CN110263998A - Multi-source numerical weather forecast set bilayer modification method - Google Patents
Multi-source numerical weather forecast set bilayer modification method Download PDFInfo
- Publication number
- CN110263998A CN110263998A CN201910531956.7A CN201910531956A CN110263998A CN 110263998 A CN110263998 A CN 110263998A CN 201910531956 A CN201910531956 A CN 201910531956A CN 110263998 A CN110263998 A CN 110263998A
- Authority
- CN
- China
- Prior art keywords
- data
- weather forecast
- numerical weather
- wind
- cluster
- 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.)
- Granted
Links
- 238000002715 modification method Methods 0.000 title claims abstract description 7
- 238000000034 method Methods 0.000 claims abstract description 20
- 238000012937 correction Methods 0.000 claims abstract description 10
- 238000005259 measurement Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 230000006870 function Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 2
- 238000012217 deletion Methods 0.000 claims 1
- 230000037430 deletion Effects 0.000 claims 1
- 230000008030 elimination Effects 0.000 claims 1
- 238000003379 elimination reaction Methods 0.000 claims 1
- 238000010248 power generation Methods 0.000 abstract description 2
- 238000003066 decision tree Methods 0.000 description 9
- 238000012549 training Methods 0.000 description 5
- 238000007637 random forest analysis Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 230000005611 electricity Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004992 fission Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
A kind of multi-source numerical weather forecast set bilayer modification method is proposed, technical field of new energy power generation is belonged to, which comprises acquisition multi-source numerical weather forecast data survey wind data with corresponding reality;The set and corresponding practical survey wind data, logarithm data of weather forecast formed according to multi-source numerical weather forecast data carries out first layer amendment;Different weather scenes is divided according to the revised multi-source numerical weather forecast data of first layer;Analyze the correlation under different weather scene at each spatial position between multi-source numerical weather forecast data and corresponding practical survey wind data;The mutual correction model of numerical weather forecast is established for each weather scene according to relevance ranking, logarithm data of weather forecast carries out second layer amendment.This method improves the accuracy of weather forecast, improves the precision of prediction of wind power.
Description
Technical field
The invention belongs to technical field of new energy power generation more particularly to a kind of multi-source numerical weather forecast set bilayer to correct
Method.
Background technique
Extensive renewable energy access power grid is one of the prominent features of various countries' electric system future development.With wind-powered electricity generation
Power supply accounting in countries in the world rises year by year, and the intrinsic intermittence of wind-powered electricity generation and fluctuation lead to large-scale wind power integration meeting
Great impact is caused to the safe and stable and economical operation of power grid, accurate wind power prediction is to solve the problems, such as this necessity
One of means.Numerical weather forecast (NWP) is the key input and wind-powered electricity generation function of wind power prediction and its analysis of uncertainty
The main source of rate prediction error.In most cases, numerical weather forecast bring error is twice of prediction algorithm or more,
Therefore improving numerical weather forecast precision is to improve the key link of wind farm power prediction precision.
The amendment of existing numerical weather forecast mainly combines actual measurement wind speed to single position by statistical method both at home and abroad
The systematic error of point value weather forecast is modified.Such method mostly just uses the numerical weather forecast of single source,
It is closed by establishing the mapping in a certain location point historical series between synchronization numerical weather forecast data and measured data respectively
System, and then the numerical weather forecast data of future time instance are modified.But China has a vast territory, climate type and orographic condition
Complicated multiplicity, considerably increases the difficulty of numerical weather forecast (NWP) and subsequent wind power prediction.Existing Numerical Weather is pre-
Report correction method for repairment to have the following problems: single source numerical weather forecast is limited to the adaptability of complicated weather condition, becomes shadow
One of an important factor for ringing wind power prediction precision;From the point of view of physical modeling process, the calculating of Mesoscale Meteorology is differentiated
Rate is too low, and the Regional factors such as landform, roughness, wake flow can not be embodied in its described flow field, and single position is caused to be counted
Value data of weather forecast has often lacked the time and space usage relationship of practical wind wave process within the scope of Large Scale Wind Farm Integration, or needs to consume
Take a large amount of computing resources to be solved, needs to consider the correlative character of practical wind regime at multi-point for above-mentioned minute yardstick factor
It is corrected again.
