CN112348255B - Ultra-short-term wind power prediction method based on wavelet time-frequency imaging - Google Patents

Ultra-short-term wind power prediction method based on wavelet time-frequency imaging Download PDF

Info

Publication number
CN112348255B
CN112348255B CN202011232819.2A CN202011232819A CN112348255B CN 112348255 B CN112348255 B CN 112348255B CN 202011232819 A CN202011232819 A CN 202011232819A CN 112348255 B CN112348255 B CN 112348255B
Authority
CN
China
Prior art keywords
time
layer
wavelet
frequency
input data
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
Application number
CN202011232819.2A
Other languages
Chinese (zh)
Other versions
CN112348255A (en
Inventor
何洪英
罗滇生
高思远
蒋宇翔
尹希浩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan University
Original Assignee
Hunan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hunan University filed Critical Hunan University
Priority to CN202011232819.2A priority Critical patent/CN112348255B/en
Publication of CN112348255A publication Critical patent/CN112348255A/en
Application granted granted Critical
Publication of CN112348255B publication Critical patent/CN112348255B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/14—Fourier, Walsh or analogous domain transformations, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms
    • G06F17/148—Wavelet transforms
    • 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
    • G06N3/045—Combinations of networks
    • 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
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy 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
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
    • Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Software Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Human Resources & Organizations (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Mathematics (AREA)
  • Marketing (AREA)
  • Computing Systems (AREA)
  • Mathematical Optimization (AREA)
  • General Business, Economics & Management (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Molecular Biology (AREA)
  • Quality & Reliability (AREA)
  • Primary Health Care (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Development Economics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses an ultra-short-term wind power prediction method based on wavelet time-frequency imaging, which comprises the steps of obtaining an original input data set, and preprocessing input data of the original input data set; performing wavelet time-frequency analysis on the processed input data, extracting time domain features and frequency domain features of historical time sequences, and generating a corresponding time-frequency diagram; building a neural network prediction model according to the characteristics of the time-frequency diagram, the size of an original input data set and the condition of a hardware platform; and adjusting parameters of the neural network prediction model according to the image feature extraction result, the training error result and the final prediction result to obtain an optimal prediction result. According to the method, on one hand, time-frequency information of a time sequence can be fully utilized, and on the other hand, excellent two-dimensional convolution characteristics can be fully utilized, so that accurate ultra-short-term prediction of the time sequence of wind power is realized.

