CN113111592A - Short-term wind power prediction method based on EMD-LSTM - Google Patents
Short-term wind power prediction method based on EMD-LSTM Download PDFInfo
- Publication number
- CN113111592A CN113111592A CN202110491290.4A CN202110491290A CN113111592A CN 113111592 A CN113111592 A CN 113111592A CN 202110491290 A CN202110491290 A CN 202110491290A CN 113111592 A CN113111592 A CN 113111592A
- Authority
- CN
- China
- Prior art keywords
- wind power
- historical
- data
- processed
- short
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 27
- 238000010606 normalization Methods 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 22
- 230000006403 short-term memory Effects 0.000 claims abstract description 14
- 230000007787 long-term memory Effects 0.000 claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 6
- 230000006641 stabilisation Effects 0.000 claims abstract description 6
- 238000011105 stabilization Methods 0.000 claims abstract description 6
- 210000002569 neuron Anatomy 0.000 claims description 21
- 239000013598 vector Substances 0.000 claims description 19
- 230000006870 function Effects 0.000 claims description 12
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000012549 training Methods 0.000 claims description 8
- 230000015654 memory Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 6
- 238000009472 formulation Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims 1
- 238000010248 power generation Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000005070 sampling 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
- 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
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Geometry (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Wind Motors (AREA)
Abstract
The application provides a short-term wind power prediction method based on EMD-LSTM, which comprises the steps of obtaining historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data; carrying out normalization processing on the historical data to obtain historical data to be processed; decomposing historical data to be processed according to empirical mode decomposition, and carrying out stabilization processing to obtain a plurality of groups of subsequences; performing correlation screening on each group of subsequence to screen out n +1 groups of subsequence to be processed; predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed; and performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result. The method for predicting the wind power of the wind power plant has the advantages that parameters needing to be adjusted are few, meanwhile, the problem of long-time nonlinear sequence prediction can be solved, the wind power can be timely and accurately predicted, and therefore the wind power plant can be accurately scheduled and operated.
Description
Technical Field
The application relates to the technical field of wind power prediction, in particular to a short-term wind power prediction method based on EMD-LSTM.
Background
Renewable energy sources such as wind power, photovoltaic power generation and the like are rapidly developed in recent years, and more renewable energy sources are connected to a power grid, so that new challenges are brought to the power grid.
Wind power is a renewable energy source with intermittence and volatility, certain adverse effects are brought to grid connection, scheduling and the like due to the characteristic, and the wind power generation prediction technology becomes an effective mode for relieving the adverse effects and can perform day-ahead or real-time scheduling according to the prediction data of wind power generation. However, the prediction precision of the traditional wind power generation prediction method is low, and the wind power cannot be accurately predicted, so that the scheduling of the wind power plant is not accurate according to the prediction value.
Disclosure of Invention
The application provides a short-term wind power prediction method based on EMD-LSTM, which aims to solve the technical problem.
In order to achieve the above purpose, the embodiments of the present application adopt the following technical solutions:
the short-term wind power prediction method based on the EMD-LSTM is provided and comprises the following steps:
acquiring historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data;
carrying out normalization processing on the historical data to obtain historical data to be processed;
decomposing historical data to be processed according to empirical mode decomposition, and carrying out stabilization processing to obtain a plurality of groups of subsequences;
performing correlation screening on each group of subsequence to screen out n +1 groups of subsequence to be processed;
predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed;
and performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result.
Further, the historical data is normalized, and the normalization formula is as follows:
wherein y and y*Respectively representing data before and after normalization; y ismaxAnd yminRespectively representing the maximum and minimum values before normalization.
Further, the historical wind power data includes historical maximum wind power data, historical minimum wind power data, and historical average wind power data.