Summary of the invention
In order to solve the problems, such as to be proposed above, the present invention proposes a kind of multi-source numerical weather forecast set bilayer amendment side
Method.
According to an aspect of the present invention, it proposes a kind of multi-source numerical weather forecast set bilayer modification method, the party
Method includes: step 1: it acquires the multi-source numerical weather forecast data at multiple spatial positions and surveys wind data with corresponding reality, it is right
Above-mentioned data are cleaned and are pre-processed;Step 2: being directed to the spatial position, pre- according to the multi-source Numerical Weather at the position
Set (NWP data) and corresponding practical survey wind data of the count off according to formation, establish numerical weather forecast from local correction model, logarithm
It is worth data of weather forecast and carries out first layer amendment;Step 3: according to the revised multi-source numerical weather forecast data set of first layer
It closes result and divides different weather scenes;Step 4: multi-source Numerical Weather at each spatial position under analysis different weather scene
Forecast data aggregated result and the corresponding practical correlation surveyed between wind data;Step 5: it according to relevance ranking, chooses not on the same day
Numerical value is established for each weather scene with the actual measurement the most similar numerical weather forecast data of air speed data under gas field scape
The mutual correction model of weather forecast, logarithm data of weather forecast carry out second layer amendment.
According to an aspect of the present invention, the multi-source numerical weather forecast data refer to different initial fields, different parameters
Numerical weather forecast result when scheme, difference under secondary or different computation models.
According to an aspect of the present invention, in step 1, the cleaning and pretreatment include identification to abnormal data
Interpolation with deleting and to missing data.
According to an aspect of the present invention, in step 2, with the multi-source numerical weather forecast number at the spatial position
It is combined into input according to the collection of formation, the practical survey wind data in corresponding position is learning objective, establishes the custom holotype of numerical weather forecast
Type eliminates the Systematic Errors in numerical simulation.
According to an aspect of the present invention, the practical wind data of surveying includes that anemometer measurement is simultaneously in wind power plant operation data
The actual measurement wind data that anemometer measures and records at the practical survey wind data of record and anemometer tower position.
According to an aspect of the present invention, in step 3, using Fuzzy C-Means Cluster Algorithm by numerical weather forecast number
According to different weather scenes is divided into, numerical weather forecast data set X={ x is given1,x2,…,xn, wherein n is the number of sample
Amount, k are the class number of weather scene, mj(j=1,2 ..., k) is the center of each cluster, μj(xi) it is that i-th of sample is corresponding
The subordinating degree function of jth class, then the cluster loss function J based on subordinating degree functionfAre as follows:
Wherein b > 1, for the constant of the fog-level of control cluster result.
Enable JfTo mjAnd μj(xi) partial derivative be 0, acquire the necessary condition of above formula minimum:
Then the step of Fuzzy C-Means Cluster Algorithm are as follows:
(1) cluster classification number C is given, iteration convergence condition is set, initializes each cluster centre;
(2) following operation is repeated, until the angle value that is subordinate to of each sample is stablized:
A. subordinating degree function is calculated according to formula (4) with current cluster centre;
B. the center of each cluster is recalculated according to formula (3) with current subordinating degree function.
When algorithmic statement, just obtains all kinds of cluster centres and each sample and angle value is subordinate to for all kinds of, thus
Fuzzy clustering is completed to divide.
According to an aspect of the present invention, in step 4, by calculating multi-source numerical weather forecast data acquisition system result
With the practical Pearson correlation coefficient for surveying wind data, Two-Dimensional Correlativity evaluations matrix is constructed.