Description

Ultra-short-term wind power prediction method based on wavelet time-frequency imaging
Technical Field
The invention relates to the technical field of time sequence prediction, in particular to an ultra-short-term wind power prediction method based on wavelet time-frequency imaging.
Background
The wavelet transformation is proposed by the french physicist Moler in the 80 s of the 20 th century, is a time-scale (time-frequency) analysis method of signals, and is a time-frequency localization analysis method of which the window size is fixed, but the shape can be changed, and the time window and the frequency window can be changed.
The resolution of the traditional Fourier transform in the time domain and the frequency domain is fixed, if the time window is too long, a plurality of periods of the high-frequency signal are included in one time window, and finally, the average value of the plurality of periods is output, so that the local characteristics of the high-frequency signal cannot be focused; if the time window is too short, although the local features of the high frequency signal may be well focused, the features of the low frequency signal may not be detected within one period.
The wavelet analysis has good local property, so that the wavelet analysis has good characterization capability of signal local characteristics in both time domain and frequency domain, and the inherent defect of single resolution of the traditional Fourier transform is overcome.
Convolutional neural networks (Convolutional Neural Networks, CNN) are one of the most important methods of deep learning and are a popular research direction in the fields of image recognition and speech analysis. Its weight sharing network structure is similar to biological neural network, and can reduce complexity and weight of network model.
If the network inputs a multidimensional image, the image can be directly put into the network, so that complex characteristic extraction and data reconstruction processes in the traditional recognition algorithm can be avoided. As a multilayer perceptron designed specifically for identifying two-dimensional shapes, convolutional neural networks have certain processing advantages for having a high degree of invariance to panning, zooming, tilting, or deforming of pictures.
Ultra-short-term predictions of time series are present in a variety of engineering applications. At present, the ultra-short-term prediction of the time sequence mainly has two problems: on the one hand, the signal has certain nonlinearity and non-stationarity, the time-frequency local property is just the most important characteristic of the non-stationary signal, the existing prediction method cannot simultaneously utilize the time domain characteristics and the frequency domain characteristics of the time sequence, and most of the time sequence belongs to the non-stationary signal, so that a certain lifting space is provided in the aspect of extracting the time-frequency characteristics of the time sequence simultaneously; on the other hand, the existing ultra-short-term prediction models have certain limitations, such as unstable results of the continuous prediction method, difficulty in estimating noise amount by a Kalman filtering method, lack of sparsity of an LS-SVM model and the like, limit improvement of the accuracy of the prediction results, and therefore the final prediction results are inaccurate.
Disclosure of Invention
The invention mainly aims to provide a method for predicting ultra-short-term wind power based on wavelet time-frequency imaging, which aims to solve the problem that the prediction result of the existing time sequence ultra-short-term prediction model is not accurate enough.
In order to achieve the purpose, the ultra-short-term wind power prediction method based on wavelet time-frequency imaging provided by the invention comprises the following steps of:
acquiring historical wind power data as an original input data set, and preprocessing the input data of the original input data set;
performing wavelet time-frequency analysis on the processed input data, acquiring wavelet coefficients, and generating a corresponding two-dimensional time-frequency diagram according to time domain features and frequency domain features;
building a neural network prediction model for processing the time-frequency diagram according to the characteristics of the time-frequency diagram, the size of the original input data set and the condition of a hardware platform;
according to an original input data set and the neural network prediction model, an image feature extraction result, a training error result and a final prediction result are obtained, and parameters of the neural network prediction model are adjusted according to the image feature extraction result, the training error result and the final prediction result, so that an optimal wind power ultra-short-term prediction result is obtained.
Preferably, the step of acquiring historical wind power data as an original input data set and preprocessing the input data of the original input data set includes the following steps:
acquiring historical wind power data as an original input data set, and acquiring the maximum value and the minimum value of input data in the original input data set;
and normalizing the input data in the original input data set according to the maximum value of the input data and the minimum value of the input data.
Preferably, the step of performing wavelet time-frequency analysis on the processed input data and obtaining wavelet coefficients, and generating a corresponding two-dimensional time-frequency diagram according to time domain features and frequency domain features includes the following steps:
according to the processed input data, wavelet coefficients under different frequencies are obtained to extract time domain features and frequency domain features of historical time sequences;
and combining the wavelet coefficients under different frequencies according to the time frequency to generate a wavelet time-frequency diagram.
Preferably, the step of building a neural network prediction model for processing the time-frequency diagram according to the characteristics of the time-frequency diagram, the size of the original input data set and the condition of a hardware platform includes the following steps:
Establishing an input layer according to the characteristics of the time-frequency diagram;
establishing an alternating structure of a plurality of convolution layers and pooling layers according to the input layer so as to extract image features of the time-frequency diagram;
establishing a full-connection layer according to the convolution layer and the pooling layer, wherein the full-connection layer reduces the dimension of a characteristic image obtained by the last pooling layer, converts the characteristic image into a characteristic vector, and outputs a final prediction result;
and building a neural network prediction model according to the size of the original input data set, the condition of a hardware platform, the input layer, the full-connection layer, a plurality of convolution layers and a plurality of pooling layers which are alternately arranged.
Preferably, the step of reducing the dimension of the feature image obtained by the last pooling layer processing by the full connection layer and converting the feature image into a feature vector includes the following steps:
and converting the feature image obtained by the last pooling layer into a one-dimensional feature vector through a flat algorithm.
Preferably, the step of establishing an alternating structure of a plurality of convolution layers and pooling layers according to the input layer comprises the steps of:
according to the input layer, a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a third convolution layer and a third pooling layer which are sequentially connected are established; the first convolution layer, the second convolution layer and the third convolution layer are used for eliminating invalid information of the time-frequency diagram layer by layer and extracting key features; the first pooling layer, the second pooling layer and the third pooling layer are used for further extracting image features and compressing feature images;
Respectively establishing convolution kernels for extracting image features for the first convolution layer, the second convolution layer and the third convolution layer, wherein each convolution kernel corresponds to one image feature;
and determining a pooling method of the first pooling layer, the second pooling layer and the third pooling layer.
Preferably, the activation function of the first convolution layer is a hyperbolic tangent activation function, and the activation function of the second convolution layer is a linear rectification activation function; the activation function of the third convolution layer is a hyperbolic tangent activation function.
Preferably, the convolution kernel size of the first convolution layer, the convolution kernel size of the second convolution layer, and the convolution kernel size of the third convolution layer decrease in order.
Preferably, the first pooling layer, the second pooling layer and the third pooling layer each use a maximum combining method to extract the time-frequency features.
Preferably, the step of obtaining an image feature extraction result, a training error result and a final prediction result according to an original input data set and the neural network prediction model, and adjusting parameters of the neural network prediction model according to the image feature extraction result, the training error result and the final prediction result to obtain an optimal prediction result includes the following steps:
Establishing a test set, a training set and a model input set according to the original input data set;
acquiring the image feature extraction result, the training error result and the final prediction result according to the test set, the training set and the neural network prediction model so as to adjust the learning rate, the training times and the convolution kernel size of each convolution layer of the neural network prediction model;
and according to the model input set and the wind power prediction result obtained by the adjusted neural network prediction model.