Further, decomposing the historical data to be processed according to empirical mode decomposition includes:
determining local maximum value points and minimum value points of various feature vectors according to the historical average wind power data;
constructing an upper envelope line a (t) and a lower envelope line b (t) of each feature vector by a cubic spline curve function;
determining the average value of each feature vector through the upper envelope line and the lower envelope line
Calculating an original historical average wind power sequence y (t), and making a difference between the original historical average wind power sequence y (t) and an average value c (t) to obtain a difference value d (t) of the feature vector, y (t) -c (t);
if the difference d (t) of the feature vector meets the component condition of empirical mode decomposition, the difference d (t) of the feature vector is the maximum frequency component l of the original historical average wind power sequence y (t)1(t)；
For the original historical average wind power sequence y (t) and the maximum frequency component li(t) differencing to obtain a residual component sequence ri(t)(i＝1，2，…，n)；
If the remaining component sequence riAnd (t) ending the empirical mode decomposition process when the (t) is a monotonic function or a constant.
Further, the predicting the to-be-processed subsequence and the to-be-processed historical data through a long-short term memory network model, wherein the optimizing the long-short term memory network model comprises:
the number M of the neurons between the input layer and the output layer is determined by the characteristics of the training set data;
selecting the number M of the neurons under the condition of optimal neuron number formula and evaluation index, wherein the neuron number formula is as follows:
in the formula, n and m are the node numbers of the output layer and the input layer respectively, and a is a constant between [0 and 10 ];
gradually increasing the number of network layers according to the number M of the neurons to test the evaluation indexes of the model and the average relative error to obtain the network layers;
and obtaining an optimized long-term and short-term memory network model according to the network layer.
The application provides a short-term wind power prediction method based on EMD-LSTM, which is used for acquiring historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data; carrying out normalization processing on the historical data to obtain historical data to be processed; decomposing historical data to be processed according to empirical mode decomposition, and carrying out stabilization processing to obtain a plurality of groups of subsequences; performing correlation screening on each group of subsequence to screen out n +1 groups of subsequence to be processed; predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed; and performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result. The method for predicting the wind power of the wind power plant has the advantages that parameters needing to be adjusted are few, meanwhile, the problem of long-time nonlinear sequence prediction can be solved, the wind power can be timely and accurately predicted, and therefore the wind power plant can be accurately scheduled and operated.
Drawings
In order to more clearly illustrate the embodiments of the present application 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 application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart of a short-term wind power prediction method based on EMD-LSTM according to the present application;
FIG. 2 is a prediction flow chart of another short-term wind power prediction method based on EMD-LSTM in the present application;
FIG. 3 is a schematic empirical mode decomposition of the EMD-LSTM predictive model of the present application;
FIG. 4 is a comparison graph of the wind power prediction result and the actual power result of the EMD-LSTM model of the present application.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of this application and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In this embodiment, the related terms are described as follows:
empirical Mode Decomposition (EMD) method
Cyclic Neural Network prediction model, RNN for short, Current Neural Network
Recurrent neural network prediction model based on empirical mode decomposition (EMD-RNN for short)
Support vector machine prediction model, SVM for short
Support vector machine prediction model based on empirical mode decomposition (EMD-SVM for short)
Long Short Term Memory network prediction model, LSTM for Short, Long Short-Term Memory
Empirical mode decomposition-based long-short term memory network prediction model, EMD-LSTM for short
The present application is described in further detail below with reference to the attached drawing figures:
the embodiment of the application provides a short-term wind power prediction method based on EMD-LSTM, as shown in fig. 1 and 2, the method comprises the following steps:
step S101, acquiring historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data; the historical wind power data comprises historical maximum wind power data, historical minimum wind power data and historical average wind power data.
Step S102, normalization processing is carried out on the historical data to obtain historical data to be processed; and carrying out normalization processing on the historical data, wherein a normalization formula is as follows:
wherein y and y*Respectively representing data before and after normalization; y ismaxAnd yminRespectively representing the maximum and minimum values before normalization.
