CN107194507A - A kind of short-term wind speed forecasting method of wind farm based on combination SVMs - Google Patents
A kind of short-term wind speed forecasting method of wind farm based on combination SVMs Download PDFInfo
- Publication number
- CN107194507A CN107194507A CN201710347542.XA CN201710347542A CN107194507A CN 107194507 A CN107194507 A CN 107194507A CN 201710347542 A CN201710347542 A CN 201710347542A CN 107194507 A CN107194507 A CN 107194507A
- Authority
- CN
- China
- Prior art keywords
- wind
- wind speed
- svms
- model
- 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
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 32
- 238000013277 forecasting method Methods 0.000 title claims abstract description 21
- 230000006870 function Effects 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000013480 data collection Methods 0.000 claims abstract description 9
- 230000000739 chaotic effect Effects 0.000 claims abstract description 8
- 238000004364 calculation method Methods 0.000 claims abstract description 7
- 239000002245 particle Substances 0.000 claims abstract description 5
- 238000005457 optimization Methods 0.000 claims abstract description 4
- 238000005070 sampling Methods 0.000 claims description 7
- 238000005259 measurement Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 abstract description 6
- 238000000034 method Methods 0.000 description 13
- 230000005611 electricity Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000007935 neutral effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000854350 Enicospilus group Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Quality & Reliability (AREA)
- Evolutionary Computation (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Operations Research (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
It is a kind of that the chaotic characteristic of original wind series is analyzed based on the short-term wind speed forecasting method of wind farm for combining SVMs, with the Embedded dimensions m and delay time T of C C algorithm calculation of wind speed sequences, build inputoutput data collection and be divided into training set and checking collection；The SVMs individual event forecast model based on different kernel functions is set up on training set, and key parameter is determined with integrated learning strategy particle swarm optimization algorithm；The weight coefficient of each Single model is determined using the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator on checking collection, is predicted respectively, to the weighted sum that predicts the outcome, obtains a step forecasting wind speed result.More rational mode input vector is determined, it is ensured that the otherness in combined prediction between each individual event forecast model；Combined weight number improves combined forecasting precision according to Single model precision of prediction adaptive change.Be particularly suitable for use in predicting wind speed of wind farm.
Description
Technical field
The present invention relates to a kind of short-term wind speed forecasting method of wind farm based on combination SVMs, belong to wind power plant wind
Fast prediction field.
Background technology
Wind-powered electricity generation obtained extensive development in China in recent years as a kind of renewable and clean energy resource, meanwhile, wind-power electricity generation
Randomness, the intermittent and fluctuation being had bring potential safety hazard to the stabilization of power network and economical operation.Accurate wind-powered electricity generation work(
Rate prediction can provide important evidence for power scheduling, effectively mitigate influence of the wind-powered electricity generation to power network.Due to wind power and wind speed
There is the relation directly determined, wind power prediction can be realized on the basis of forecasting wind speed, so accurately wind farm wind velocity is pre-
Survey particularly significant.But due to the non-stationary and randomness of wind speed itself so that existing technological means was to its minute rank
Short-term forecast precision is not high.The information that combined prediction can be included using different Single models, is reduced and scattered individual event is pre-
The approximation and uncertainty of model are surveyed, the overall predicated error of reduction, is the study hotspot of short-term wind speed forecasting.
Retrieve and find through the open source literature to prior art, in short-term wind speed combination forecasting method, to each Single model
Selection focus primarily upon BP neural network, RBF neural, wavelet neural network etc., but neutral net is based on Local Search,
Local optimum is easily trapped into, over-fitting occurs；Generalization ability is limited.The combined method predicted the outcome to Single model is main
Weigthed sums approach or nonlinear weight method are used, these method weights are fixed, it is impossible to adapt to wind speed change, make forecasting wind speed
Precision is limited.It can be seen that, in the combination forecasting method of the short-term wind speed of wind power plant, in the selection of Single model and combination all
There is improved space, further to improve precision of prediction.
The content of the invention
It is an object of the invention to provide a kind of step clearly, the short-term wind speed combined prediction sides of wind power plant of reliable results
The problem of short-term wind speed forecasting precision is not high caused by method, the randomness and uneven stability of solution wind farm wind velocity sequence.