In the technical scheme of the invention, a wavelet analysis is adopted to convert a time sequence into a time-frequency diagram sequence so as to fully utilize the time domain characteristics and the frequency domain characteristics of the time sequence, then a convolutional neural network is adopted to extract the time-frequency diagram characteristics, a back propagation algorithm is utilized to optimize the network structure, and finally the final output is obtained through regression. According to the method, on one hand, time-frequency information of a time sequence can be fully utilized, and on the other hand, excellent two-dimensional convolution characteristics can be fully utilized, so that accurate ultra-short-term prediction of the time sequence is realized, and an accurate ultra-short-term prediction result of wind power is obtained.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to the structures shown in these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a time series ultrashort-term prediction model based on wavelet analysis and convolutional neural network in the present invention.
Fig. 2 is a schematic diagram of a convolutional neural network structure.
FIG. 3 is a graph of the relationship between wind power and data points for month 12 of 2014
Fig. 4 is a time-frequency diagram of 2014, month 2 and 12 wind power.
FIG. 5 is a graph of wind power versus data points for month 13 of 2014.
Fig. 6 is a time-frequency diagram of 2014, 2-month 13-year wind power.
FIG. 7 is a graph of wind power versus data points for month 14 of 2014.
Fig. 8 is a time-frequency diagram of 2014, 2 months and 14 # wind power.
Fig. 9 is a schematic diagram of convolution operation.
Fig. 10 is a flowchart of three different pooling layer sub-sampling training.
Fig. 11 is a view of a visualization result of each convolution layer of a Convolutional Neural Network (CNN) model for performing ultra-short-term prediction on wind power by using the ultra-short-term wind power prediction method based on wavelet time-frequency imaging.
FIG. 12 shows the recognition accuracy corresponding to the case that different convolution kernels are selected for the first convolution layer, the second convolution layer and the third convolution layer for carrying out ultra-short-term prediction on wind power based on the wavelet time-frequency imaging ultra-short-term wind power prediction method.
FIG. 13 is a graph of training error of a prediction model when the model learning rate of ultra-short-term prediction of wind power is 0.01, based on the ultra-short-term wind power prediction method based on wavelet time-frequency imaging.
Fig. 14 is a training error curve of a prediction model when the model learning rate of ultra-short-term prediction of wind power by the ultra-short-term wind power prediction method based on wavelet time-frequency imaging is 0.012.
FIG. 15 is a graph of a predicted model training error when the model learning rate of ultra-short-term prediction of wind power is 0.014, based on a wavelet time-frequency imaging ultra-short-term wind power prediction method of the present invention.
FIG. 16 is a graph of prediction result of a prediction model for performing ultra-short-term prediction on wind power by the ultra-short-term wind power prediction method based on wavelet time-frequency imaging.
The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that all directional indicators (such as up, down, left, right, front, and rear … …) in the embodiments of the present invention are merely used to explain the relative positional relationship, movement, etc. between the components in a particular posture (as shown in the drawings), and if the particular posture is changed, the directional indicator is changed accordingly.
Furthermore, descriptions such as those referred to as "first," "second," and the like, are provided for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implying an order of magnitude of the indicated technical features in the present disclosure. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
In the present invention, unless specifically stated and limited otherwise, the terms "connected," "affixed," and the like are to be construed broadly, and for example, "affixed" may be a fixed connection, a removable connection, or an integral body; can be mechanically or electrically connected; either directly or indirectly, through intermediaries, or both, may be in communication with each other or in interaction with each other, unless expressly defined otherwise. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In addition, the technical solutions of the embodiments of the present invention may be combined with each other, but it is necessary to be based on the fact that those skilled in the art can implement the technical solutions, and when the technical solutions are contradictory or cannot be implemented, the combination of the technical solutions should be considered as not existing, and not falling within the scope of protection claimed by the present invention.
Referring to fig. 1 of fig. 1 to 16, in order to achieve the above objective, the ultra-short-term wind power prediction method based on wavelet time-frequency imaging provided by the present invention includes the following steps:
s100, acquiring historical wind power data as an original input data set, and preprocessing the input data of the original input data set;
s200, carrying out wavelet time-frequency analysis on the processed input data, obtaining wavelet coefficients, and generating a corresponding two-dimensional time-frequency diagram according to time domain features and frequency domain features;
s300, building a neural network prediction model for processing the time-frequency diagram according to the characteristics of the time-frequency diagram, the original input data set and the conditions of a hardware platform;
s400, acquiring an image feature extraction result, a training error result and a final prediction result according to an original input data set and the neural network prediction model, and adjusting parameters of the neural network prediction model according to the acquired image feature extraction result, the training error result and the final prediction result to acquire an optimal wind power ultra-short-term prediction result. In the technical scheme of the invention, a wavelet analysis is adopted to convert a time sequence into a time-frequency diagram, the time domain characteristics and the frequency domain characteristics of the time sequence are fully utilized, then a convolutional neural network is adopted to extract the time-frequency diagram characteristics, a back propagation algorithm is utilized to optimize the network structure, and finally the final output is obtained through regression. According to the method, on one hand, time-frequency information of a time sequence can be fully utilized, and on the other hand, excellent two-dimensional convolution characteristics can be fully utilized, so that accurate ultra-short-term prediction of the time sequence is realized, and an accurate ultra-short-term prediction result of wind power is obtained.
Specifically, the method comprises the steps of carrying out normalization pretreatment on the original time sequence, facilitating the subsequent use of data and the rapid convergence of a prediction program, simultaneously extracting the time domain features and the frequency domain features of the sequence through wavelet time-frequency analysis on the pretreated time sequence, imaging the time domain features and the frequency domain features of the time sequence, and accurately reflecting the local features of the time sequence in a time domain.
Because of the weight sharing structure of the convolutional neural network, the method has high invariance to the translation, the scaling and the tilting of images and other forms of deformation, effectively avoids the pre-processing process of the images, and can directly process the original images, so that the time-frequency diagram obtained by wavelet time-frequency analysis processed by the convolutional neural network can fully utilize the excellent characteristics of the convolutional neural network, extract the characteristics of the time-frequency diagram, complete the accurate prediction of time sequences, avoid the problems that the noise statistics are difficult to estimate and the LS-SVM lacks sparsity in the common Kalman filtering method, and have wide application prospects.
Referring to fig. 1, in a first embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, in a second embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the step S100 includes the following steps:
S110, acquiring historical wind power data as an original input data set, and acquiring the maximum value and the minimum value of input data in the original input data set;
s120, carrying out normalization processing on the input data in the original input data set according to the maximum value of the input data and the minimum value of the input data.
Specifically, in order to avoid the influence of the overlarge unit level of the input data on the accuracy of the subsequent prediction, the data needs to be normalized in advance, and this step is to facilitate the subsequent data processing and ensure the rapid convergence of the program.
The input data and the output data subjected to normalization processing are controlled in the range of [0,1], and the specific normalization operation is as shown in formula (1).
Wherein: p is p n Is the original input power data; p is p min Representing a minimum value in the original dataset; p is p max Representing the maximum in the original dataset;is normalized power data.
Referring to fig. 3 to 8, in a third embodiment of the method for predicting ultra-short-term wind power based on wavelet time-frequency imaging according to the present invention, the step S200 includes the following steps:
S210, acquiring wavelet coefficients under different frequencies according to the processed input data so as to extract time domain features and frequency domain features of a historical time sequence;
s220, combining the wavelet coefficients at different frequencies according to the time frequency to generate a wavelet time frequency diagram.
Specifically, when a function ψ (t) ∈L 2 (R)(L 2 (R) represents the Fourier transform ψ (ω) of a space where the signal energy is limited * When the condition shown in the following formula (2) is satisfied, the function is a wavelet basis function, the basis function can obtain a wavelet sequence through translation and stretching, the continuous wavelet sequence is shown in the formula (3), and the discrete wavelet sequence is shown in the formula (4).