S103, decomposing the historical data to be processed according to empirical mode decomposition and carrying out stabilization processing to obtain a plurality of groups of subsequences; performing empirical mode decomposition on the normalized historical average wind power data with non-stationarity, obtaining each subsequence IMF1-IMFn and residual component RES, and decomposing the historical data to be processed according to the empirical mode decomposition, wherein the method comprises the following steps:
determining local maximum value points and minimum value points of various feature vectors according to the historical average wind power data;
constructing an upper envelope line a (t) and a lower envelope line b (t) of each feature vector by a cubic spline curve function;
determining the average value of each feature vector through the upper envelope line and the lower envelope line
Calculating an original historical average wind power sequence y (t), and making a difference between the original historical average wind power sequence y (t) and an average value c (t) to obtain a difference value d (t) of the feature vector, y (t) -c (t);
and if the difference d (t) of the feature vectors does not meet the IMF component condition of the empirical mode decomposition, taking the difference as the original historical average wind power sequence, and repeating the steps until the IMF component condition is met, wherein the obtained d (t) is one of the IMF components of the original historical average wind power sequence y (t).
If the difference d (t) of the feature vector meets the IMF component condition of empirical mode decomposition, the difference d (t) of the feature vector is the maximum frequency component l of the original historical average wind power sequence y (t)1(t)；
For the original historical average wind power sequence y (t) and the maximum frequency component li(t) differencing to obtain a residual component sequence ri(t)(i＝1，2，…，n)；
The residual component sequence ri(t) as the original sequence, repeating the above steps continuously, and judging ri(t) (i ═ 1, 2, …, n) whether or not a termination condition is satisfied;
if the remaining component sequence riAnd (t) ending the empirical mode decomposition process when the (t) is a monotonic function or a constant.
Meanwhile, decomposing the original concentration sequence y (t) to obtain n IMF components and a residual component rn(t), as shown in the following equation:
residual component rnAnd (t) is the average trend of the original historical average wind power sequence y (t), and IMF components from 1 to n are respectively reflected by different characteristic scale signal components from high to low in frequency.
S104, performing correlation screening on each group of subsequences to screen out n +1 groups of subsequences to be processed;
s105, predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed; wherein, optimizing the long-term and short-term memory network model comprises:
the number M of the neurons between the input layer and the output layer is determined by the characteristics of the training set data;
selecting the number M of the neurons under the condition of optimal neuron number formula and evaluation index, wherein the neuron number formula is as follows:
in the formula, n and m are the node numbers of the output layer and the input layer respectively, and a is a constant between [0 and 10 ];
gradually increasing the number of network layers according to the number M of the neurons to test the evaluation indexes of the model and the average relative error to obtain the network layers;
and obtaining an optimized long-term and short-term memory network model according to the network layer.
And S106, performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result.
The long-short term memory network model LSTM constructed in step S105 is optimized to ensure the average relative percentage error yMAPERoot mean square error yRMSEMinimum, prediction accuracy yFAOn the premise of the highest, determining the optimal parameters of a long-term and short-term memory network prediction model LSTM; the method comprises the following specific steps: the method is characterized in that a ReLU function is used as an activation function in a hidden layer and is suitable for time series nonlinear prediction, the initial learning rate is set to be 0.001, the rejection rate of each layer of network nodes is 0.2, overfitting is prevented, the iteration frequency is 100, the number of neurons between an input layer and an output layer is determined by the characteristics of training set data, the number of the selected neurons is 15 under the condition that the neuron number formula and the evaluation index are optimal, the number of the neurons is controlled to be 15 under the condition that the number of the neurons is not changed, the number of network layers is gradually increased to test a model, and finally, the network layer is selected to be 3 in combination with the evaluation index of average relative.
And counting the output predicted value, performing reverse normalization processing on the predicted value output by all the long-term and short-term memory network models, and superposing all the data subjected to reverse normalization processing to obtain a final predicted result.
