptotic mean squared integral variance of the initial probability density function;
calculating the optimal bandwidth of the initial probability density function;
and obtaining an error probability density function based on the average integral square difference of the initial probability density function, the asymptotic average integral square difference of the initial probability density function and the optimal bandwidth of the initial probability density function.
4. The wind speed prediction method of claim 3, wherein the wind speed data set further comprises a test set, the method further comprising:
inputting the test set into the interpretable network model for calculation to obtain a second prediction result;
and evaluating the second prediction result through the error cumulative distribution function to obtain an upper boundary and a lower boundary of probability interval prediction.
5. The method according to claim 4, wherein the step of evaluating the second prediction result by the cumulative error distribution function to obtain the upper boundary and the lower boundary of the probability interval prediction specifically comprises:
evaluating the second prediction result by the error cumulative distribution function;
the upper and lower bounds of the probability interval prediction are calculated with a confidence level of 100 (1- α)% where α is greater than or equal to 0 and α is less than or equal to 1.
6. A wind speed prediction device for wind speed interval prediction of a wind farm, the device comprising:
the wind speed data set comprises a wind speed scalar and a wind speed vector; the wind speed dataset further comprises a validation set comprising real values;
an extraction module for extracting M guide pairs from the wind speed data set;
a training module, configured to train M interpretable networks respectively according to the M guide pairs to obtain an interpretable network model, where one guide pair is used to train one interpretable network, the interpretable network includes interpretable parameters affecting wind speed, and the interpretable network model includes M trained interpretable networks; the verification set is also used for inputting the verification set into the interpretable network model for calculation to obtain a first prediction result; comparing the first prediction result with the real value to obtain a prediction error data set; performing statistical analysis calculation on the prediction error data set by a nonparametric kernel density estimation method to obtain an error probability density function and an error cumulative distribution function; the expression of the probability density function is as follows:
in the above expression, s i Is a sample point, h is the bandwidth, and K is the kernel function;
and the prediction module is used for predicting the wind speed interval of the wind power plant through the interpretable network model.
7. An electronic device, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, the processor when executing the computer program implementing the steps in the wind speed prediction method according to any of claims 1 to 5.
8. A computer-readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, carries out the steps in the wind speed prediction method according to any of the claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110287049.XA CN112949201B (en) | 2021-03-17 | 2021-03-17 | Wind speed prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110287049.XA CN112949201B (en) | 2021-03-17 | 2021-03-17 | Wind speed prediction method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112949201A CN112949201A (en) | 2021-06-11 |
| CN112949201B true CN112949201B (en) | 2023-03-21 |
Family
ID=76228784
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110287049.XA Active CN112949201B (en) | 2021-03-17 | 2021-03-17 | Wind speed prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112949201B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114298444B (en) * | 2022-03-09 | 2022-05-27 | 广东电网有限责任公司佛山供电局 | Wind speed probability prediction method, device, equipment and storage medium |
| CN116644866A (en) * | 2023-07-27 | 2023-08-25 | 国网江西省电力有限公司经济技术研究院 | Comprehensive energy system robust optimization method and system considering wind-light uncertainty |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365033A (en) * | 2020-10-26 | 2021-02-12 | 中南大学 | Wind power interval prediction method, system and storage medium |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018002953A1 (en) * | 2016-06-28 | 2018-01-04 | Indian Institute Of Technology Bombay | Integrated decision support system and method for deriving inferences from data sets |
| CN109063939B (en) * | 2018-11-01 | 2020-08-18 | 华中科技大学 | Wind speed prediction method and system based on neighborhood gate short-term memory network |
| CN111461445B (en) * | 2020-04-09 | 2022-03-15 | 深圳大学 | Short-term wind speed prediction method and device, computer equipment and storage medium |
| CN112308322B (en) * | 2020-11-03 | 2023-11-24 | 华翔翔能科技股份有限公司 | Multi-wind-field space-time wind speed prediction method and device and electronic equipment |
-
2021
- 2021-03-17 CN CN202110287049.XA patent/CN112949201B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365033A (en) * | 2020-10-26 | 2021-02-12 | 中南大学 | Wind power interval prediction method, system and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112949201A (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112202736B (en) | Communication network anomaly classification method based on statistical learning and deep learning | |
| CN106951695A (en) | Plant equipment remaining life computational methods and system under multi-state | |
| CN112949201B (en) | Wind speed prediction method and device, electronic equipment and storage medium | |
| Du et al. | Fast cascading outage screening based on deep convolutional neural network and depth-first search | |
| CN112734128A (en) | 7-day power load peak value prediction method based on optimized RBF | |
| CN116245033B (en) | Artificial intelligent driven power system analysis method and intelligent software platform | |
| Kaplan et al. | A novel method based on Weibull distribution for short-term wind speed prediction | |
| CN114386537A (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN114065634A (en) | Data-driven power quality monitoring and stationing optimization method and device | |
| CN111340287A (en) | Power distribution cabinet operation state prediction method and device | |
| CN117592870B (en) | Comprehensive analysis system based on water environment monitoring information | |
| CN114548498A (en) | Wind speed prediction method and system for local area of overhead transmission line | |
| Ren et al. | Online anomaly detection using machine learning and HPC for power system synchrophasor measurements | |
| CN113151842B (en) | Method and device for determining conversion efficiency of wind-solar complementary water electrolysis hydrogen production | |
| CN117318033B (en) | Power grid data management method and system combining data twinning | |
| CN115858606A (en) | Method, device and equipment for detecting abnormity of time series data and storage medium | |
| CN116739172B (en) | Method and device for ultra-short-term prediction of offshore wind power based on climbing identification | |
| CN113891342A (en) | Base station inspection method and device, electronic equipment and storage medium | |
| CN116578436A (en) | Real-time online detection method based on asynchronous multielement time sequence data | |
| CN113779892B (en) | Method for predicting wind speed and wind direction | |
| CN114970674A (en) | Time sequence data concept drift adaptation method based on relevance alignment | |
| CN114443738A (en) | Abnormal data mining method, device, equipment and medium | |
| CN114510469A (en) | Method, device, equipment and medium for identifying bad data of power system | |
| CN117131947B (en) | Overhead transmission line fault prediction method, device, equipment and storage medium | |
| CN114169704B (en) | Quality control method, device, equipment and storage medium for product production process |
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 | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A wind speed prediction method, device, electronic device, and storage medium Effective date of registration: 20231106 Granted publication date: 20230321 Pledgee: China Co. truction Bank Corp Yiyang branch Pledgor: Huaxiang XiangNeng Technology Co.,Ltd. Registration number: Y2023980063452 |