A kind of short-term wind speed forecasting method of wind farm based on combination SVMs, is sensed by anemobiagraph or wind speed
Device equipment surveys the air speed data of wind power plant, obtains original wind series；The chaotic characteristic of original wind series is analyzed, C-C is used
The Embedded dimensions m and delay time T of algorithm calculation of wind speed sequence, the inputoutput data collection of forecast model is built according to m and τ,
Data set is divided into training set and checking collects；The SVMs individual event based on different kernel functions is set up on the training set pre-
Model is surveyed, and determines with integrated learning strategy particle swarm optimization algorithm the key parameter of each supporting vector machine model；Tested described
The power of each Single model is determined on card collection using the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator
Coefficient, is carried out in advance to the wind speed of prediction time respectively with the SVMs individual event forecast model based on different kernel functions
Survey, predict the outcome weighted sum to each Single model, obtains a step forecasting wind speed result.
Preferably, after new sampling instant arrives, the wind speed number for the actual measurement wind power plant that the checking collection basis is newly sampled
Updated according to rolling, the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator is used on the checking collection
The weight coefficient of each Single model is determined, with the SVMs individual event forecast model based on different kernel functions respectively to prediction
The wind speed at moment is predicted, and predict the outcome weighted sum to each Single model, combined prediction subsequent time air speed value.
Preferably, after new sampling instant arrives, wind power plant is surveyed by anemobiagraph or air velocity transducer equipment
Air speed data, the wind series after being updated；The chaotic characteristic of wind series after analysis renewal, wind is calculated with C-C algorithms
The Embedded dimensions m and delay time T of fast sequence, the inputoutput data collection of forecast model are built according to m and τ, by the checking
Collection is rolled according to the air speed data for the actual measurement wind power plant newly sampled to be updated.
Preferably, the Embedded dimensions m and delay time T of the use C-C algorithms calculation of wind speed sequence, is to carry out phase space
Reconstruct, and the inputoutput data collection of each Single model is determined according to Embedded dimensions m and delay time T, data set is divided into instruction
Practice collection and checking collection.
After technology proposed by the present invention is taken, the wind power plant based on combination SVMs according to embodiments of the present invention
Short-term wind speed forecasting method, the supporting vector machine model based on different kernel functions makes each list as the Single model of combined prediction
Item model is while possessing stronger generalization ability, additionally it is possible to ensure the otherness between Single model；Checking collection, which is followed, to be adopted
The sample moment, which rolls, to be updated, and the weight coefficient of each Single model is verifying the precision of prediction adaptive change on collecting according to each model,
Combining weights are provided for the forecasting wind speed of prediction time, short-term wind speed combined forecasting precision is further improved.
After technology proposed by the present invention is taken, the wind power plant based on combination SVMs according to embodiments of the present invention
Short-term wind speed forecasting method, SVMs of the selection with different kernel functions is based on knot as Single model, SVMs
Structure risk minimization criterion, compared with neutral net, possesses outstanding generalization ability and global optimizing ability, meanwhile, it is different
Kernel function can ensure the otherness between each Single model；Combination forecasting method, which is used, is based on induced order weighted harmonic mean
The variable weight coefficient combination of operator, different prediction time each Single model weight coefficients are different, to adapt to the change of wind series.This
The Forecasting Methodology of invention further increases short-term wind speed forecasting precision, with practicality.
After technology proposed by the present invention is taken, the wind power plant based on combination SVMs according to embodiments of the present invention
Short-term wind speed forecasting method, according to the chaotic characteristic phase space reconstruction of wind series, it is determined that more rational mode input to
Amount；Single model selects the supporting vector machine model based on different kernel functions, it is ensured that each individual event forecast model in combined prediction
Between otherness；Combined weight number improves combined forecasting precision according to Single model precision of prediction adaptive change.This
Invention be applied to predicting wind speed of wind farm field, with precision of prediction it is high the characteristics of.
Brief description of the drawings
Fig. 1 is a kind of short-term wind speed forecasting method of wind farm flow chart based on combination SVMs；
Fig. 2 is individual event SVMs forecasting wind speed illustraton of model；
Fig. 3 is prediction of wind speed and the comparison figure of actual wind speed.
Embodiment
Below with reference to accompanying drawings to the present invention each be preferred embodiment described.There is provided referring to the drawings
Description, to help the understanding of the example embodiment of the invention to being limited by appended claims and their equivalents.It includes side
Assistant solution various details, but they can only be counted as it is exemplary.Therefore, it would be recognized by those skilled in the art that
Embodiment described herein can be made various changes and modifications, without departing from scope and spirit of the present invention.Moreover, in order to
Make specification more clear succinct, will omit pair it is well known that function and the detailed description of construction.