After step S106, the error analysis of the obtained final prediction result and the short-term wind power actual value is used for evaluating the model prediction performance, and the evaluation standard is adoptedTwo indexes of average relative percentage error and root mean square error are used. Wherein the average relative percentage error yMAPERoot mean square error yRMSEThe expression of (a) is as follows:
wherein n represents the total number of predictions; xact(i) And Xpred(i) The real value and the predicted value of the wind power at the moment i are respectively, wherein the smaller the average relative percentage error and the root mean square error are, the better the prediction result is represented.
In order to verify the effectiveness of the short-term wind power prediction method based on the EMD-LSTM combination, 20-day numerical weather forecast data and historical wind power data (historical maximum wind power data, historical minimum wind power data and historical average wind power data) of a certain wind power plant generator set are selected as input data of an EMD-LSTM model, a wind power predicted value is used as output, modeling analysis is carried out, and the data sampling time interval is 15 min. And selecting 1824 data in the first 19 days as a training set for training the EMD-LSTM model, and selecting 96 data in the next day as a test set for the model to perform tests to predict the wind power 24h in the future.
In the EMD-LSTM method, the EMD signal processing technology is utilized to decompose historical average wind power data to decompose a series of subsequences with different characteristic frequency scales, and the decomposition result is shown in figure 3.
The wind power sub-sequence prediction model decomposed by the empirical mode can be obtained by training the long-short term memory network prediction model, the prediction results of the sub-sequence components are further superposed, the wind power prediction value of two days in the future can be obtained, the prediction error evaluation index of the EMD-LSTM prediction method is shown in the table 1, and the prediction result is shown in the graph 4.
And comparing and researching the EMD-LSTM prediction model with the RNN model and the SVM model respectively to compare and verify the effectiveness of the EMD-LSTM prediction model. The neuron of the RNN model takes a value of 10, the activation function adopts a relu function, the learning rate is 0.01, and the training times are 200. The kernel function of the SVM model selects a radial basis kernel function (RBF), and the penalty factor is 1.0. The performance pair ratios of the prediction models are shown in table 1.
TABLE 1 wind power prediction error evaluation index/% under different models
| Model (model) | Average relative percentage error | Root mean square error |
| EMD-LSTM | 0.751 | 4.351 |
| EMD-RNN | 1.237 | 5.628 |
| EMD-SVM | 1.982 | 7.519 |
| LSTM | 2.416 | 8.359 |
| RNN | 3.526 | 9.153 |
| SVM | 3.843 | 10.124 |
From table 1, it can be known that the average relative percentage error and root mean square error of the LSTM prediction model are lower than those of the RNN model and the SVM model, and thus the LSTM prediction model has certain advantages. Meanwhile, compared with the EMD-LSTM, EMD-RNN and EMD-SVM models which are not subjected to empirical mode decomposition, the average relative percentage error and the root mean square error of the EMD-LSTM, EMD-RNN and EMD-SVM models which are subjected to empirical mode decomposition are reduced, so that the empirical mode decomposition method can effectively decompose the fluctuating wind power sequence into relatively stable subsequences, and has high prediction reliability.
The short-term wind power prediction method of the EMD-LSTM solves the problem of mutual influence in decomposition results in the prior art, can accurately predict the change of wind power, provides a basis for the operation condition of a wind power plant, and provides a reference for providing an operation plan for dispatchers.
The application provides a short-term wind power prediction method based on EMD-LSTM, which is used for acquiring historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data; carrying out normalization processing on the historical data to obtain historical data to be processed; decomposing historical data to be processed according to empirical mode decomposition, and carrying out stabilization processing to obtain a plurality of groups of subsequences; performing correlation screening on each group of subsequence to screen out n +1 groups of subsequence to be processed; predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed; and performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result. The method for predicting the wind power of the wind power plant has the advantages that parameters needing to be adjusted are few, meanwhile, the problem of long-time nonlinear sequence prediction can be solved, the wind power can be timely and accurately predicted, and therefore the wind power plant can be accurately scheduled and operated.
