CN113344270A - Wind resource prediction method and system based on integrated extreme learning machine - Google Patents
Wind resource prediction method and system based on integrated extreme learning machine Download PDFInfo
- Publication number
- CN113344270A CN113344270A CN202110620101.9A CN202110620101A CN113344270A CN 113344270 A CN113344270 A CN 113344270A CN 202110620101 A CN202110620101 A CN 202110620101A CN 113344270 A CN113344270 A CN 113344270A
- Authority
- CN
- China
- Prior art keywords
- extreme learning
- learning machine
- wind speed
- data
- prediction
- 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 44
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000012935 Averaging Methods 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims abstract description 8
- 239000011159 matrix material Substances 0.000 claims description 19
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 4
- 230000007246 mechanism Effects 0.000 abstract description 3
- 230000007547 defect Effects 0.000 abstract description 2
- 230000002349 favourable effect Effects 0.000 abstract 1
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 8
- 238000013473 artificial intelligence Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000001680 brushing effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- 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
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- General Engineering & Computer Science (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Development Economics (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Wind Motors (AREA)
Abstract
The invention provides a wind resource prediction method based on an integrated extreme learning machine, which comprises the following steps: initializing a system and inputting a wind speed sample to be predicted; carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set; training the data sets respectively by using an integrated extreme learning machine to obtain trained models; and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result. The invention also provides a wind resource prediction system based on the integrated extreme learning machine, which has the advantages of fast prediction by utilizing a plurality of extreme learning machines to independently predict the short-term wind speed and then carrying out weighted averaging on all prediction results, overcomes the defect of unstable prediction of single extreme learning machine, still retains the advantage of fast prediction speed of the extreme learning machine due to simple mechanism without damaging the structure of the single extreme learning machine, is favorable for the application of the extreme learning machine in practical engineering, and has high followability and prediction feasibility on the wind speed.
Description
Technical Field
The invention relates to the technical field of wind resource prediction, in particular to a wind resource prediction method and a wind resource prediction system based on an integrated extreme learning machine.
Background
Among various new energy sources, wind energy is highly emphasized and rapidly developed in recent years due to its characteristics of good inertia, wide distribution, short construction period of relevant base stations, and the like. However, the intermittency, volatility and uncertainty of wind power generation pose a serious challenge to the stable operation of a power system, and accurate offshore wind energy prediction can help a ship power system and a port microgrid to effectively utilize offshore wind resources. The offshore wind speed is the most direct influence factor of offshore wind resources, and the high-precision offshore wind speed prediction is beneficial to timely adjusting a power generation plan of a ship power system, optimizing reserve capacity and reducing the operation cost of a power grid, and has important strategic significance on the development of offshore wind power.
The prediction of offshore wind resources is still rarely researched, a complete prediction invention is not formed, and most of related prediction methods refer to a land wind power prediction method. At present, wind power prediction methods mainly include numerical weather prediction and artificial intelligence-based prediction methods. The numerical weather forecast predicts the future wind resource condition based on the weather forecast and historical weather data, but the numerical weather forecast excessively depends on the weather data, and the prediction error for medium and long periods is large. In recent years, with the rapid development of artificial intelligence technology, many attention has been paid to artificial intelligence-based wind resource prediction methods, and currently, common artificial intelligence-based prediction methods include an autoregressive model, a support vector machine, a neural network, and the like. However, the wind power prediction method based on the support vector machine and the neural network needs to be trained for many times, and parameters are more, so that the training time is longer, and higher requirements are imposed on wind resource historical data and computer computing capacity.
Through retrieval, patent document CN112100922A discloses a wind resource prediction method based on WRF and CNN convolutional neural networks, which includes the following steps: downloading GFS meteorological data; operating a WRF to simulate a target regional wind field to obtain grid meteorological data of a research region; acquiring actual measurement data of the wind speed of the anemometer tower; obtaining a WRF mesoscale operation result at the anemometer tower; performing CNN model modeling through CNN convolutional neural network and peripheral anemometer tower wind speed actual measurement; obtaining a WRF simulation wind speed result at a fan site; and summing and averaging the result of the WRF mesoscale data linear interpolation and the CNN simulated and predicted wind speed result at the fan site to obtain more accurate predicted wind speed. The wind speed prediction method based on the WRF and CNN convolutional neural networks in the prior art cannot solve the problem that rapid and violent wind speed fluctuation cannot be captured when the wind speed changes suddenly, and still has large prediction deviation.