According to an aspect of the present invention, in step 5, for certain weather scene, with numerical value day under the weather scene
At gas forecast data and actual measurement air speed data the numerical weather forecast data and other positions to be verified of similar location the most
Numerical weather forecast data be input, the practical survey wind data at other positions to be verified is learning objective, establishes the day
The mutual Knowledge Verification Model of numerical weather forecast under gas field scape.
It can be seen that the present invention is aiming at the problems existing in the prior art, using the numerical weather forecast data in a variety of sources
Its adaptability to complicated weather condition is improved, using numerical weather forecast data at multiple spatial positions, in independent position
The multi-point numerical weather forecast that different weather scene is carried out from the basis of correcting mutually verifies, to correct minute yardstick factor, this
Outside, numerical weather forecast amendment of the invention includes the double-deck amendment, i.e., surveys wind data using history and carry out from correcting and multi-point
The mutual verification of numerical weather forecast improves the precision of prediction of wind power to improve the accuracy of weather forecast.
Detailed description of the invention
Fig. 1 is the flow chart of multi-source numerical weather forecast set bilayer modification method.
Specific embodiment
Technical solution of the present invention is described in further detail with specific embodiment with reference to the accompanying drawing.
Fig. 1 is the flow chart of multi-source numerical weather forecast set bilayer modification method, as shown in Figure 1, the method includes
Following steps:
Step 1: acquiring the multi-source numerical weather forecast data at multiple spatial positions and survey wind data with corresponding reality,
Above-mentioned data are cleaned and pre-processed.
Specifically, acquisition spatial position multi-source numerical weather forecast data (NWP-1, NWP-2 ..., NWP-n) with it is right
It answers the actual measurement at position to survey wind data, abnormal data is identified according to range check, correlation test and trend test criterion and incites somebody to action
It is deleted, and using nearest neighbor interpolation algorithm and linear interpolation algorithm interpolation missing data, restores reducible data, after guaranteeing
Continuous step learning model input data is accurate corresponding with label data.At the beginning of the multi-source numerical weather forecast data refer to difference
Beginning field, different parameters scheme, it is different when time or different computation models under numerical weather forecast as a result, specifically include wind speed,
Wind direction, temperature, air pressure, humidity and atmospheric density data.
Step 2: being directed to the spatial position, the set formed according to the multi-source numerical weather forecast data at the position
(NWP data) and it is corresponding it is practical survey wind data, establish numerical weather forecast from local correction model, logarithm data of weather forecast into
The amendment of row first layer.
Specifically, it is directed to some spatial position, the set formed with the multi-source numerical weather forecast data at the position
For input, the practical survey wind data in corresponding position (wind speed, wind direction, temperature, air pressure, humidity and atmospheric density data) is study mesh
Mark establishes numerical weather forecast from local correction model, the systematicness eliminated in numerical simulation is missed using random forests algorithm
Difference.As shown in Figure 1, the actual measurement wind data (wind speed, wind direction etc.) for having instrument for wind measurement to measure and record in wind power plant operation data, is surveyed
Wind tower data are the actual measurement wind datas (wind speed, wind direction etc.) that instrument for wind measurement measures and records at anemometer tower position, and herein is more
A spatial position refers to each Wind turbines and each anemometer tower.
N is randomly selected from numerical weather forecast and practical survey in wind data training set S using the Bootstrap methods of sampling
A training sample subset S1,S2,…,SNFor constructing N decision tree, the size of each training subset is about original training set
2/3rds, sampling is random and sampling with replacement every time, so that there are certain repetition, mesh for the sample in training subset
Be in order to make the decision tree in forest be unlikely to generate locally optimal solution;Then these decision trees are combined, with each
The result that is returned as random forest of average value of tree output result;To guarantee randomness when decision tree building, avoided
Fitting problems randomly select m1 kind as random character variable participative decision making when each decision tree constructs from m attribute
The sampling of the fission process of tree node, referred to as random attribute subspace, wherein m1 takes less than or equal to log2(m+1) maximum is just whole
Number;Furthermore the quantity N needs of stochastic decision tree adjust;Finally based on the regression result of random forests algorithm are as follows:
In formula: X is input variable, h (X;θk) it is single decision-tree model, wherein k=1,2 ..., N, N are decision tree
Number, θkFor the parameter of single decision tree.