Wherein: a is a scale factor; b is a time shift factor.
ψ j,t (t)＝2 -j/2 ψ(2 -j -k)，j,k∈R；j≠0 (4)；
The wavelet basis functions due to the self-contained scale factors and time shift factors can focus different portions of the data by varying the length and location of the wavelet.
Thus, at low frequencies, wavelet analysis has a higher frequency resolution and a lower time resolution; in the high frequency part, the wavelet analysis method has a higher time resolution and a lower frequency resolution.
The high-precision local feature extraction function is the most important and practical feature of wavelet analysis. The wavelet analysis method can select different wavelet basis functions for different engineering objects.
In the wavelet operation process, a wavelet function starts from a starting point of a signal, convolves the signal with window signals at different positions continuously on a time axis, obtains wavelet coefficients corresponding to local signals through comparison until the whole signal completes convolution operation, stretches the selected wavelet function by one unit, and repeats the process until all scales complete the steps. The magnitude of the wavelet coefficients represents the degree to which the waveform of the wave function resembles the waveform of the local signal at that moment. The larger the wavelet coefficients, the higher the similarity of the selected wavelet waveform and the waveform of the local signal at that moment.
In the frequency domain, wavelets of different lengths and frequencies are obtained by stretching and compressing the length of the wavelets. In the high frequency band, the wavelet length is compressed, the wavelet coefficient is reduced, the window is narrowed, the time resolution is improved, and the frequency resolution is reduced; in the low frequency band, the wavelet length is stretched, the window is widened, the time resolution is reduced, and the frequency resolution is improved.
The wavelet time-frequency diagram is obtained by combining wavelet coefficients at different frequencies. Different wavelet basis functions are selected for different time series. The dbN wavelet has outstanding capability of extracting the characteristic of the unstable signal and is widely applied to the analysis of the unstable signal.
Referring to fig. 1 and 2, in a first embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, in a fourth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the step S300 includes:
s310, an input layer is established according to the characteristics of the time-frequency diagram;
s320, establishing an alternating structure of a plurality of convolution layers and a pooling layer according to the input layer so as to extract the image characteristics of the time-frequency diagram;
s330, establishing a full-connection layer according to the convolution layer and the pooling layer, wherein the full-connection layer reduces the dimension of a feature image obtained by the last pooling layer, converts the feature image into a feature vector, and outputs a final prediction result;
and S340, building a neural network prediction model according to the size of the original input data set, the condition of a hardware platform, the input layer, the full-connection layer, a plurality of convolution layers and a plurality of pooling layers which are alternately arranged.
Specifically, as can be seen from fig. 2, the basic structure of the convolutional neural network includes several parts of an input layer, a convolutional layer, a pooling layer, a full-connection layer and an output layer, each of the convolutional neural networks includes a plurality of alternating structures of the convolutional layer and the pooling layer, and the structure is equivalent to an "image filter" for extracting image features, and the conversion from low-dimensional features to high-dimensional features of the image is realized through continuous filtration of the filter. The full-connection layer is equivalent to a classifier, receives the image transmitted by the front end, reduces the dimension of the feature image transmitted by the front end through a corresponding algorithm, converts the feature image into a feature vector, and then outputs a final prediction result. The convolution layer is a key part of the convolution neural network, which is different from the traditional neural network, and is mainly used for extracting the characteristics of the image. In image processing, a filter (also called convolution kernel) is required to perform convolution operation on an image to extract image features.
Specific operations are shown in fig. 9, the matrix formed by the adjacent pixels of the image is multiplied by the matrix corresponding elements of the filter, and then the obtained products are added to obtain the processed pixel value, and the filter is slid in turn until all the pixel points complete convolution operation. For the image edge, there is no method to directly perform convolution operation, and edge elements need to be filled, and the filled pixel value is generally selected to be 0 or a central pixel value so as to ensure the normal operation of convolution operation.
Referring to fig. 1, in a fifth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the fourth embodiment of the invention, the step of reducing the dimension of the feature image obtained by processing the last pooling layer by the full-connection layer in the step S330 and converting the feature image into a feature vector includes the following steps:
and converting the feature image obtained by the last pooling layer into a one-dimensional feature vector through a flat algorithm.
Specifically, the last pooling layer converts the processed two-dimensional image into a one-dimensional feature vector through a flat algorithm, and finally ultra-short-term prediction of the time sequence is realized through a full-connection layer.
Referring to fig. 1, in a sixth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the fourth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the step of establishing an alternating structure of a plurality of convolution layers and pooling layers according to the input layer in the step S320 includes:
s321, establishing a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a third convolution layer and a third pooling layer which are sequentially connected according to the input layer; the first convolution layer, the second convolution layer and the third convolution layer are used for eliminating invalid information of the time-frequency diagram layer by layer, extracting key features, and the first pooling layer, the second pooling layer and the third pooling layer are used for further extracting image features and compressing feature images;
s322, respectively establishing convolution kernels for extracting image features for the first convolution layer, the second convolution layer and the third convolution layer, wherein each convolution kernel corresponds to one image feature;
s323, determining the pooling methods of the first pooling layer, the second pooling layer and the third pooling layer.
Specifically, the model increases network capacity, reduces feature resolution and the number of network parameters to be optimized through an alternating structure of a convolution layer and a pooling layer, and realizes feature extraction of a time-frequency feature map. As shown in fig. 1, a 7-layer convolutional neural network model (including a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a third convolutional layer, a third pooling layer, and a full connection layer connected in sequence) is built, as shown in fig. 1.
Referring to fig. 1, in a seventh embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the activation function of the first convolution layer is a hyperbolic tangent activation function (tanh) activation function, and the activation function of the second convolution layer is a linear rectification function (ReLU) activation function; the activation function of the third convolution layer is a hyperbolic tangent activation function (tanh) activation function.
Specifically, the neural network generally solves the problem of nonlinearity, and the activation function is to compensate the expression capability of the linear model and add a certain nonlinear factor to the neural network.
Common activation functions are the sigmoid function, the hyperbolic tangent (tanh) function, and the linear rectification (ReLU) function.
The sigmoid function is shown as follows:
the output interval is (0, 1), and the output result is monotonously continuous without 0 as the output center, but the function itself has soft saturation, so that the problem of gradient disappearance is easy to occur, and the function is generally used for an output layer.
The hyperbolic tangent function is shown in the following formula:
the output interval is (-1, 1), 0 is taken as the output center, and the hyperbolic tangent function has the problem of gradient disappearance like the sigmoid function, but the training speed is improved to a certain extent, and the method is generally used for hidden layers.
The linear rectified ReLU function is shown as follows:
when the input is smaller than 0, the output is 0, and when the input is larger than 0, the data is output, so that the gradient explosion problem of the first two activation functions is solved, the training speed is improved, the performance is good in unsupervised learning, and the method is an activation function widely applied at the present stage.
Referring to fig. 1 and 9, in an eighth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the convolution kernel size of the first convolution layer, the convolution kernel size of the second convolution layer, and the convolution kernel size of the third convolution layer are sequentially reduced.
Specifically, the different sizes of the convolution kernels have different training effects, and the smaller the convolution kernel scale is, the faster the corresponding model training speed is, the higher the model training loss is, and the lower the feature recognition accuracy is; the larger the convolution kernel scale is, the more gradual the corresponding model training speed is, the lower the model training loss is, and the feature recognition accuracy (array) is relatively improved. And selecting the convolution kernel scale of each convolution layer according to the feature recognition accuracy of the convolution kernels of different scales in different convolution layers.