In the document C-C and ELM rapid prediction method for short-term wind speed of a wind power plant, Su Ying et al records the prediction of the short-term wind speed of the wind power plant in a text published by a power system and an automation newspaper thereof, and provides a rapid prediction method combining a C-C method and an extreme learning machine. The literature combines various data processing and neural network methods, is beneficial to improving prediction precision and accelerating prediction speed, does not consider randomness of prediction of an extreme learning machine, does not consider influence of space-time transformation on prediction of offshore wind resources, and is not wide in application range and high in data universality.
In the literature, "short-term wind power generation prediction based on extreme learning machine", Zhu kang et al published and recorded in the electric power science and technology newspaper and proposed a prediction technology based on an extreme learning machine algorithm aiming at wind speed prediction. However, the disclosure only aims at a single extreme learning machine model, the application of an integrated extreme learning machine is not considered, and the prediction performance needs to be further improved.
Therefore, it is necessary to develop and design a wind resource prediction method capable of effectively balancing the contradiction between prediction precision and training complexity to accurately predict offshore wind resources.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a wind resource prediction method and a system based on an integrated extreme learning machine, which can effectively balance the contradiction between prediction precision and training complexity, can accurately predict offshore wind resources, is beneficial to more reasonably utilizing the offshore wind resources of ships and ports, improves the energy utilization rate, and fundamentally realizes the effects of energy conservation and emission reduction.
The invention provides a wind resource prediction method based on an integrated extreme learning machine, which comprises the following steps:
step S1: initializing a system and inputting a wind speed sample to be predicted;
step S2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set;
step S3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models;
step S4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
Preferably, the wind speed samples to be predicted in step S1 include wind speed variation samples with different variation trends.
Preferably, step S2 normalizes the prediction data using a normalization method;
normalizing the value range of the input variable to be in a [0,1] range by adopting a minimum maximization method, wherein a conversion formula is as follows:
wherein x is*Is a normalized data value; x represents wind speed sample data; x is the number ofmaxAnd xminThe maximum value and the minimum value of the original wind speed sample data are respectively.
Preferably, step S2 randomly arranges the wind speed samples to be predicted when normalizing the prediction data.
Preferably, the data set in step S3 is based on the predicted results of a plurality of individual extreme learning machines, and the predicted results are obtained by weighted averaging a plurality of results.
Preferably, in step S3, the trained model is obtained by performing weighted averaging calculation on the results after the data sets are respectively trained.
The invention provides a wind resource prediction system based on an integrated extreme learning machine, which comprises:
module M1: initializing a system and inputting a wind speed sample to be predicted;
module M2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set;
module M3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models;
module M4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
Preferably, the integrated extreme learning machine is a model which is trained and obtained by using a plurality of extreme learning machines to predict the short-term wind speed independently and then performing weighted averaging on all prediction results.
Preferably, the single extreme learning machine comprises an input layer, a hidden layer and an output layer, the input layer is in signal connection with the hidden layer, and the hidden layer is in signal connection with the output layer.
Preferably, the input layer contains grouped historical wind speed data, stored in matrix X; obtaining an output matrix of the hidden layer by the input data of the hidden layer through an activation function, wherein the input data of the hidden layer is obtained by summing a weight matrix omega and a bias matrix b on the data of the input layer; and the weight matrix beta of the output layer is obtained by calculating the target output T and the hidden layer output H of the training set.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the method, various characteristics of the wind speed can be effectively covered through the characteristic extraction of the wind speed data, and the accurate prediction of the ultra-short-term wind speed under different natural conditions is facilitated.
2. According to the invention, the prediction performance is enhanced through the prediction model of the integrated extreme learning machine, so that the prediction method is more robust, the randomness of the prediction result of a single extreme learning machine is reduced, and the provided prediction method is still effective for the wind speed with strong fluctuation.
3. In addition, the mechanism is simple, the structure of a single extreme learning machine is not damaged, the advantage of high prediction speed of the extreme learning machine is still kept, the extreme learning machine is favorably applied to actual engineering, and the extreme learning machine prediction model has high follow-up performance and prediction feasibility in wind speed.
4. The prediction method disclosed by the invention is low in data dependence, has wide applicability and good ductility, and can be applied to the prediction of offshore wind resources in different geographic environments.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a flow chart of the steps of the wind resource prediction method based on the integrated extreme learning machine according to the present invention;
FIG. 2 is a schematic diagram of an extreme learning machine deep learning neural network structure according to the present invention;
FIG. 3 is a wind speed prediction result obtained by the wind resource prediction method based on the integrated extreme learning machine according to the present invention.
Detailed Description
The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.