Step 3: different day gas fields is divided according to the revised multi-source numerical weather forecast data acquisition system result of first layer
Scape.
In one embodiment, numerical weather forecast data are divided into not by step 3 using Fuzzy C-Means Cluster Algorithm
Same weather scene gives numerical weather forecast data set X={ x1,x2,…,xn, wherein n is the quantity of sample, and k is weather
The class number of scene, mj(j=1,2 ..., k) is the center of each cluster, μj(xi) it is that i-th of sample corresponds to being subordinate to for jth class
Function is spent, then the cluster loss function J based on subordinating degree functionfIt is writeable are as follows:
Wherein b > 1, for the constant of the fog-level of control cluster result.
Enable JfTo mjAnd μj(xi) partial derivative be 0, acquire the necessary condition of above formula minimum:
Then the step of Fuzzy C-Means Cluster Algorithm are as follows:
(1) cluster classification number C is given, iteration convergence condition is set, initializes each cluster centre;
(2) following operation is repeated, until the angle value that is subordinate to of each sample is stablized:
A. subordinating degree function is calculated according to formula (4) with current cluster centre;
B. the center of each cluster is recalculated according to formula (3) with current subordinating degree function;
When algorithmic statement, just obtains all kinds of cluster centres and each sample and angle value is subordinate to for all kinds of, thus
Fuzzy clustering is completed to divide.
Step 4: under analysis different weather scene at each spatial position multi-source numerical weather forecast data acquisition system result with
The corresponding practical correlation surveyed between wind data.
Two-Dimensional Correlativity evaluations matrix is constructed by calculating the Pearson correlation coefficient of the two according to one embodiment.
Step 5: according to relevance ranking, choose under different weather scene with the actual measurement the most similar numerical value of air speed data
Data of weather forecast, i.e., with the practical numerical weather forecast for surveying the most similar position of wind data as mutual at different spatial
The input of Knowledge Verification Model establishes the mutual correction model of numerical weather forecast, logarithm weather forecast number for each weather scene
According to progress second layer amendment.
According to one embodiment, for certain weather scene, with numerical weather forecast data under the weather scene and actual measurement
The numerical weather forecast number at numerical weather forecast data and other points to be verified at air speed data point the most similar
According to input, the practical survey wind data at other points to be verified establishes the day using random forests algorithm for learning objective
The mutual Knowledge Verification Model of numerical weather forecast under gas field scape, further, it is also possible to be modified using other algorithms, such as deep learning
Stacking automatic coding machine in algorithm etc..
Above-mentioned specific example is only to illustrate, not as limiting the scope of the invention.Those skilled in the art can root
According to needing the specific example proposed to the application to modify or adjust, these modifications or adjustment equally fall into the application and are wanted
Within the scope of asking protection.
Claims (8)
1. a kind of multi-source numerical weather forecast set bilayer modification method, which is characterized in that this method comprises:
Step 1: the multi-source numerical weather forecast data at multiple spatial positions are acquired and survey wind data with corresponding reality, to upper
Data are stated to be cleaned and pre-processed;
Step 2: being directed to the spatial position, the set (NWP formed according to the multi-source numerical weather forecast data at the position
Data) and it is corresponding it is practical survey wind data, establish numerical weather forecast from local correction model, logarithm data of weather forecast carries out the
One layer of amendment;
Step 3: different weather scenes is divided according to the revised multi-source numerical weather forecast data acquisition system result of first layer;
Step 4: multi-source numerical weather forecast data acquisition system result and corresponding at each spatial position under analysis different weather scene
The practical correlation surveyed between wind data;
Step 5: according to relevance ranking, choose under different weather scene with the actual measurement the most similar Numerical Weather of air speed data
Forecast data establishes the mutual correction model of numerical weather forecast for each weather scene, and logarithm data of weather forecast carries out
Second layer amendment.