Referring to fig. 1 and fig. 2, according to a sixth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging of the present invention, in a ninth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging of the present invention, the first pooling layer, the second pooling layer and the third pooling layer all use a maximum merging method to extract time-frequency characteristics.
Specifically, the pooling layer after the convolution layer is a structure for further extracting image features, compressing images, reducing the calculated amount and preventing the model from being over fitted. The pooling layer achieves the purpose of reducing the data scale mainly through sub-sampling.
For example, sub-sampling a 4×4 image by 2×2 results in a 2×2 image, which is equivalent to extracting a representative pixel from every 4 pixels in the original image as an output, greatly reducing the computational effort of the overall model.
Common pooling methods are maximum merging, average merging and random merging, and most commonly, maximum merging is used, as shown in fig. 10.
Referring to fig. 1, in a tenth embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to the present invention, the step S400 includes the following steps:
S410, a test set, a training set and a model input set are established according to the original input data set;
s420, according to the test set, the training set and the neural network prediction model, obtaining the image feature extraction result, the training error result and the final prediction result, and adjusting the learning rate, the training times and the convolution kernel size of each convolution layer of the neural network prediction model;
s430, obtaining an optimal wind power ultra-short-term prediction result according to the model input set and the adjusted neural network prediction model. Specifically, the whole training process of the prediction model adopts a batch gradient descent algorithm, so that the network structure can be optimized by using an inverse regression algorithm, and the BP algorithm (i.e. a back propagation algorithm) is suitable for a learning algorithm of the multi-layer neuron network and is based on a gradient descent method. The input-output relationship of the BP network is essentially a mapping relationship: an n-input m-output BP neural network performs the function of a continuous mapping from n-dimensional Euclidean space to a finite field in m-dimensional Euclidean space, which mapping is highly nonlinear. Its information processing capability is derived from multiple complex of simple nonlinear functions, and thus has a strong function reproduction capability.
In another embodiment of the ultra-short-term wind power prediction method based on wavelet time-frequency imaging, the actual calculation example of ultra-short-term prediction of wind power is performed by using the ultra-short-term wind power prediction method based on wavelet time-frequency imaging.
The data set used in the method is winter data from 2014, 1 month, 1 to 2014, 3 month, 1 of a certain wind farm in northeast China, the weather of the wind farm does not change obviously in the period, and wind power generation data are less affected by weather. The wind power plant collects wind power values at 15-minute intervals, a total of 5000 data points are selected as an original input data set, normalization processing is carried out on the data, and the obtained data are used for subsequent prediction.
The data set of 1 month is used herein as the training set and the data of 2 months is used as the test set. The input set of the model is a three-day wind power history time sequence, the image size of the obtained time-frequency diagram is uniformly adjusted to 300 multiplied by 150 for facilitating subsequent processing, and the output is the power of the wind power plant 15 minutes in the future.
The dbN wavelet has outstanding capability of extracting the characteristic of the unstable signal and is widely applied to the analysis of the unstable signal. Because the wind power history time sequence belongs to an unstable signal, the embodiment adopts a tightly supported orthogonal wavelet (db 6 wavelet) to extract time-frequency characteristics of the wind power history time sequence under different frequencies, and the db6 wavelet has excellent characteristics of good regularity, large vanishing moment and the like. And (3) performing time-frequency analysis on wind power in different frequency bands through telescopic translation of db6 wavelets, and extracting local frequency domain characteristics of data.
1. Feature extraction analysis
And selecting a wind power history time sequence of from No. 12 to No. 14 of 2 months in 2014, and performing wavelet time-frequency joint analysis on the wind power history time sequence of 96 data points (24 multiplied by 60 and 15) per day, wherein a time-frequency analysis chart is shown in the following figures 3-8.
As shown in fig. 1 and 11, the first, second and third convolution layers establish 4, 8 and 16 convolution kernels, respectively, one convolution kernel corresponding to each image feature.
Therefore, the first convolution layer, the second convolution layer and the third convolution layer respectively extract 4, 8 and 16 image features, in order to further understand the characteristic extraction process of the convolution neural network, the image features extracted by each layer are visualized by a code inversion method, the image features of the first convolution layer, the second convolution layer and the third convolution layer are respectively visualized, and the extracted features of different convolution kernels are also different, as shown in fig. 6.
The first convolution layer extracts edge features of the image, retains basic information of the image, can see the outline of the image, and has certain differences between feature_maps corresponding to different convolution kernels, so that different convolution kernels can be proved to recognize different image features.
With the continuous deepening of the layer number, the image features extracted by the convolution layers are more and more abstract, invalid information is further filtered, and the second convolution kernel and the third convolution kernel have more blank contents because no matched image texture features are found. And removing invalid information layer by the time-frequency image through the first convolution layer, the second convolution layer and the third convolution layer, and extracting key features.
The recognition accuracy corresponding to different convolution kernels is selected for the first convolution layer, the second convolution layer and the third convolution layer as shown in fig. 12.
The accuracy of selecting different convolution kernels for each convolution layer is shown in table 1.
Table 1 convolutional layer selects feature recognition accuracy corresponding to different convolutional kernels
As can be seen from fig. 7 and table 1, the feature recognition accuracy is highest, which is 94.912% when the convolution kernel size of the first convolution layer is 8×8; the second convolution layer has the highest characteristic recognition accuracy rate of 91.791 percent when the convolution kernel size is 4 multiplied by 4; the third convolution layer has the highest feature recognition accuracy of 92.413% when the convolution kernel size is 2×2. Thus, the first convolution layer, the second convolution layer and the third convolution layer sequentially select convolution kernels with the sizes of 8×8, 4×4 and 2×2.
2. Training error analysis
The predictive model training error curves are shown in fig. 13-15.
As is clear from fig. 14, when the learning rate is 0.012, the model training set and the test set have the smallest error, and the test set error is lower than the training set, so the learning rate of the model is 0.012.
The whole training process of the prediction model adopts a batch gradient descent algorithm, and the error loss of the model is given by an absolute average error.
The training times of the model are important factors influencing the prediction accuracy, and excessive training times can cause the overfitting of training data; too low a number of training results in a lack of fit of the training data.
As can be seen from fig. 14, the training error and the test error of the model always maintain a steady downward trend, and the model error is stabilized around 0.01 at about 80 times. Thus, the number of training of the model was chosen 80 times.
3. Prediction result analysis
The prediction result of the prediction model is shown in fig. 16.
As can be seen from fig. 16, the fitting effect of the prediction curve based on the model output and the actual wind power curve basically reaches the expectation, and although some places come in and go out, the trend of the general trend is basically consistent, so that the expectation can be reached, and the model prediction effect is relatively good.
The root mean square error (square root of the ratio of the square of the deviation of the predicted value from the true value to the number of observations) of the model was 0.241.
The method provided by the invention has the advantages that the accuracy of the ultra-short-term prediction of the wind power time sequence can be effectively improved by combining the wavelet time-frequency joint analysis with the convolutional neural network.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part in the form of a software product stored in a computer readable storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device to enter the method according to the embodiments of the present invention.
In the description of the present specification, the descriptions of the terms "one embodiment," "another embodiment," "other embodiments," or "first through X-th embodiments," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, method steps or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
The foregoing description of the preferred embodiments of the present invention should not be taken as limiting the scope of the invention, but rather should be utilized to describe the same or different equivalent structures or processes as may be utilized by the present invention or to adapt it to other technical fields related thereto, either directly or indirectly.