The above-mentioned contents are only for explaining the technical idea of the present application, and the protection scope of the present application is not limited thereby, and any modification made on the basis of the technical idea presented in the present application falls within the protection scope of the claims of the present application.
Additionally, the order in which elements and sequences of the processes described herein are processed, the use of alphanumeric characters, or the use of other designations, is not intended to limit the order of the processes and methods described herein, unless explicitly claimed. While various presently contemplated embodiments have been discussed in the foregoing disclosure by way of example, it should be understood that such detail is solely for that purpose and that the appended claims are not limited to the disclosed embodiments, but, on the contrary, are intended to cover all modifications and equivalent arrangements that are within the spirit and scope of the embodiments herein. For example, although the system components described above may be implemented by hardware devices, they may also be implemented by software-only solutions, such as installing the described system on an existing server or mobile device.
Similarly, it should be noted that in the preceding description of embodiments of the application, various features are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the embodiments. This method of disclosure, however, is not intended to require more features than are expressly recited in the claims. Indeed, the embodiments may be characterized as having less than all of the features of a single embodiment disclosed above.
The entire contents of each patent, patent application publication, and other material cited in this application, such as articles, books, specifications, publications, documents, and the like, are hereby incorporated by reference into this application. Except where the application is filed in a manner inconsistent or contrary to the present disclosure, and except where the claim is filed in its broadest scope (whether present or later appended to the application) as well. It is noted that the descriptions, definitions and/or use of terms in this application shall control if they are inconsistent or contrary to the statements and/or uses of the present application in the material attached to this application.
Claims (5)
1. A short-term wind power prediction method based on EMD-LSTM is characterized by comprising the following steps:
acquiring historical data of a wind field, wherein the historical data comprises weather forecast data and historical wind power data;
carrying out normalization processing on the historical data to obtain historical data to be processed;
decomposing historical data to be processed according to empirical mode decomposition, and carrying out stabilization processing to obtain a plurality of groups of subsequences;
performing correlation screening on each group of subsequence to screen out n +1 groups of subsequence to be processed;
predicting the subsequence to be processed and the historical data to be processed through a long-term and short-term memory network model to obtain n +1 predicted values, wherein the predicted values are predicted values of the subsequence to be processed;
and performing inverse normalization processing on the n +1 predicted values, and overlapping to obtain a final predicted result.
2. The short-term wind power prediction method based on EMD-LSTM according to claim 1, characterized in that the historical data is normalized by the formula:
wherein y and y*Respectively representing data before and after normalization; y ismaxAnd yminRespectively representing the maximum and minimum values before normalization.
3. The EMD-LSTM based short-term wind power prediction method of claim 2, wherein the historical wind power data comprises historical maximum wind power data, historical minimum wind power data, and historical average wind power data.
4. The EMD-LSTM-based short-term wind power prediction method of claim 3, wherein decomposing historical data to be processed according to empirical mode decomposition comprises:
determining local maximum value points and minimum value points of various feature vectors according to the historical average wind power data;
constructing an upper envelope line a (t) and a lower envelope line b (t) of each feature vector by a cubic spline curve function;
determining the average value of each feature vector through the upper envelope line and the lower envelope line
Calculating an original historical average wind power sequence y (t), and making a difference between the original historical average wind power sequence y (t) and an average value c (t) to obtain a difference value d (t) of the feature vector, y (t) -c (t);
if the difference d (t) of the feature vector meets the component condition of empirical mode decomposition, the difference d (t) of the feature vector is the maximum frequency component l of the original historical average wind power sequence y (t)1(t)；
For the original historical average wind power sequence y (t) and the maximum frequency component li(t) differencing to obtain a residual component sequence ri(t)(i＝1，2，…，n)；
If the remaining component sequence riAnd (t) ending the empirical mode decomposition process when the (t) is a monotonic function or a constant.