2. according to the method described in claim 1, its feature exists:
The multi-source numerical weather forecast data refer to secondary or different calculating when different initial fields, different parameters scheme, difference
Numerical weather forecast result under model.
3. according to the method described in claim 1, it is characterized by:
In step 1, the cleaning and pretreatment include identification to abnormal data and deletion and insert to missing data
It mends.
4. according to the method described in claim 1, it is characterized by:
In step 2, input, corresponding position are combined into the collection that the multi-source numerical weather forecast data at the spatial position are formed
Setting the practical wind data of surveying in place is learning objective, and establishing numerical weather forecast from local correction model, elimination numerical simulation is
System property error.
5. according to the method described in claim 4, it is characterized by:
The practical wind data of surveying includes the practical survey wind data and survey wind that anemometer measures and records in wind power plant operation data
The actual measurement wind data that anemometer measures and records at tower position.
6. according to the method described in claim 1, it is characterized by:
In step 3, numerical weather forecast data are divided into using Fuzzy C-Means Cluster Algorithm by different weather scenes, are given
Fixed number value data of weather forecast collection X={ x1,x2,…,xn, wherein n is the quantity of sample, and k is the class number of weather scene,
mj(j=1,2 ..., k) is the center of each cluster, μj(xi) it is the subordinating degree function that i-th of sample corresponds to jth class, then based on person in servitude
The cluster loss function J of category degree functionfAre as follows:
Wherein b > 1, for the constant of the fog-level of control cluster result;
Enable JfTo mjAnd μj(xi) partial derivative be 0, acquire the necessary condition of above formula minimum:
Then the step of Fuzzy C-Means Cluster Algorithm are as follows:
(1) cluster classification number C is given, iteration convergence condition is set, initializes each cluster centre;
(2) following operation is repeated, until the angle value that is subordinate to of each sample is stablized:
A. subordinating degree function is calculated according to formula (4) with current cluster centre;
B. the center of each cluster is recalculated according to formula (3) with current subordinating degree function;
When algorithmic statement, all kinds of cluster centres and each sample is just obtained and angle value are subordinate to for all kinds of, to complete
Fuzzy clustering divides.
7. according to the method described in claim 1, it is characterized by:
It is related to the practical survey Pearson came of wind data by calculating multi-source numerical weather forecast data acquisition system result in step 4
Coefficient constructs Two-Dimensional Correlativity evaluations matrix.
8. according to the method described in claim 1, it is characterized by:
In step 5, for certain weather scene, with numerical weather forecast data under the weather scene and actual measurement air speed data
The numerical weather forecast data of similar location and the numerical weather forecast data at other positions to be verified the most are input,
Practical survey wind data at other positions to be verified is learning objective, establishes the mutual school of numerical weather forecast under the weather scene
Test model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910531956.7A CN110263998B (en) | 2019-06-19 | 2019-06-19 | Double-layer correction method for multisource numerical weather forecast set |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910531956.7A CN110263998B (en) | 2019-06-19 | 2019-06-19 | Double-layer correction method for multisource numerical weather forecast set |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110263998A true CN110263998A (en) | 2019-09-20 |
CN110263998B CN110263998B (en) | 2024-02-13 |