Claims (8)

1. The ultra-short-term wind power prediction method based on wavelet time-frequency imaging is characterized by comprising the following steps of:
acquiring historical wind power data as an original input data set, and preprocessing the input data of the original input data set;
Performing wavelet time-frequency analysis on the processed input data, acquiring wavelet coefficients, and generating a corresponding two-dimensional time-frequency diagram according to time domain features and frequency domain features;
according to the characteristics of the time-frequency diagram, the size of the original input data set and the conditions of a hardware platform, a neural network prediction model for processing the time-frequency diagram is built, and the neural network prediction model adopts a BP neural network;
acquiring an image feature extraction result, a training error result and a final prediction result according to an original input data set and the neural network prediction model, and adjusting parameters of the neural network prediction model according to the image feature extraction result, the training error result and the final prediction result to acquire an optimal wind power ultra-short-term prediction result;
the step of performing wavelet time-frequency analysis on the processed input data, obtaining wavelet coefficients, and generating a corresponding two-dimensional time-frequency diagram according to time domain features and frequency domain features, comprises the following steps: according to the processed input data, wavelet coefficients under different frequencies are obtained to extract time domain features and frequency domain features of historical time sequences; combining the wavelet coefficients under different frequencies according to the time frequency to generate a wavelet time-frequency diagram;
When a function psi (t) epsilon L 2 Fourier transform ψ (ω) of (R) * When the condition shown in the following formula (2) is satisfied, the function is a wavelet basis function, the basis function obtains a wavelet sequence through translation and stretching, the continuous wavelet sequence is shown in the formula (3), and the discrete wavelet sequence is shown in the formula (4);
wherein: a is a scale factor; b is a time shift factor;
ψ j,t (t)＝2 -j/2 ψ(2 -j -k)，j,k∈R；j≠0(4)；
the wavelet basis function with the scale factor and time shift factor focuses different portions of the data by varying the length and position of the wavelet;
wherein L is 2 (R) represents a space where signal energy is limited;
in the wavelet operation process, a wavelet function starts from a starting point of a signal, convolves the window signals at different positions through continuous movement on a time axis, obtains wavelet coefficients corresponding to local signals through comparison until the whole signal completes convolution operation, stretches the selected wavelet function by one unit, and repeats the process until all scales complete the steps;
the wavelet time-frequency diagram is obtained by combining wavelet coefficients at different frequencies; selecting different wavelet basis functions for different time sequences;
the step of building a neural network prediction model for processing the time-frequency diagram according to the characteristics of the time-frequency diagram, the size of the original input data set and the condition of a hardware platform comprises the following steps:
Establishing an input layer according to the characteristics of the time-frequency diagram;
establishing an alternating structure of a plurality of convolution layers and pooling layers according to the input layer so as to extract image features of the time-frequency diagram;
establishing a full-connection layer according to the convolution layer and the pooling layer, wherein the full-connection layer reduces the dimension of a characteristic image obtained by the last pooling layer, converts the characteristic image into a characteristic vector, and outputs a final prediction result;
and building a neural network prediction model according to the size of the original input data set, the condition of a hardware platform, the input layer, the full-connection layer, a plurality of convolution layers and a plurality of pooling layers which are alternately arranged.
2. The ultra-short term wind power prediction method based on wavelet time-frequency imaging according to claim 1, wherein the step of acquiring historical wind power data as an original input data set and preprocessing input data of the original input data set comprises the steps of:
acquiring historical wind power data as an original input data set, and acquiring the maximum value and the minimum value of input data in the original input data set;
And normalizing the input data in the original input data set according to the maximum value of the input data and the minimum value of the input data.
3. The ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to claim 1, wherein the step of reducing the dimension of the feature image obtained by the last pooling layer processing by the full-connection layer and converting the feature image into a feature vector comprises the following steps:
and converting the feature image obtained by the last pooling layer into a one-dimensional feature vector through a flat algorithm.
4. The ultra-short term wind power prediction method based on wavelet time-frequency imaging according to claim 1, wherein said step of establishing an alternating structure of a plurality of convolution layers and pooling layers according to said input layer comprises the steps of:
according to the input layer, a first convolution layer, a first pooling layer, a second convolution layer, a second pooling layer, a third convolution layer and a third pooling layer which are sequentially connected are established; the first convolution layer, the second convolution layer and the third convolution layer are used for eliminating invalid information of the time-frequency diagram layer by layer and extracting key features; the first pooling layer, the second pooling layer and the third pooling layer are used for further extracting image features and compressing feature images;
Respectively establishing convolution kernels for extracting image features for the first convolution layer, the second convolution layer and the third convolution layer, wherein each convolution kernel corresponds to one image feature;
and determining a pooling method of the first pooling layer, the second pooling layer and the third pooling layer.
5. The ultra-short term wind power prediction method based on wavelet time-frequency imaging according to claim 4, wherein the activation function of the first convolution layer is a hyperbolic tangent activation function, and the activation function of the second convolution layer is a linear rectification activation function; the activation function of the third convolution layer is a hyperbolic tangent activation function.
6. The ultra-short term wind power prediction method based on wavelet time-frequency imaging according to claim 4, wherein the convolution kernel size of the first convolution layer, the convolution kernel size of the second convolution layer, and the convolution kernel size of the third convolution layer decrease in order.
7. The ultra-short term wind power prediction method based on wavelet time-frequency imaging according to claim 4, wherein the first pooling layer, the second pooling layer and the third pooling layer each use a maximum merging method to extract time-frequency characteristics.
8. The ultra-short-term wind power prediction method based on wavelet time-frequency imaging according to any one of claims 1-7, wherein the step of obtaining an image feature extraction result, a training error result and a final prediction result according to an original input data set and the neural network prediction model, and adjusting parameters of the neural network prediction model according to the image feature extraction result, the training error result and the final prediction result to obtain an optimal prediction result comprises the following steps:
establishing a test set, a training set and a model input set according to the original input data set;
acquiring the image feature extraction result, the training error result and the final prediction result according to the test set, the training set and the neural network prediction model so as to adjust the learning rate, the training times and the convolution kernel size of each convolution layer of the neural network prediction model;
and obtaining an optimal wind power ultra-short-term prediction result according to the model input set and the adjusted neural network prediction model.
CN202011232819.2A 2020-11-06 2020-11-06 Ultra-short-term wind power prediction method based on wavelet time-frequency imaging Active CN112348255B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011232819.2A CN112348255B (en) 2020-11-06 2020-11-06 Ultra-short-term wind power prediction method based on wavelet time-frequency imaging