As shown in figure 1, a kind of short-term wind speed forecasting method of wind farm based on combination SVMs, specific steps are such as
Under：
1. surveying the air speed data of wind power plant by anemobiagraph or air velocity transducer equipment, obtaining actual wind speed sequence is
10 minutes of 31 days sampling air speed datas, totally 4464 data.First 30 days totally 4320 sample points are selected, using the present invention's
Method sets up combination forecasting；Selecting the data of the 31st day, totally 144 sample points are as forecast set, to examine precision of prediction.
2. the chaotic characteristic of the original wind series of analysis, during with the Embedded dimensions m of C-C algorithm calculation of wind speed sequences with delay
Between τ, the inputoutput data collection of forecast model is built according to m and τ, and inputoutput data collection is divided into training set and checking
Collection.Step is as follows：
2.1 couples of wind series X={ x1,x2,...x4320Its Embedded dimensions m=4 is calculated with C-C algorithms, during delay
Between τ=43.
2.2 decimally calculate Liapunov exponent λ=0.3877 according to amount method>0, it is known that wind series have chaos special
Property.
2.3 Embedded dimensions drawn according to step 2.1 and time delay carry out chaos phase space weight to wind speed time series
Structure, the input vector for obtaining SVMs is
XI (i)=[x (i), x (i+43), x (i+86), (i+129)] i=1 ..., 4190
It is output as
Y (i)=x (i+130) i=1 ..., 4190
2.4 determine that training set is { XI (i), Y (i) } i=1,2 ..., 3470, and checking collection is { XI (i), Y (i) } i=
3471,3472,...,4190。
3. on the training set that step 2 is drawn, the SVMs based on different kernel functions is set up as individual event and predicts mould
Type to ensure the otherness between Single model, to the parameter of each model with the optimizing of integrated learning strategy particle cluster algorithm choose with
Ensure the accuracy of each Single model.Set up respectively in the present embodiment and be based on gaussian radial basis function, exponential type radial direction base core
Function, Polynomial kernel function, the supporting vector machine model of multi-layer perception (MLP) kernel function and line style kernel function, individual event SVMs
Model is as shown in Fig. 2, and to the parameter integrated learning strategy particle cluster algorithm optimizing of each model.
4. on checking collection, determined using the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator
The weight coefficient of each Single model, and checking collection and weight coefficient follow the wind-speed sample moment to roll renewal.Finally to prediction time
Each Single model predict the outcome weighted sum, calculate prediction time air speed value and obtain forecasting wind speed result.Specific steps
It is as follows：
The 4.1 five supporting vector machine models N=720 samplings common to checking collection upper 3471~4190 set up with step 3
Air speed data on point predicts that predicted value is designated as x respectivelyit, represent the predicted value of i-th kind of Forecasting Methodology t, i=1,
2 ..., 5, t=1,2 ..., N；Each model precision of prediction on each future position is calculated, p is designated asit, represent i-th kind of prediction side
The precision of prediction of method t, i=1,2 ..., 5, t=1,2 ..., N；If L=(l1,l2,l3,l4,l5)TIt is pre- for five kinds of individual events
The weight coefficient in combined prediction is surveyed, it meets regression nature and nonnegativity.
4.2 precision of prediction pitRegard predicted value x asitInduction value, five kinds of individual event Forecasting Methodology ts are predicted
Precision sequence p1t,p2t,p3t,p4t,p5tBy order arrangement from big to small, if p-index (it) is the i-th big precision of prediction
Subscript, it is induced order weighted harmonic mean operator, the i.e. group as the t produced by precision of prediction sequence to define IOWHA
Close predicted value, such as following formula.It can be seen that, the weight coefficient of combined prediction and each individual event Forecasting Methodology at a time on precision of prediction it is big
It is small closely related.
The 4.3 combined prediction reciprocal errors quadratic sum F for making N phases on training set total are minimum, and the optimization for solving following formula is asked
Topic, so that it may obtain each weighted value L.
4.4 calculate the wind speed value of subsequent time with each Single model respectively, are designated as xi, it is calculated as follows and draws
The combined prediction value of subsequent time
4.5 after the actual wind speed at last samples moment is obtained, and is determined further according to the N number of sampled point nearest from future position
New checking collection, calculates according to step 4.1~4.4 and obtains new weight, then carry out the combined prediction of next prediction time.Can
See, the weight of each future position is, according to newest wind-speed sample data adaptive change, to imply newest wind speed information, be conducive to
Improve precision of prediction.