Family
ID=67919417
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910531956.7A Active CN110263998B (en) | 2019-06-19 | 2019-06-19 | Double-layer correction method for multisource numerical weather forecast set |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110263998B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110934061A (en) * | 2019-12-26 | 2020-03-31 | 裕华生态环境股份有限公司 | Garden irrigation water-saving system |
CN111309852A (en) * | 2020-03-16 | 2020-06-19 | 青岛百洋智能科技股份有限公司 | Method, system, apparatus and storage medium for generating a visual decision tree (set) model |
CN113537648A (en) * | 2021-09-16 | 2021-10-22 | 国能日新科技股份有限公司 | Wind speed prediction method and device based on set data |
CN113610287A (en) * | 2021-07-27 | 2021-11-05 | 远景智能国际私人投资有限公司 | Optical power forecasting method and device, computer equipment and storage medium |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009138523A (en) * | 2007-12-03 | 2009-06-25 | Mitsubishi Electric Corp | Method of estimating output of wind power generation |
CN102269124A (en) * | 2011-06-30 | 2011-12-07 | 内蒙古电力勘测设计院 | Ultra-short term wind power station generated power forecasting system |
CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
US20140025354A1 (en) * | 2012-07-20 | 2014-01-23 | Tata Consultancy Services Limited | Method and system for adaptive forecast of wind resources |
CN104951846A (en) * | 2015-06-02 | 2015-09-30 | 国电南京自动化股份有限公司 | Microgrid short-term power and load prediction system and dynamic error classification and correction method |
CN106339568A (en) * | 2015-07-08 | 2017-01-18 | 中国电力科学研究院 | Numerical weather prediction method based on mixed ambient field |
CN106803129A (en) * | 2015-11-26 | 2017-06-06 | 中国电力科学研究院 | A kind of wind power ensemble prediction method based on multi-source numerical weather forecast |
CN107506878A (en) * | 2017-10-11 | 2017-12-22 | 华中科技大学 | A kind of power system multi-source scheduling method for considering honourable smoothing effect |
CN108009689A (en) * | 2017-12-22 | 2018-05-08 | 北京金风慧能技术有限公司 | The training of wind speed correction model, wind speed modification method, device and wind power generating set |
CN108205717A (en) * | 2017-12-30 | 2018-06-26 | 国网江苏省电力公司无锡供电公司 | A kind of photovoltaic generation power Multiple Time Scales Forecasting Methodology |
CN109685257A (en) * | 2018-12-13 | 2019-04-26 | 国网青海省电力公司 | A kind of photovoltaic power generation power prediction method based on Support vector regression |
-
2019
- 2019-06-19 CN CN201910531956.7A patent/CN110263998B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009138523A (en) * | 2007-12-03 | 2009-06-25 | Mitsubishi Electric Corp | Method of estimating output of wind power generation |
CN102269124A (en) * | 2011-06-30 | 2011-12-07 | 内蒙古电力勘测设计院 | Ultra-short term wind power station generated power forecasting system |
US20140025354A1 (en) * | 2012-07-20 | 2014-01-23 | Tata Consultancy Services Limited | Method and system for adaptive forecast of wind resources |
CN103268366A (en) * | 2013-03-06 | 2013-08-28 | 辽宁省电力有限公司电力科学研究院 | Combined wind power prediction method suitable for distributed wind power plant |
CN104951846A (en) * | 2015-06-02 | 2015-09-30 | 国电南京自动化股份有限公司 | Microgrid short-term power and load prediction system and dynamic error classification and correction method |
CN106339568A (en) * | 2015-07-08 | 2017-01-18 | 中国电力科学研究院 | Numerical weather prediction method based on mixed ambient field |
CN106803129A (en) * | 2015-11-26 | 2017-06-06 | 中国电力科学研究院 | A kind of wind power ensemble prediction method based on multi-source numerical weather forecast |
CN107506878A (en) * | 2017-10-11 | 2017-12-22 | 华中科技大学 | A kind of power system multi-source scheduling method for considering honourable smoothing effect |
CN108009689A (en) * | 2017-12-22 | 2018-05-08 | 北京金风慧能技术有限公司 | The training of wind speed correction model, wind speed modification method, device and wind power generating set |
CN108205717A (en) * | 2017-12-30 | 2018-06-26 | 国网江苏省电力公司无锡供电公司 | A kind of photovoltaic generation power Multiple Time Scales Forecasting Methodology |