As shown in fig. 1, the present invention provides a wind resource prediction method based on an integrated extreme learning machine, which includes the following steps:
step S1: initializing a system and inputting a wind speed sample to be predicted; the wind speed samples to be predicted comprise wind speed change samples with different change trends. The sample data considers different variation trends of wind speed variation at the beginning of selection, and comprises data integrity and data diversity.
Step S2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set; the invention is different from the previous research, the normalized sample data is firstly sorted according to the input and output formats (a group of sample data comprises input data and prediction data), and then the sample data is randomly disordered, so as to enhance the diversity of the sample. Namely:
normalizing the value range of the input variable to be in a [0,1] range by adopting a minimum maximization method, wherein a conversion formula is as follows:
wherein x is*Is a normalized data value; x represents wind speed sample data; x is the number ofmaxAnd xminThe maximum value and the minimum value of the original wind speed sample data are respectively.
And randomly arranging the wind speed samples to be predicted when normalization processing is carried out on the predicted data. And reducing the difference between the extreme data and the normal data and weakening the influence of the extreme data on the prediction. In the method, extreme data can be understood as mutation data, for example, the wind speed is 5m/s at the last moment, and the wind speed is 7m/s at the next moment, and we consider that 7m/s is extreme data, because the wind speed is influenced by weather and time, and sometimes mutation occurs, and even if the wind speed mutates, the integrated prediction method provided by the invention can obtain a better prediction result. According to the method, the extreme data are not selected in a brushing mode, the raw data are normalized, all the data are normalized to the [0,1] interval, the extreme data in the [0,1] interval are only reflected in the condition that the numerical value is relatively large, and the influence of the extreme data on prediction is weakened in such a mode. If the normalization processing is not adopted, the prediction error is high when the wind speed suddenly changes, but the prediction error is not particularly large even if the wind speed suddenly changes after normalization.
Step S3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models; the data set is based on the prediction results of a plurality of individual extreme learning machines, and the prediction results are obtained by weighted averaging a plurality of results. And respectively carrying out weighting and averaging calculation on the results of the data sets after training to obtain the trained model.
Step S4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
The invention also provides a wind resource prediction system based on the integrated extreme learning machine, which comprises: module M1: initializing a system and inputting a wind speed sample to be predicted; module M2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set; module M3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models; module M4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
As shown in fig. 2, the deep learning neural network of the extreme learning machine mainly includes an input layer, a hidden layer and an output layer. The input layer contains the grouped historical wind speed data, which is stored in the computer in matrix form (matrix X). Specifically, the input layer data is weighted (weight matrix ω) and summed with the resulting offset (offset matrix b) to obtain the hidden layer input.
Specifically, the input layer weights and the bias matrix of the extreme learning machine are randomly generated and are not obtained by iterative calculation.
Wherein H0Representing an input layer; w is aiRepresenting the weight of the ith neuron; biRepresents the bias of the ith neuron; x is the number ofjRepresenting jth data of the sample data; the weight matrix ω is composed of wi.
And the input of the hidden layer is subjected to an activation function to obtain a hidden layer output matrix. The weight matrix beta of the output layer is calculated by the target output T and the hidden layer output H of the training set according to the following formula:
H=g(H0) (2)
β=H+·T (3)
wherein H represents the hidden layer output, H+A generalized inverse matrix representing H; t represents the prediction result, and g represents the mapping.
The parameters of the deep learning neural network of the extreme learning machine are obtained by calculation, and the traditional neural network training process is replaced, so that the method is not only the main characteristic of prediction of the extreme learning machine, but also one of the main reasons that the calculation speed is high.
According to the invention, the adopted wind speed data of a certain place in the whole year is real, and the wind speed data of the whole year has different wind speed change trends in different seasons, different months and the like.
The invention solves the problems that in the past energy prediction, the long-time scale is year, and the short-time prediction is a short prediction board of one day or hour level generally, but the invention is a minute-level wind speed prediction, namely, the wind speed of 10 minutes in the future is predicted by using the historical wind speed, and belongs to the ultra-short-term wind speed prediction.
The mechanism of the integrated extreme learning machine adopted by the invention is that the prediction results of a plurality of single extreme learning machines are weighted and averaged, so that the prediction robustness of the extreme learning machine can be enhanced, and the randomness of the prediction results of the single extreme learning machine is reduced.
The main innovation point of the invention is that extreme learning machines are integrated, most of the previous inventions adopt a single extreme learning machine to predict energy, the invention is based on the prediction results of a plurality of single extreme learning machines, and a plurality of results are subjected to mathematical operation (weighted average) to finally obtain the prediction results.