5. The EMD-LSTM based short-term wind power prediction method of claim 1, wherein the to-be-processed sub-sequences and the to-be-processed historical data are predicted by a long-short term memory network model, wherein the optimization of the long-short term memory network model comprises:
the number M of the neurons between the input layer and the output layer is determined by the characteristics of the training set data;
selecting the number M of the neurons under the condition of optimal neuron number formula and evaluation index, wherein the neuron number formula is as follows:
in the formula, n and m are the node numbers of the output layer and the input layer respectively, and a is a constant between [0 and 10 ];
gradually increasing the number of network layers according to the number M of the neurons to test the evaluation indexes of the model and the average relative error to obtain the network layers;
and obtaining an optimized long-term and short-term memory network model according to the network layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110491290.4A CN113111592A (en) | 2021-05-06 | 2021-05-06 | Short-term wind power prediction method based on EMD-LSTM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110491290.4A CN113111592A (en) | 2021-05-06 | 2021-05-06 | Short-term wind power prediction method based on EMD-LSTM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113111592A true CN113111592A (en) | 2021-07-13 |
Family
ID=76721350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110491290.4A Pending CN113111592A (en) | 2021-05-06 | 2021-05-06 | Short-term wind power prediction method based on EMD-LSTM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113111592A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115514439A (en) * | 2022-09-26 | 2022-12-23 | 华工未来科技(江苏)有限公司 | Channel air interface utilization rate prediction method, system, electronic equipment and medium |
| CN117013534A (en) * | 2023-09-26 | 2023-11-07 | 宁德时代新能源科技股份有限公司 | Power generation power prediction method, power prediction model training method, device and equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102951A (en) * | 2014-05-05 | 2014-10-15 | 南方电网科学研究院有限责任公司 | Short-term wind power prediction method based on EMD (Empirical Mode Decomposition) historical data preprocessing |
| CN109214566A (en) * | 2018-08-30 | 2019-01-15 | 华北水利水电大学 | Short-term wind power prediction method based on shot and long term memory network |
| CN110363360A (en) * | 2019-07-24 | 2019-10-22 | 广东工业大学 | A kind of short-term wind power forecast method, device and equipment |
| CN110458316A (en) * | 2019-05-21 | 2019-11-15 | 武汉大学 | A kind of offshore wind farm power short term prediction method based on set empirical mode decomposition and LSTM network |
| CN112488415A (en) * | 2020-12-14 | 2021-03-12 | 国网江苏省电力有限公司经济技术研究院 | Power load prediction method based on empirical mode decomposition and long-and-short-term memory network |
| CN112686464A (en) * | 2021-01-07 | 2021-04-20 | 云南电力技术有限责任公司 | Short-term wind power prediction method and device |
-
2021
- 2021-05-06 CN CN202110491290.4A patent/CN113111592A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104102951A (en) * | 2014-05-05 | 2014-10-15 | 南方电网科学研究院有限责任公司 | Short-term wind power prediction method based on EMD (Empirical Mode Decomposition) historical data preprocessing |
| CN109214566A (en) * | 2018-08-30 | 2019-01-15 | 华北水利水电大学 | Short-term wind power prediction method based on shot and long term memory network |
| CN110458316A (en) * | 2019-05-21 | 2019-11-15 | 武汉大学 | A kind of offshore wind farm power short term prediction method based on set empirical mode decomposition and LSTM network |
| CN110363360A (en) * | 2019-07-24 | 2019-10-22 | 广东工业大学 | A kind of short-term wind power forecast method, device and equipment |