CN109685257A (en) * | 2018-12-13 | 2019-04-26 | 国网青海省电力公司 | A kind of photovoltaic power generation power prediction method based on Support vector regression |
Non-Patent Citations (5)
Title |
---|
刘晓楠等: "基于非参数核密度估计与数值天气预报的风速预测修正方法", vol. 37, no. 10, pages 15 - 20 * |
张浩: "基于深度学习的风电功率预测方法研究", no. 3 * |
杨家然等: "基于数值天气预报及模糊聚类的风电功率智能组合预测", vol. 38, no. 3, pages 669 - 675 * |
王一妹: "基于CFD流场预计算的复杂地形风电场功率预测方法研究", no. 1 * |
阎洁: "风电场功率预测不确定性分析方法及其应用研究", no. 12 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110934061A (en) * | 2019-12-26 | 2020-03-31 | 裕华生态环境股份有限公司 | Garden irrigation water-saving system |
CN111309852A (en) * | 2020-03-16 | 2020-06-19 | 青岛百洋智能科技股份有限公司 | Method, system, apparatus and storage medium for generating a visual decision tree (set) model |
CN111309852B (en) * | 2020-03-16 | 2021-09-03 | 青岛百洋智能科技股份有限公司 | Method, system, device and storage medium for generating visual decision tree set model |
CN113610287A (en) * | 2021-07-27 | 2021-11-05 | 远景智能国际私人投资有限公司 | Optical power forecasting method and device, computer equipment and storage medium |
CN113537648A (en) * | 2021-09-16 | 2021-10-22 | 国能日新科技股份有限公司 | Wind speed prediction method and device based on set data |
Also Published As
Publication number | Publication date |
---|---|
CN110263998B (en) | 2024-02-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110263998A (en) | Multi-source numerical weather forecast set bilayer modification method | |
CN106779226B (en) | Fan batch power prediction method based on mixed kernel machine learning | |
CN112529282A (en) | Wind power plant cluster short-term power prediction method based on space-time graph convolutional neural network | |
CN105868853B (en) | Method for predicting short-term wind power combination probability | |
CN104036121B (en) | Wind measurement data wind speed correction method based on probability distribution transfer | |
CN108428017A (en) | Wind power interval prediction method based on core extreme learning machine quantile estimate | |
CN110766200A (en) | Method for predicting generating power of wind turbine generator based on K-means mean clustering | |
CN109167387A (en) | Wind field wind power forecasting method | |
CN113657662B (en) | Downscaling wind power prediction method based on data fusion | |
CN109902340A (en) | It is a kind of meter and complicated meteorology coupled characteristic multi-source-lotus combine scene generating method | |
CN110460038A (en) | It is a kind of to be related to more scene method for expansion planning of power transmission network of new-energy grid-connected | |
CN112862274A (en) | WRF-LES and Deepar combined wind power prediction method | |
CN113379142A (en) | Short-term wind power prediction method based on wind speed correction and fusion model | |
CN112651576A (en) | Long-term wind power prediction method and device | |
CN116702937A (en) | Photovoltaic output day-ahead prediction method based on K-means mean value clustering and BP neural network optimization | |
CN115204444A (en) | Photovoltaic power prediction method based on improved cluster analysis and fusion integration algorithm | |
CN117060407B (en) | Wind power cluster power prediction method and system based on similar day division | |
US20240054267A1 (en) | Method for planning a layout of a renewable energy site | |
Chicco et al. | Characterization of solar irradiance profiles for photovoltaic system studies through data rescaling in time and amplitude | |
CN115544895B (en) | Photovoltaic power station annual output guarantee rate model optimization method | |
CN111062509A (en) | Wind power medium-short term prediction method and system based on improved neural network | |
CN113887839A (en) | Clustering-based wind power plant ultra-short-term wind power prediction method and system | |
CN113591399B (en) | Short-term wind power prediction method | |
CN111027816B (en) | Photovoltaic power generation efficiency calculation method based on data envelope analysis | |
CN113610285A (en) | Power prediction method for distributed wind power |
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 |