CN113095547B - Short-term wind power prediction method based on GRA-LSTM-ICE model - Google Patents
Short-term wind power prediction method based on GRA-LSTM-ICE model Download PDFInfo
- Publication number
- CN113095547B CN113095547B CN202110288897.2A CN202110288897A CN113095547B CN 113095547 B CN113095547 B CN 113095547B CN 202110288897 A CN202110288897 A CN 202110288897A CN 113095547 B CN113095547 B CN 113095547B
- Authority
- CN
- China
- Prior art keywords
- wind power
- short
- prediction
- term
- variable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000013528 artificial neural network Methods 0.000 claims abstract description 11
- 238000012937 correction Methods 0.000 claims abstract description 7
- 238000005516 engineering process Methods 0.000 claims abstract description 7
- 238000010219 correlation analysis Methods 0.000 claims abstract description 3
- 238000011156 evaluation Methods 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000014509 gene expression Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000013179 statistical model Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
- H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/40—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation wherein a plurality of decentralised, dispersed or local energy generation technologies are operated simultaneously
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/76—Power conversion electric or electronic aspects
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Human Resources & Organizations (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Computational Mathematics (AREA)
- Power Engineering (AREA)
- Pure & Applied Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Optimization (AREA)
- Operations Research (AREA)
- Marketing (AREA)
- Mathematical Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
Abstract
The invention discloses a short-term wind power prediction method based on GRA-LSTM-ICE, which comprises the following steps of: firstly, gray correlation analysis (GRA) is carried out on meteorological data collected in a wind farm to obtain meteorological variables with higher correlation with wind power; secondly, aiming at the problem of low short-term prediction precision of the current wind power, a wind power prediction model is built based on a long-short term memory (LSTM) neural network, and the short-term prediction precision of the wind power is improved; finally, aiming at the obtained wind power prediction result, a wind power prediction correction model is established based on an information credibility assessment (ICE) technology, and the wind power short-term prediction precision is further improved. By the implementation of the method, the prediction accuracy of short-term wind power can be greatly improved, and the economic and stable operation of the wind power grid with high proportion can be realized.
Description
Technical Field
The invention relates to a wind power prediction technology in the technical field of wind power generation, in particular to a short-term wind power prediction method based on a GRA-LSTM-ICE model.
Background
With the increase of the popularity of wind power generation, in order to ensure timely and reliable power supply, it becomes important to accurately predict the future wind power integration amount in the power system. Due to inaccurate predictions, unbalance in the power system may increase drastically and may even lead to problems of frequency stability etc. Therefore, improving the short-term prediction accuracy of wind power is of great importance for the stable operation of modern power systems.
In recent years, there are physical modeling methods and statistical modeling methods as methods commonly used for wind power prediction. The physical model uses atmospheric parameters such as wind speed and direction, temperature and pressure, physical characteristics such as terrain robustness index and wind farm layout, and numerical weather forecast as inputs to the complex meteorological model to predict future parameters. Statistical models are purely mathematical models, using mainly past observations, sometimes aided by numerical weather forecast information. For statistical models, machine learning methods are widely used, with Artificial Neural Networks (ANNs) being one of the most common techniques for short-term prediction. However, the current method cannot effectively analyze the nonlinear correlation between the multidimensional weather variable and the wind power variable, and cannot effectively and timely predict the wind power in a short term with high precision over the whole range according to weather forecast.
Disclosure of Invention
Because in the current wind power short-term prediction method, nonlinear correlation between multidimensional weather variables and wind power variables cannot be effectively analyzed, and the wind power output cannot be effectively and accurately predicted in time according to weather forecast. Thus, the present invention has for its research the following objectives: and the nonlinear correlation between the multidimensional meteorological variable and the wind power variable is accurately expressed, and a high-precision short-term wind power predicted value is obtained.
The invention adopts the technical proposal for solving the technical problems in the prior art that: a short-term wind power prediction method based on a GRA-LSTM-ICE model, the method comprising the steps of:
step 1: based on a gray correlation analysis (GRA) method, respectively calculating the correlation between wind power variable and wind speed, wind direction, temperature, relative humidity, air pressure, rainfall, snowfall and cloud layer thickness variable, and taking a meteorological variable with a correlation coefficient larger than 0.5 as an input variable of a prediction model;
step 2: according to the calculation result of the relevance coefficient, taking a meteorological variable with the relevance coefficient larger than 0.5 as an input variable of a prediction model, taking a wind power variable as an output variable of the model, establishing the prediction model based on a long-term short-term memory (LSTM) neural network, and carrying out preliminary prediction on the output of the fan;
step 3: and according to the preliminary short-term wind power predicted value, a wind power prediction correction model is established based on an information reliability assessment (ICE) technology, the preliminary predicted value is corrected, a large-range high-precision wind power short-term predicted result is obtained, and the short-term wind power prediction precision is further improved.