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011232819.2A CN112348255B (en) 2020-11-06 2020-11-06 Ultra-short-term wind power prediction method based on wavelet time-frequency imaging

Publications (2)

Publication Number Publication Date
CN112348255A CN112348255A (en) 2021-02-09
CN112348255B true CN112348255B (en) 2024-04-09

Family

ID=74429739

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011232819.2A Active CN112348255B (en) 2020-11-06 2020-11-06 Ultra-short-term wind power prediction method based on wavelet time-frequency imaging

Country Status (1)

Country Link
CN (1) CN112348255B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115081489A (en) * 2022-07-13 2022-09-20 重庆大学 Time sequence classification method based on wavelet decomposition matrix and residual error network
CN117313927A (en) * 2023-09-19 2023-12-29 华能澜沧江水电股份有限公司 Wind power generation power prediction method and system based on wavelet neural network

Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1288766A2 (en) * 2001-08-31 2003-03-05 Sony United Kingdom Limited Digital content distribution
CN102102626A (en) * 2011-01-30 2011-06-22 华北电力大学 Method for forecasting short-term power in wind power station
JP2013114625A (en) * 2011-11-30 2013-06-10 A T Communications Co Ltd Mobile communication terminal, settlement system, settlement method, and program
CN105184391A (en) * 2015-08-19 2015-12-23 国网山东省电力公司电力科学研究院 Method for predicting wind speed and power of wind farm based on wavelet decomposition and support vector machine
CN106384170A (en) * 2016-09-24 2017-02-08 华北电力大学（保定） Wavelet decomposition and reconstruction-based time sequence wind speed prediction method
CN107292446A (en) * 2017-07-03 2017-10-24 西南交通大学 A kind of mixing wind speed forecasting method based on consideration component relevance wavelet decomposition
CN107507097A (en) * 2017-07-03 2017-12-22 上海电力学院 A kind of short-term wind power prediction method
WO2018028255A1 (en) * 2016-08-11 2018-02-15 深圳市未来媒体技术研究院 Image saliency detection method based on adversarial network
CN108448610A (en) * 2018-03-12 2018-08-24 华南理工大学 A kind of short-term wind power prediction method based on deep learning
CN109214575A (en) * 2018-09-12 2019-01-15 河海大学 A kind of super short-period wind power prediction technique based on small wavelength short-term memory network
CN109272156A (en) * 2018-09-12 2019-01-25 河海大学 A kind of super short-period wind power probability forecasting method
CN110045348A (en) * 2019-05-05 2019-07-23 应急管理部上海消防研究所 A kind of human motion state classification method based on improvement convolutional neural networks
CN110082841A (en) * 2019-04-18 2019-08-02 东华大学 A kind of short-term wind speed forecasting method
CN110415180A (en) * 2019-06-10 2019-11-05 西安电子科技大学 A kind of SAR image denoising method based on wavelet convolution neural network
CN110618353A (en) * 2019-10-22 2019-12-27 北京航空航天大学 Direct current arc fault detection method based on wavelet transformation + CNN
CN111091233A (en) * 2019-11-26 2020-05-01 江苏科技大学 Wind power plant short-term wind power prediction modeling method based on wavelet analysis and multi-model AdaBoost depth network
WO2020156348A1 (en) * 2019-01-31 2020-08-06 青岛理工大学 Structural damage identification method based on ensemble empirical mode decomposition and convolution neural network
CN111507884A (en) * 2020-04-19 2020-08-07 衡阳师范学院 Self-adaptive image steganalysis method and system based on deep convolutional neural network