4.6 predictions obtain the 31st day 144 data points, are compared with actual wind speed sampled value, are predicted as shown in Figure 3
The comparison figure of wind speed and actual wind speed.
Short-term wind speed forecasting method of wind farm based on combination SVMs according to embodiments of the present invention, based on difference
The supporting vector machine model of kernel function makes each Single model possess stronger generalization ability as the Single model of combined prediction
While, additionally it is possible to ensure the otherness between Single model；The checking collection following sampling moment, which rolls, to be updated, each Single model
Precision of prediction adaptive change of the weight coefficient according to each model on checking collection, combination is provided for the forecasting wind speed of prediction time
Weight, makes short-term wind speed combined forecasting precision further improve.
Short-term wind speed forecasting method of wind farm based on combination SVMs according to embodiments of the present invention, selection has
The SVMs of different kernel functions is as Single model, and SVMs is based on empirical risk minimization, with nerve net
Network is compared, and possesses outstanding generalization ability and global optimizing ability, meanwhile, different kernel functions can ensure each Single model it
Between otherness；Combination forecasting method is different pre- using the variable weight coefficient combination based on induced order weighted harmonic mean operator
Moment each Single model weight coefficient is surveyed different, to adapt to the change of wind series.The Forecasting Methodology of the present invention is further improved
Short-term wind speed forecasting precision, with practicality.
Short-term wind speed forecasting method of wind farm based on combination SVMs according to embodiments of the present invention, according to wind speed
The chaotic characteristic phase space reconstruction of sequence, it is determined that more rational mode input vector；Single model selection is based on different IPs
The supporting vector machine model of function, it is ensured that the otherness in combined prediction between each individual event forecast model；Combined weight number
According to Single model precision of prediction adaptive change, combined forecasting precision is improved.The present invention is applied to predicting wind speed of wind farm
Field, with precision of prediction it is high the characteristics of.
The present invention is described in detail above, principle and embodiment party of the specific case used herein to the present invention
Formula is set forth, and the explanation of above example is only intended to the method and its core concept for helping to understand the present invention；Meanwhile, it is right
In those of ordinary skill in the art, according to the thought of the present invention, change is had in specific embodiments and applications
Part, in summary, this specification content should not be construed as limiting the invention.
Through the above description of the embodiments, those skilled in the art can be understood that the present invention can be real
Apply.Certainly, above listed situation is merely illustrative, and the present invention is not limited to this.It should be appreciated by those skilled in the art root
According to other deformations or simplified of technical solution of the present invention, the present invention can be suitably applied to, and this hair should be included in
In bright scope.
Claims (4)
1. a kind of short-term wind speed forecasting method of wind farm based on combination SVMs, it is characterised in that
The air speed data of wind power plant is surveyed by anemobiagraph or air velocity transducer equipment, original wind series are obtained；
The chaotic characteristic of original wind series is analyzed, with the Embedded dimensions m and delay time T of C-C algorithm calculation of wind speed sequences, according to
The inputoutput data collection of forecast model is built according to m and τ, data set is divided into training set and checking collects；
The SVMs individual event forecast model based on different kernel functions is set up on the training set, and uses integrated learning strategy
Particle swarm optimization algorithm determines the key parameter of each supporting vector machine model；
Determined on the checking collection using the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator each
The weight coefficient of Single model, with the SVMs individual event forecast model based on different kernel functions respectively to prediction time
Wind speed is predicted, and predict the outcome weighted sum to each Single model, obtains a step forecasting wind speed result.
2. a kind of short-term wind speed forecasting method of wind farm based on combination SVMs according to claim 1, it is special
Levy and be,
After new sampling instant arrives, the checking collection is rolled according to the air speed data for the actual measurement wind power plant newly sampled to be updated；
Determined on the checking collection using the variable weight coefficient combinatorial forecast based on induced order weighted harmonic mean operator each
The weight coefficient of Single model, with the SVMs individual event forecast model based on different kernel functions respectively to prediction time
Wind speed is predicted, and predict the outcome weighted sum to each Single model, combined prediction subsequent time air speed value.
3. a kind of short-term wind speed forecasting method of wind farm based on combination SVMs according to claim 2, it is special
Levy and be,
After new sampling instant arrives, the air speed data of wind power plant is surveyed by anemobiagraph or air velocity transducer equipment, is obtained
Wind series after must updating；
The chaotic characteristic of wind series after analysis renewal, Embedded dimensions m and time delay with C-C algorithm calculation of wind speed sequences
τ, the inputoutput data collection of forecast model is built according to m and τ, by wind of the checking collection according to the actual measurement wind power plant newly sampled
Fast data scrolling updates.