Step 4: the prediction results were evaluated based on Mean Absolute Error (MAE) and Root Mean Square Error (RMSE).
Further, in the step 1, the correlation between the wind speed, wind direction, temperature, relative humidity, air pressure, rainfall, snowfall, cloud layer thickness variable and wind power variable is calculated respectively based on the GRA method, and the method comprises the following steps:
step 1.1: normalizing meteorological variables provided by numerical weather forecast and wind power variables collected by a wind farm, and eliminating physical dimensions among different variables;
step 1.2: setting the multidimensional weather variable as X i The wind power variable is set as Y;
step 1.3: calculating Y and X i Gray correlation coefficient ζ of (2) i ；
Step 1.4: calculating Y and X i Gray correlation coefficient R of (2) i 。
Further, in the step 2, a prediction model is built based on the LSTM neural network, and the fan output is primarily predicted, including the following steps:
step 2.1: calculating the number of hidden layers of the LSTM network and the number of hidden layer neurons;
step 2.2: calculating initial weights and bias items of the LSTM network;
step 2.3: training a network, and updating weights and bias items;
step 2.4: establishing a wind power prediction model, and calculating a wind power predicted value P meeting the precision requirement T 。
Further, in the step 3, a wind power prediction correction model is established based on the ICE technology, and a short-term wind power preliminary predicted value is corrected to obtain a large-range high-precision wind power short-term predicted result, which comprises the following steps:
step 3.1: calculating a credible evaluation value A of weather data at the moment T T Trusted evaluation value B of stand-alone wind power prediction T ；
Step 3.2: calculating a trusted value C of single machine wind power prediction at time T T ；
Step 3.3: calculating a confidence value P of a single machine short-term wind power prediction C ；
Step 3.4: calculating high-precision predicted value P of large-range short-term wind power N 。
Further, the gray correlation coefficient R in step 1.4 i The expression is：
Wherein: r is R i Is the gray correlation coefficient of the ith meteorological variable; y is Y t The value of the wind power variable at the time t; x is X it Is the value of the ith meteorological variable at the time t; y is Y m Is the average value of the wind power variation; x is X im Is the average of the ith meteorological variable.
Further, the LSTM neural network in step 2.1 includes 1 input layer, 3 hidden layers, 1 output layer, and 1 connection layer.
Further, the trusted evaluation value A in step 3.1 T And B T The expression of (2) is:
wherein: a is that T Is the credible evaluation value of the weather data at the moment T; b (B) T Is the credible evaluation value of the single-machine wind power prediction; t (T) T Is the actual value at time T; p (P) T Is the predicted value at time T.
Further, the confidence value C of the single machine wind power prediction in the step 3.2 T The expression of (2) is:
C T ＝A T +B T
wherein: c (C) T Is the credible value of the single machine wind power prediction at the moment T.
Further, the trusted value P of the stand-alone short-term wind power prediction in step 3.3 C The expression of (2) is:
wherein: p (P) C Is the credible value of the single machine short-term wind power prediction; m is the entire short-term prediction interval.
Further, stepHigh-precision prediction value P of the large-range short-term wind power in step 3.4 N The expression of (2) is:
wherein: p (P) N Is a high-precision predicted value of large-range short-term wind power; n is the total number of fans for the entire range.
Further, the expressions of MAE and RMSE in step 4 are:
wherein: p (P) Ni Is a predicted value of wind power; t (T) Ni Is the actual value of the wind power.
The invention has the advantages and beneficial effects that: step 1, a GRA-based method obtains a correlation coefficient of a wind power variable and a meteorological variable, avoids the influence of factors with low correlation on modeling, and improves the prediction accuracy of wind power; step 2, a wind power prediction model considering multidimensional meteorological factors is established based on an LSTM neural network method, the problem that the accuracy of the existing model is low is solved, and the short-term prediction accuracy of wind power is improved; and 3, based on the ICE method, a wind power prediction correction model is established, and the short-term prediction accuracy of the large-range wind power is improved.