| CN112488415A (en) * | 2020-12-14 | 2021-03-12 | 国网江苏省电力有限公司经济技术研究院 | Power load prediction method based on empirical mode decomposition and long-and-short-term memory network |
| CN112686464A (en) * | 2021-01-07 | 2021-04-20 | 云南电力技术有限责任公司 | Short-term wind power prediction method and device |
Non-Patent Citations (4)
| Title |
|---|
| 朱玥 等: "基于EMD-LSTM 的光伏发电预测模型", 《电力工程技术》 * |
| 赵倩;黄景涛;: "基于EMD-SA-SVR的超短期风电功率预测研究", 电力***保护与控制 * |
| 韩朋;张晓琳;张飞;王永平;: "基于AM-LSTM模型的超短期风电功率预测", 科学技术与工程 * |
| 魏骜;茅大钧;韩万里;吕彬;: "基于EMD和长短期记忆网络的短期电力负荷预测研究", 热能动力工程 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115514439A (en) * | 2022-09-26 | 2022-12-23 | 华工未来科技(江苏)有限公司 | Channel air interface utilization rate prediction method, system, electronic equipment and medium |
| CN117013534A (en) * | 2023-09-26 | 2023-11-07 | 宁德时代新能源科技股份有限公司 | Power generation power prediction method, power prediction model training method, device and equipment |
| CN117013534B (en) * | 2023-09-26 | 2024-02-20 | 宁德时代新能源科技股份有限公司 | Power generation power prediction method, power prediction model training method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Raza et al. | An ensemble framework for day-ahead forecast of PV output power in smart grids | |
| Alencar et al. | Hybrid approach combining SARIMA and neural networks for multi-step ahead wind speed forecasting in Brazil | |
| CN111260030B (en) | A-TCN-based power load prediction method and device, computer equipment and storage medium | |
| CN103117546B (en) | A kind of Ultrashort-term slide prediction method for wind power | |
| CN111860982A (en) | Wind power plant short-term wind power prediction method based on VMD-FCM-GRU | |
| CN111027775A (en) | Step hydropower station generating capacity prediction method based on long-term and short-term memory network | |
| CN110443417A (en) | Multiple-model integration load forecasting method based on wavelet transformation | |
| CN110956312A (en) | Photovoltaic power distribution network voltage prediction method based on EMD-CNN deep neural network | |
| CN113111592A (en) | Short-term wind power prediction method based on EMD-LSTM | |
| CN111242355A (en) | Photovoltaic probability prediction method and system based on Bayesian neural network | |
| CN116307291B (en) | Distributed photovoltaic power generation prediction method and prediction terminal based on wavelet decomposition | |
| CN114492922A (en) | Medium-and-long-term power generation capacity prediction method | |
| Li et al. | GMM-HMM-based medium-and long-term multi-wind farm correlated power output time series generation method | |
| CN115688993A (en) | Short-term power load prediction method suitable for power distribution station area | |
| CN111697560B (en) | Method and system for predicting load of power system based on LSTM | |
| CN115860797B (en) | Electric quantity demand prediction method suitable for new electricity price reform situation | |
| CN115907131A (en) | Method and system for building electric heating load prediction model in northern area | |
| Hossain | Application of Gaussian mixture regression model for short-term wind speed forecasting | |
| CN112734073A (en) | Photovoltaic power generation short-term prediction method based on long and short-term memory network | |
| CN115660893A (en) | Transformer substation bus load prediction method based on load characteristics | |
| Bantupalli et al. | Wind Speed forecasting using empirical mode decomposition with ANN and ARIMA models | |
| Paulin et al. | SOLAR PHOTOVOLTAIC OUTPUT POWER FORECASTING USING BACK PROPAGATION NEURAL NETWORK. | |
| CN113191526A (en) | Short-term wind speed interval multi-objective optimization prediction method and system based on random sensitivity | |
| Yang et al. | A hybrid vmd-based ARIMA-LSTM model for day-ahead pv prediction and uncertainty analysis | |
| Katranji et al. | Short-Term Wind Speed Prediction for Saudi Arabia via 1D-CNN |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210713 |