4. a kind of short-term wind speed forecasting method of wind farm based on combination SVMs according to claim 1, it is special
Levy and be, the Embedded dimensions m and delay time T of the use C-C algorithms calculation of wind speed sequence are to carry out phase space reconfiguration, and
The inputoutput data collection of each Single model is determined according to Embedded dimensions m and delay time T, data set is divided into training set and tested
Card collection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710347542.XA CN107194507A (en) | 2017-05-17 | 2017-05-17 | A kind of short-term wind speed forecasting method of wind farm based on combination SVMs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710347542.XA CN107194507A (en) | 2017-05-17 | 2017-05-17 | A kind of short-term wind speed forecasting method of wind farm based on combination SVMs |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107194507A true CN107194507A (en) | 2017-09-22 |
Family
ID=59873927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710347542.XA Pending CN107194507A (en) | 2017-05-17 | 2017-05-17 | A kind of short-term wind speed forecasting method of wind farm based on combination SVMs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194507A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108930632A (en) * | 2017-05-25 | 2018-12-04 | 株式会社日立制作所 | Adaptability electric generation management |
| CN110659672A (en) * | 2019-09-02 | 2020-01-07 | 国电新能源技术研究院有限公司 | Wind turbine generator output step uncertainty prediction method and device |
| CN111753368A (en) * | 2020-05-18 | 2020-10-09 | 重庆长安汽车股份有限公司 | Method for predicting sound absorption performance in vehicle |
| CN111931981A (en) * | 2020-07-06 | 2020-11-13 | 安徽天尚清洁能源科技有限公司 | Photovoltaic power generation ultra-short-term prediction method based on machine learning multi-model combination |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663513A (en) * | 2012-03-13 | 2012-09-12 | 华北电力大学 | Combination forecast modeling method of wind farm power by using gray correlation analysis |
| CN103065202A (en) * | 2012-12-24 | 2013-04-24 | 电子科技大学 | Wind power plant ultrashort term wind speed prediction method based on combination kernel function |
| CN103400052A (en) * | 2013-08-22 | 2013-11-20 | 武汉大学 | Combined method for predicting short-term wind speed in wind power plant |
| CN103903067A (en) * | 2014-04-09 | 2014-07-02 | 上海电机学院 | Short-term combination forecasting method for wind power |
| CN104899432A (en) * | 2015-05-19 | 2015-09-09 | 上海大学 | Kernel function combination-based PSO-LSSVM fluctuating wind speed prediction method |
| CN105303250A (en) * | 2015-09-23 | 2016-02-03 | 国家电网公司 | Wind power combination prediction method based on optimal weight coefficient |
| CN105809293A (en) * | 2016-03-29 | 2016-07-27 | 国网青海省电力公司 | Multi-model combined prediction method for short-term power of wind farm |
| CN106126906A (en) * | 2016-06-22 | 2016-11-16 | 重庆科技学院 | Short-term wind speed forecasting method based on C C Yu ELM |
-
2017
- 2017-05-17 CN CN201710347542.XA patent/CN107194507A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663513A (en) * | 2012-03-13 | 2012-09-12 | 华北电力大学 | Combination forecast modeling method of wind farm power by using gray correlation analysis |
| CN103065202A (en) * | 2012-12-24 | 2013-04-24 | 电子科技大学 | Wind power plant ultrashort term wind speed prediction method based on combination kernel function |
| CN103400052A (en) * | 2013-08-22 | 2013-11-20 | 武汉大学 | Combined method for predicting short-term wind speed in wind power plant |
| CN103903067A (en) * | 2014-04-09 | 2014-07-02 | 上海电机学院 | Short-term combination forecasting method for wind power |
| CN104899432A (en) * | 2015-05-19 | 2015-09-09 | 上海大学 | Kernel function combination-based PSO-LSSVM fluctuating wind speed prediction method |
| CN105303250A (en) * | 2015-09-23 | 2016-02-03 | 国家电网公司 | Wind power combination prediction method based on optimal weight coefficient |
| CN105809293A (en) * | 2016-03-29 | 2016-07-27 | 国网青海省电力公司 | Multi-model combined prediction method for short-term power of wind farm |
| CN106126906A (en) * | 2016-06-22 | 2016-11-16 | 重庆科技学院 | Short-term wind speed forecasting method based on C C Yu ELM |
Non-Patent Citations (4)
| Title |
|---|
| 唐小我: "组合预测计算方法研究", 《预测》 * |