Aiming at the problem of low short-term prediction precision of the current wind power, the invention establishes a wind power prediction model based on the LSTM neural network, thereby improving the short-term prediction precision of the wind power; aiming at the obtained wind power prediction result, a wind power prediction correction model is established based on the ICE technology, so that the short-term wind power prediction precision is further improved, and the short-term wind power prediction precision can be greatly improved through the implementation of the method, and the economic and stable operation of a power grid containing high-proportion wind power can be realized; the short-term wind power prediction method based on the GRA-LSTM-ICE model reduces prediction errors, improves overall prediction accuracy, and effectively verifies the effectiveness, accuracy and applicability of the method.
Drawings
FIG. 1 is a technical roadmap of a short-term wind power prediction method based on GRA-LSTM-ICE model of the invention.
FIG. 2 is a diagram of a prediction model structure in a short-term wind power prediction method based on GRA-LSTM-ICE model.
FIG. 3 is a diagram of a network hidden layer neuron structure in a short-term wind power prediction method based on a GRA-LSTM-ICE model.
FIG. 4 is a graph of short-term prediction contrast of large-scale wind power based on GRA-LSTM-ICE model of the present invention.
Detailed Description
In order to further describe the technical method adopted by the invention, the short-term prediction method for improving the wind power according to the invention is described in detail below with reference to the accompanying drawings and the embodiments.
In this embodiment, a method for improving short-term wind power prediction accuracy based on a GRA-LSTM-ICE model is provided, as shown in fig. 1, and the method for improving wind power prediction accuracy includes the following steps:
step 1: based on the GRA method, respectively calculating the correlation between the wind power variable and the wind speed, wind direction, temperature, relative humidity, air pressure, rainfall, snowfall and cloud layer thickness variable, and taking the meteorological variable with the correlation coefficient larger than 0.5 as an input variable of a prediction model;
step 1.1: normalizing meteorological variables provided by numerical weather forecast and wind power variables collected by a wind farm, and eliminating physical dimensions among different variables;
step 1.2: setting the multidimensional weather variable as X i The wind power variable is set as Y;
step 1.3: calculating Y and X i Gray correlation coefficient ζ of (2) i ；
Step 1.4: calculating Y and X i Gray correlation coefficient R of (2) i 。
Step 2: according to the calculation result of the relevance coefficient, taking a meteorological variable with the relevance coefficient larger than 0.5 as an input variable of a prediction model, taking a wind power variable as an output variable of the model, establishing the prediction model based on an LSTM neural network, and carrying out preliminary prediction on the output of the fan as shown in fig. 2;
step 2.1: calculating the number of hidden layers of the LSTM network and the number of hidden layer neurons, wherein the hidden layer neurons are structured as shown in figure 3;
step 2.2: calculating initial weights and bias items of the LSTM network;
step 2.3: training a network, and updating weights and bias items;
step 2.4: establishing a wind power prediction model, and calculating a wind power predicted value P meeting the precision requirement T 。
Step 3: and according to the preliminary short-term wind power predicted value, a wind power prediction correction model is established based on the ICE technology, the preliminary predicted value is corrected, a large-range high-precision wind power short-term predicted result is obtained, and the short-term wind power predicted precision is further improved.
Step 3.1: calculating a credible evaluation value A of weather data at the moment T T Trusted evaluation value B of stand-alone wind power prediction T ；
Step 3.2: calculating a trusted value C of single machine wind power prediction at time T T ；
Step 3.3: calculating a confidence value P of a single machine short-term wind power prediction C ；
Step 3.4: calculating high-precision predicted value P of large-range short-term wind power N 。
Step 4: the prediction results were evaluated using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE).
Examples
In order to verify the effectiveness, accuracy and applicability of the short-term wind power prediction method provided by the invention, the method is applied to a Hua Dianda forest stand wind power plant, the wind power of a large-range area of the wind power plant is subjected to daily prediction research, and SVM and LSTM algorithms are selected for comparison analysis. The multi-dimensional meteorological variable and wind power variable of 2019 and 9 are selected as training and testing data of the model, the time resolution is 15 minutes, and 20 fans are included in the selected range.
According to the gray correlation coefficient R between the multidimensional meteorological variable and the wind power variable obtained in the step 1 i As shown in table 1:
table 1: gray correlation coefficient of multidimensional meteorological variable and wind power variable
According to grey correlation coefficient R i And (3) selecting the wind speed, wind direction, temperature and humidity variables as input variables of the prediction model. The short-term prediction comparison result of the large-range wind power based on the GRA-LSTM-ICE model obtained according to the steps 2 and 3 is shown in fig. 4, and the comparison evaluation index is shown in table 2.