The specific embodiment is as follows:
according to the application case, 2015-year wind speed data provided by Solargis company is used, data with 10 minutes as intervals are selected to produce a plurality of samples as training sets and testing sets, wherein the proportion of the training sets to the testing sets is 70% and 30%, each sample comprises wind speed data (20 historical data values in total) of the first predicted 200 minutes, and wind speed of the last 10 minutes is predicted, namely wind speed of the last 200 minutes is predicted at sea of the last 10 minutes.
And (3) respectively initializing and training the application prediction cases by adopting 20 extreme learning machines, selecting the optimal hidden layer neuron number of the test set under respective parameters, and storing the result with the highest precision. The prediction results of the deep learning neural networks of the 20 extreme learning machines are weighted and averaged to obtain the final prediction result, as shown in fig. 3.
As can be seen from FIG. 3, the coincidence degree between the predicted value and the true value is very high, and even when the wind speed is suddenly changed, the prediction invention based on the integrated extreme learning machine can still capture rapid and severe wind speed fluctuation. Meanwhile, as can be seen from the prediction error in table 1, the prediction method provided by the invention is more effective for offshore wind resources.
TABLE 1 accuracy of predicted results based on integrated extreme learning machine
Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, units provided by the present invention as pure computer readable program code, the system and its various devices, modules, units provided by the present invention can be fully implemented by logically programming method steps in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units included in the system for realizing various functions can also be regarded as structures in the hardware component; means, modules, units for performing the various functions may also be regarded as structures within both software modules and hardware components for performing the method.
The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.
Claims (10)
1. A wind resource prediction method based on an integrated extreme learning machine is characterized by comprising the following steps:
step S1: initializing a system and inputting a wind speed sample to be predicted;
step S2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set;
step S3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models;
step S4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
2. The integrated extreme learning machine-based wind resource prediction method according to claim 1, wherein the wind speed samples to be predicted in the step S1 comprise wind speed variation samples with different variation trends.
3. The integrated extreme learning machine-based wind resource prediction method according to claim 1, wherein the step S2 normalizes the prediction data by using a normalization method;
normalizing the value range of the input variable to be in a [0,1] range by adopting a minimum maximization method, wherein a conversion formula is as follows:
wherein x is*Is a normalized data value; x represents wind speed sample data; x is the number ofmaxAnd xminThe maximum value and the minimum value of the original wind speed sample data are respectively.
4. The integrated extreme learning machine-based wind resource prediction method according to claim 3, wherein the step S2 randomly arranges the wind speed samples to be predicted when normalizing the prediction data.
5. The integrated extreme learning machine-based wind resource prediction method according to claim 1, wherein the data set in step S3 is based on prediction results of a plurality of individual extreme learning machines, and the prediction results are obtained by weighted averaging of a plurality of results.
6. The integrated extreme learning machine-based wind resource prediction method according to claim 1, wherein the trained model is obtained by performing weighted averaging calculation on the results after the data sets are respectively trained in step S3.
7. An integrated extreme learning machine-based wind resource prediction system, comprising:
module M1: initializing a system and inputting a wind speed sample to be predicted;
module M2: carrying out data characteristic extraction on a wind speed sample to be predicted to form a data set;
module M3: training the data sets respectively by using an integrated extreme learning machine to obtain trained models;
module M4: and applying the trained model to the wind speed data acquired in real time to output an integrated prediction result.
8. The integrated extreme learning machine-based wind resource prediction system of claim 7 wherein the integrated extreme learning machine is a trained model obtained by using a plurality of extreme learning machines to predict the short-term wind speed individually and then performing weighted averaging on all prediction results.
9. The integrated extreme learning machine based wind resource prediction system of claim 8 wherein the single extreme learning machine comprises an input layer, a hidden layer and an output layer, the input layer being in signal connection with the hidden layer, the hidden layer being in signal connection with the output layer.