| 张亚军: "变权系数组合预测模型在电力系统负荷预测中的应用研究", 《中国优秀硕士学位论文全文数据库》 * |
| 李涛: "支持向量机在风电功率预测中的应用研究", 《中国优秀硕士学位论文全文数据库》 * |
| 陈华友等: "IOWHA算子及其在组合预测中的应用", 《中国管理科学》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108930632A (en) * | 2017-05-25 | 2018-12-04 | 株式会社日立制作所 | Adaptability electric generation management |
| CN108930632B (en) * | 2017-05-25 | 2020-10-27 | 株式会社日立制作所 | Adaptive power generation management |
| CN110659672A (en) * | 2019-09-02 | 2020-01-07 | 国电新能源技术研究院有限公司 | Wind turbine generator output step uncertainty prediction method and device |
| CN110659672B (en) * | 2019-09-02 | 2023-09-26 | 国电新能源技术研究院有限公司 | Method and device for predicting step-by-step uncertainty of output of wind turbine generator |
| CN111753368A (en) * | 2020-05-18 | 2020-10-09 | 重庆长安汽车股份有限公司 | Method for predicting sound absorption performance in vehicle |
| CN111753368B (en) * | 2020-05-18 | 2022-07-08 | 重庆长安汽车股份有限公司 | Method for predicting sound absorption performance in vehicle |
| CN111931981A (en) * | 2020-07-06 | 2020-11-13 | 安徽天尚清洁能源科技有限公司 | Photovoltaic power generation ultra-short-term prediction method based on machine learning multi-model combination |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ding et al. | Forecasting China's electricity consumption using a new grey prediction model | |
| Wu et al. | Probabilistic wind-power forecasting using weather ensemble models | |
| CN103279804B (en) | The Forecasting Methodology of super short-period wind power | |
| Bhaskar et al. | AWNN-assisted wind power forecasting using feed-forward neural network | |
| JP6687241B2 (en) | Power generation power prediction device, server, computer program, and power generation power prediction method | |
| Wang et al. | Decision tree based online stability assessment scheme for power systems with renewable generations | |
| CN107194507A (en) | A kind of short-term wind speed forecasting method of wind farm based on combination SVMs | |
| Islam et al. | Vertical extrapolation of wind speed using artificial neural network hybrid system | |
| CN110929963A (en) | Wind speed prediction method, wind speed prediction device, and storage medium | |
| CN106533750A (en) | System and method for predicting non-steady application user concurrency in cloud environment | |
| CN103942461A (en) | Water quality parameter prediction method based on online sequential extreme learning machine | |
| CN106875033A (en) | A kind of wind-powered electricity generation cluster power forecasting method based on dynamic self-adapting | |
| CN106651007A (en) | Method and device for GRU-based medium and long-term prediction of irradiance of photovoltaic power station | |
| CN104809532A (en) | Method for predicting generating capacity of photovoltaic system | |
| CN106126944A (en) | A kind of power transformer top-oil temperature interval prediction method and system | |
| CN110969306A (en) | Power distribution low-voltage distribution area load prediction method and device based on deep learning | |
| CN103106331B (en) | Based on the lithographic line width Intelligent Forecasting of dimensionality reduction and increment type extreme learning machine | |
| CN109886452A (en) | A kind of ultrashort-term wind power probability forecasting method and system based on experience dynamic modeling | |
| JP2024077563A (en) | Method, system and storage medium for forecasting ultra-short-term water levels of reservoirs taking into account the effects of dynamic water storage capacity | |
| JP6086875B2 (en) | Power generation amount prediction device and power generation amount prediction method | |
| CN116595394A (en) | Training method of wind speed correction model, wind speed prediction method, wind speed prediction equipment and medium | |
| CN115270239A (en) | Bridge reliability prediction method based on dynamic characteristics and intelligent algorithm response surface method | |
| CN109165770A (en) | Novel photovoltaic power forecasting method based on AFSA-Elman | |
| CN114498619A (en) | Wind power prediction method and device | |
| Dong et al. | Short-term building cooling load prediction model based on DwdAdam-ILSTM algorithm: A case study of a commercial building |
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 |
Application publication date: 20170922 |
|
| RJ01 | Rejection of invention patent application after publication |