Table 2: large-range wind power short-term prediction contrast evaluation index based on GRA-LSTM-ICE model
According to the graph, MAE and RMSE evaluation indexes of the large-range wind power short-term prediction method based on the GRA-LSTM-ICE model are reduced by 10 percent and 14.72 percent respectively compared with the SVM model, and are reduced by 2.55 percent and 3.2 percent respectively compared with the LSTM model. The short-term wind power prediction method based on the GRA-LSTM-ICE model reduces prediction errors, improves overall prediction accuracy, and effectively verifies the effectiveness, accuracy and applicability of the method.
Claims (2)
1. A short-term wind power prediction method based on a GRA-LSTM-ICE model, the method comprising the steps of:
step 1: based on a gray correlation analysis method, respectively calculating the correlation between wind power variable and wind speed, wind direction, temperature, relative humidity, air pressure, rainfall, snowfall and cloud layer thickness variable, and taking a meteorological variable with a correlation coefficient larger than 0.5 as an input variable of a prediction model;
step 1.1: normalizing meteorological variables provided by numerical weather forecast and wind power variables collected by a wind farm, and eliminating physical dimensions among different variables;
step 1.2: setting the multidimensional weather variable as X i The wind power variable is set as Y;
step 1.3: calculating Y and X i Gray correlation coefficient ζ of (2) i ；
Step 1.4: calculating Y and X i Gray correlation coefficient R of (2) i The formula is as follows:
wherein: r is R i Is the gray correlation coefficient of the ith meteorological variable; y is Y t The value of the wind power variable at the time t; x is X it Is the value of the ith meteorological variable at the time t; y is Y m Is the average value of the wind power variation; x is X im Is the average value of the ith meteorological variable;
step 2: according to the calculation result of the relevance coefficient, taking a meteorological variable with the relevance coefficient larger than 0.5 as an input variable of a prediction model, taking a wind power variable as an output variable of the model, establishing the prediction model based on a long-and-short-term memory neural network, and carrying out preliminary prediction on the output of the fan;
step 2.1: calculating the number of hidden layers of the LSTM network and the number of hidden layer neurons;
step 2.2: calculating initial weights and bias items of the LSTM network;
step 2.3: training a network, and updating weights and bias items;
and 2, step 2.4: establishing a wind power prediction model, and calculating a wind power predicted value P meeting the precision requirement T ；
Step 3: according to the preliminary short-term wind power predicted value, a wind power prediction correction model is established based on an information reliability assessment (ICE) technology, the preliminary predicted value is corrected, a large-range high-precision wind power short-term predicted result is obtained, and the short-term wind power predicted precision is further improved;
step 3.1: calculating a credible evaluation value A of weather data at the moment T T Trusted evaluation value B of stand-alone wind power prediction T The formula is as follows:
wherein: a is that T Is the credible evaluation value of the weather data at the moment T; b (B) T Is the credible evaluation value of the single-machine wind power prediction; t (T) T Is the actual value at time T; p (P) T Is a predicted value at the time T;
step 3.2: calculating a trusted value C of single machine wind power prediction at time T T The formula is as follows:
C T ＝A T +B T
wherein: c (C) T Is the credible value of the single machine wind power prediction at the moment T;
step 3.3: calculating a confidence value P of a single machine short-term wind power prediction C The formula is as follows:
wherein: p (P) C Is the credible value of the single machine short-term wind power prediction; m is the entire short-term prediction interval;
step 3.4: calculating high-precision predicted value P of large-range short-term wind power N The formula is as follows:
wherein: p (P) N Is a high-precision predicted value of large-range short-term wind power; n is the total number of fans in the whole range;
step 4: the prediction result is evaluated based on the mean absolute error and the root mean square error.