Patent Citations (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1288766A2 (en) * 2001-08-31 2003-03-05 Sony United Kingdom Limited Digital content distribution
CN102102626A (en) * 2011-01-30 2011-06-22 华北电力大学 Method for forecasting short-term power in wind power station
JP2013114625A (en) * 2011-11-30 2013-06-10 A T Communications Co Ltd Mobile communication terminal, settlement system, settlement method, and program
CN105184391A (en) * 2015-08-19 2015-12-23 国网山东省电力公司电力科学研究院 Method for predicting wind speed and power of wind farm based on wavelet decomposition and support vector machine
WO2018028255A1 (en) * 2016-08-11 2018-02-15 深圳市未来媒体技术研究院 Image saliency detection method based on adversarial network
CN106384170A (en) * 2016-09-24 2017-02-08 华北电力大学（保定） Wavelet decomposition and reconstruction-based time sequence wind speed prediction method
CN107292446A (en) * 2017-07-03 2017-10-24 西南交通大学 A kind of mixing wind speed forecasting method based on consideration component relevance wavelet decomposition
CN107507097A (en) * 2017-07-03 2017-12-22 上海电力学院 A kind of short-term wind power prediction method
CN108448610A (en) * 2018-03-12 2018-08-24 华南理工大学 A kind of short-term wind power prediction method based on deep learning
CN109214575A (en) * 2018-09-12 2019-01-15 河海大学 A kind of super short-period wind power prediction technique based on small wavelength short-term memory network
CN109272156A (en) * 2018-09-12 2019-01-25 河海大学 A kind of super short-period wind power probability forecasting method
WO2020156348A1 (en) * 2019-01-31 2020-08-06 青岛理工大学 Structural damage identification method based on ensemble empirical mode decomposition and convolution neural network
CN110082841A (en) * 2019-04-18 2019-08-02 东华大学 A kind of short-term wind speed forecasting method
CN110045348A (en) * 2019-05-05 2019-07-23 应急管理部上海消防研究所 A kind of human motion state classification method based on improvement convolutional neural networks
CN110415180A (en) * 2019-06-10 2019-11-05 西安电子科技大学 A kind of SAR image denoising method based on wavelet convolution neural network
CN110618353A (en) * 2019-10-22 2019-12-27 北京航空航天大学 Direct current arc fault detection method based on wavelet transformation + CNN
CN111091233A (en) * 2019-11-26 2020-05-01 江苏科技大学 Wind power plant short-term wind power prediction modeling method based on wavelet analysis and multi-model AdaBoost depth network
CN111507884A (en) * 2020-04-19 2020-08-07 衡阳师范学院 Self-adaptive image steganalysis method and system based on deep convolutional neural network

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于小波与最小资源分配网络的超短期风电功率预测研究;杨杰;霍志红;何永生;郭苏;邱良;许昌;;电力系统保护与控制(第09期) *
基于小波-神经网络的短期风功率预测;刘欢;张超;;电子世界(第14期) *
计及频率分析的风电场短期功率预测;林涛;赵参参;赵成林;刘航鹏;;计算机仿真(第06期) *

Also Published As

Publication number Publication date
CN112348255A (en) 2021-02-09

Similar Documents

Publication Publication Date Title
CN111012336B (en) Parallel convolutional network motor imagery electroencephalogram classification method based on spatio-temporal feature fusion
CN107977932B (en) Face image super-resolution reconstruction method based on discriminable attribute constraint generation countermeasure network
CN110232341B (en) Semi-supervised learning image identification method based on convolution-stacking noise reduction coding network
CN107316013B (en) Hyperspectral image classification method based on NSCT (non-subsampled Contourlet transform) and DCNN (data-to-neural network)
CN107633486A (en) Structure Magnetic Resonance Image Denoising based on three-dimensional full convolutional neural networks
CN112348255B (en) Ultra-short-term wind power prediction method based on wavelet time-frequency imaging
CN112233026A (en) SAR image denoising method based on multi-scale residual attention network
CN109859110B (en) Hyperspectral image panchromatic sharpening method based on spectrum dimension control convolutional neural network
CN111127316B (en) Single face image super-resolution method and system based on SNGAN network
CN108257093B (en) Single-frame image super-resolution method based on controllable kernel and Gaussian process regression
CN112700370B (en) Hyper-spectral image super-resolution method for super-parameter fidelity and depth priori combined learning
CN111812647A (en) Phase unwrapping method for interferometric synthetic aperture radar
CN111160392A (en) Hyperspectral classification method based on wavelet width learning system
CN117392153B (en) Pancreas segmentation method based on local compensation and multi-scale adaptive deformation
CN117576467B (en) Crop disease image identification method integrating frequency domain and spatial domain information
Xiong et al. Gradient boosting for single image super-resolution
CN112733589B (en) Infrared image pedestrian detection method based on deep learning
CN114820741A (en) Hyperspectral image full-waveband hyper-resolution reconstruction method
CN112634391B (en) Gray image depth reconstruction and fault diagnosis system based on compressed sensing
CN116310459B (en) Hyperspectral image subspace clustering method based on multi-view spatial spectrum combination
CN110264482B (en) Active contour segmentation method based on transformation matrix factorization of noose set
Recoskie et al. Learning filters for the 2D wavelet transform
CN114677545B (en) Lightweight image classification method based on similarity pruning and efficient module
CN115205308A (en) Fundus image blood vessel segmentation method based on linear filtering and deep learning
CN114494093A (en) Multi-modal image fusion method

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