10. The integrated extreme learning machine based wind resource prediction system of claim 9,
the input layer contains grouped historical wind speed data, and the grouped historical wind speed data is stored in a matrix X;
obtaining an output matrix of the hidden layer by the input data of the hidden layer through an activation function, wherein the input data of the hidden layer is obtained by summing a weight matrix omega and a bias matrix b on the data of the input layer;
and the weight matrix beta of the output layer is obtained by calculating the target output T and the hidden layer output H of the training set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110620101.9A CN113344270A (en) | 2021-06-03 | 2021-06-03 | Wind resource prediction method and system based on integrated extreme learning machine |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110620101.9A CN113344270A (en) | 2021-06-03 | 2021-06-03 | Wind resource prediction method and system based on integrated extreme learning machine |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113344270A true CN113344270A (en) | 2021-09-03 |
Family
ID=77473475
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110620101.9A Pending CN113344270A (en) | 2021-06-03 | 2021-06-03 | Wind resource prediction method and system based on integrated extreme learning machine |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113344270A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016101182A1 (en) * | 2014-12-23 | 2016-06-30 | 清华大学 | Interval type indicator forecasting method based on bayesian network and extreme learning machine |
US20170161606A1 (en) * | 2015-12-06 | 2017-06-08 | Beijing University Of Technology | Clustering method based on iterations of neural networks |
CN109272156A (en) * | 2018-09-12 | 2019-01-25 | 河海大学 | A kind of super short-period wind power probability forecasting method |
CN111898828A (en) * | 2020-03-09 | 2020-11-06 | 华能四川水电有限公司 | Hydroelectric power generation prediction method based on extreme learning machine |
CN112329987A (en) * | 2020-10-15 | 2021-02-05 | 南京理工大学 | Short-term wind power plant power prediction method based on Adaboost-EMD-SVM |
-
2021
- 2021-06-03 CN CN202110620101.9A patent/CN113344270A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016101182A1 (en) * | 2014-12-23 | 2016-06-30 | 清华大学 | Interval type indicator forecasting method based on bayesian network and extreme learning machine |
US20170161606A1 (en) * | 2015-12-06 | 2017-06-08 | Beijing University Of Technology | Clustering method based on iterations of neural networks |
CN109272156A (en) * | 2018-09-12 | 2019-01-25 | 河海大学 | A kind of super short-period wind power probability forecasting method |
CN111898828A (en) * | 2020-03-09 | 2020-11-06 | 华能四川水电有限公司 | Hydroelectric power generation prediction method based on extreme learning machine |
CN112329987A (en) * | 2020-10-15 | 2021-02-05 | 南京理工大学 | Short-term wind power plant power prediction method based on Adaboost-EMD-SVM |
Non-Patent Citations (1)
Title |
---|
朱抗 等: ""基于极限学习机的短期风力发电预测"", 《电力科学与技术学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wu et al. | A data mining approach combining $ K $-means clustering with bagging neural network for short-term wind power forecasting | |
CN112348292B (en) | Short-term wind power prediction method and system based on deep learning network | |
CN112686464A (en) | Short-term wind power prediction method and device | |
Wang et al. | Wind speed forecasting based on multi-objective grey wolf optimisation algorithm, weighted information criterion, and wind energy conversion system: A case study in Eastern China | |
CN103268366A (en) | Combined wind power prediction method suitable for distributed wind power plant | |
Ye et al. | Combined approach for short-term wind power forecasting based on wave division and Seq2Seq model using deep learning | |
Wu et al. | Stratification-based wind power forecasting in a high-penetration wind power system using a hybrid model | |
CN110555515A (en) | Short-term wind speed prediction method based on EEMD and LSTM | |
CN102880907A (en) | Method and device for correcting wind speed | |
CN110866633A (en) | Micro-grid ultra-short term load prediction method based on SVR support vector regression | |
Pan et al. | Probabilistic Short‐Term Wind Power Forecasting Using Sparse Bayesian Learning and NWP | |
CN113095547B (en) | Short-term wind power prediction method based on GRA-LSTM-ICE model | |
CN113435595A (en) | Two-stage optimization method for extreme learning machine network parameters based on natural evolution strategy | |
CN112307672A (en) | BP neural network short-term wind power prediction method based on cuckoo algorithm optimization | |
CN116681154A (en) | Photovoltaic power calculation method based on EMD-AO-DELM | |
CN116484743A (en) | Evaluation method and device for influence of wind power grid connection on reliability of power system | |
CN112734073A (en) | Photovoltaic power generation short-term prediction method based on long and short-term memory network | |
CN116307139A (en) | Wind power ultra-short-term prediction method for optimizing and improving extreme learning machine | |
CN113344270A (en) | Wind resource prediction method and system based on integrated extreme learning machine | |
CN114004405A (en) | Photovoltaic power prediction method and system based on Elman neural network and satellite cloud picture | |
Kehe et al. | Research of wind power prediction model based on RBF neural network | |
Al-Haija et al. | Time-series model for forecasting short-term future additions of renewable energy to worldwide capacity | |
Cheng et al. | A hybrid probabilistic estimation method for photovoltaic power generation forecasting | |
Ding et al. | Photovoltaic array power prediction model based on EEMD and PSO-KELM | |
CN113962357A (en) | GWO-WNN-based distributed photovoltaic power data virtual acquisition 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 |