2. The method for predicting short-term wind power based on GRA-LSTM-ICE model of claim 1, wherein the LSTM neural network of step 2 includes 1 input layer, 3 hidden layers, 1 output layer and 1 connection layer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110288897.2A CN113095547B (en) | 2021-03-18 | 2021-03-18 | Short-term wind power prediction method based on GRA-LSTM-ICE model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110288897.2A CN113095547B (en) | 2021-03-18 | 2021-03-18 | Short-term wind power prediction method based on GRA-LSTM-ICE model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113095547A CN113095547A (en) | 2021-07-09 |
| CN113095547B true CN113095547B (en) | 2023-09-26 |
Family
ID=76668431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110288897.2A Active CN113095547B (en) | 2021-03-18 | 2021-03-18 | Short-term wind power prediction method based on GRA-LSTM-ICE model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113095547B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113537598A (en) * | 2021-07-17 | 2021-10-22 | 沈阳工程学院 | Short-term light power prediction method based on NWP-LSTM |
| CN113673768A (en) * | 2021-08-24 | 2021-11-19 | 河海大学 | Wind power probability prediction model based on VMD and CNN-LSTM |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679298A (en) * | 2013-12-27 | 2014-03-26 | 中能电力科技开发有限公司 | Method for evaluating wind power plant short-period power prediction precision |
| CN104657786A (en) * | 2015-02-03 | 2015-05-27 | 河海大学 | Short-term wind power mixed predicting method based on Boosting algorithm |
| CN107292453A (en) * | 2017-07-24 | 2017-10-24 | 国网江苏省电力公司电力科学研究院 | A kind of short-term wind power prediction method based on integrated empirical mode decomposition Yu depth belief network |
| 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 |
| CN109359388A (en) * | 2018-10-18 | 2019-02-19 | 北京仿真中心 | A kind of Complex simulation systems credibility evaluation method |
-
2021
- 2021-03-18 CN CN202110288897.2A patent/CN113095547B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679298A (en) * | 2013-12-27 | 2014-03-26 | 中能电力科技开发有限公司 | Method for evaluating wind power plant short-period power prediction precision |
| CN104657786A (en) * | 2015-02-03 | 2015-05-27 | 河海大学 | Short-term wind power mixed predicting method based on Boosting algorithm |
| CN107292453A (en) * | 2017-07-24 | 2017-10-24 | 国网江苏省电力公司电力科学研究院 | A kind of short-term wind power prediction method based on integrated empirical mode decomposition Yu depth belief network |
| 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 |
| CN109359388A (en) * | 2018-10-18 | 2019-02-19 | 北京仿真中心 | A kind of Complex simulation systems credibility evaluation method |
Non-Patent Citations (2)
| Title |
|---|
| 基于Bagging神经网络集成的风功率预测;梁涛;石欢;崔洁;李宗琪;;水电能源科学(第04期);全文 * |
| 基于灰色理论的风电场输出功率超短期预测;高僮;何平;周莽;孟芳园;;河南科技(第19期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113095547A (en) | 2021-07-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108304623B (en) | Probability load flow online calculation method based on stack noise reduction automatic encoder | |
| CN108898251B (en) | Offshore wind farm power prediction method considering meteorological similarity and power fluctuation | |
| CN112288164B (en) | Wind power combined prediction method considering spatial correlation and correcting numerical weather forecast | |
| CN113095547B (en) | Short-term wind power prediction method based on GRA-LSTM-ICE model | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN114462718A (en) | CNN-GRU wind power prediction method based on time sliding window | |
| CN113887801A (en) | Building cold load prediction method, system, equipment and readable storage medium | |
| CN114298140A (en) | Wind power short-term power prediction correction method considering unit classification | |
| CN116526473A (en) | Particle swarm optimization LSTM-based electrothermal load prediction method | |
| Chen et al. | Research on wind power prediction method based on convolutional neural network and genetic algorithm | |
| CN116341613A (en) | Ultra-short-term photovoltaic power prediction method based on Informar encoder and LSTM | |
| CN115186923A (en) | Photovoltaic power generation power prediction method and device and electronic equipment | |
| CN113991711B (en) | Capacity configuration method for energy storage system of photovoltaic power station | |
| CN116702937A (en) | Photovoltaic output day-ahead prediction method based on K-means mean value clustering and BP neural network optimization | |
| CN117526274A (en) | New energy power prediction method, electronic equipment and storage medium in extreme climate | |
| CN114897204A (en) | Method and device for predicting short-term wind speed of offshore wind farm | |
| CN110276478B (en) | Short-term wind power prediction method based on segmented ant colony algorithm optimization SVM | |
| CN115912334A (en) | Method for establishing prediction model of output guarantee rate of wind power plant and prediction method | |
| CN115271242A (en) | Training method, prediction method and device of photovoltaic power generation power prediction model | |
| CN112036672B (en) | New energy power generation ultra-short term power prediction method and system based on iterative correction | |
| CN113723670A (en) | Photovoltaic power generation power short-term prediction method with variable time window | |
| CN113379151A (en) | Wind speed ultra-short term prediction method based on Bagging-CNN-GRU | |
| CN110543724A (en) | Satellite structure performance prediction method for overall design | |
| Singh et al. | An optimized neural network model for relative humidity prediction | |
| CN115860247A (en) | Method and device for training fan loss power prediction model in extreme weather |
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 |