CN104657787B - A kind of wind power time series combination forecasting method - Google Patents
A kind of wind power time series combination forecasting method Download PDFInfo
- Publication number
- CN104657787B CN104657787B CN201510056776.XA CN201510056776A CN104657787B CN 104657787 B CN104657787 B CN 104657787B CN 201510056776 A CN201510056776 A CN 201510056776A CN 104657787 B CN104657787 B CN 104657787B
- Authority
- CN
- China
- Prior art keywords
- wind power
- dms
- ims
- prediction
- wind
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013277 forecasting method Methods 0.000 title claims abstract description 20
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000005096 rolling process Methods 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims description 18
- 238000000176 thermal ionisation mass spectrometry Methods 0.000 claims description 17
- 238000013055 trapped ion mobility spectrometry Methods 0.000 claims description 17
- 238000002451 electron ionisation mass spectrometry Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000010248 power generation Methods 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 4
- 241001123248 Arma Species 0.000 claims description 2
- 230000005611 electricity Effects 0.000 abstract description 27
- 230000000694 effects Effects 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000002688 persistence Effects 0.000 description 6
- 238000009434 installation Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012795 verification 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
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a kind of wind power time series combination forecasting method, this method is combined prediction based on direct multi-step prediction method and rolling multi-step prediction method to the power of wind a few days ago of wind power plant.The present invention can effectively improve the precision of prediction a few days ago of wind power prediction method, solve the problems, such as that there are wind power prediction effect is poor for current engineering wind power short-term forecast technology, can be that power grid enterprises save huge reserve generation capacity buying expenses, lift power grid enterprises' operation benefits, it can effectively reduce at the same time and abandon wind-powered electricity generation amount, the effective grid connection capacity of wind-powered electricity generation is lifted, enormous profits is brought for electricity power enterprise, promotes energy-saving and emission-reduction.
Description
Technical field
The present invention relates to wind power prediction technical field during generation of electricity by new energy, during more particularly to a kind of wind power
Between combined sequence Forecasting Methodology.
Background technology
It is faced with the whole world under energy crisis and environmental crisis background, wind-power electricity generation is that following many decades are most competitive
One of regenerative resource, wind energy content is huge, is generated electricity using wind energy, can not only reduce environmental pollution, moreover it is possible to reduce power train
The fuel cost of system, brings considerable economic benefit.
Ended for the end of the year 2010, China's wind-powered electricity generation total installed capacity exceedes the U.S., leaps to the first in the world;End in December, 2014, Jiangsu
The installed capacity of power grid grid connected wind power up to 2,970,000 kilowatts, accounts for the 3.7% of Jiangsu Power Grid total installation of generating capacity, becomes and be only second to fire
Second largest main force's power supply of electricity.With swift and violent the increasing of wind-power electricity generation installation total amount, the intrinsic intermittence of wind-powered electricity generation itself, fluctuation
Become the principal element for hindering wind power integration power grid.Each main wind power base wind-abandoning phenomenon in the whole nation takes place frequently, and there is wind in wind-powered electricity generation enterprise
Electricity is not generated, has electricity can not send;Power grid enterprises are lacking high-precision wind power prediction skill to dissolve wind-powered electricity generation to greatest extent at the same time
In the case that art supports, the spare capacity of generating set can only be continuously improved in generation schedule, this necessarily causes unit generation
The continuous deterioration of economy.
Wind-powered electricity generation is accurately predicted, influence of the wind-powered electricity generation to power grid can be greatly lowered, it has been experienced that, accurately and reliably
Wind-powered electricity generation forecasting system be reduce power generation spare capacity, lifted electric system economic operation level, improve wind-powered electricity generation permeability
Key factor;Can be that generation of electricity by new energy Real-Time Scheduling, generation of electricity by new energy are planned, are generation of electricity by new energy monthly plan, new a few days ago at the same time
Energy generating capacity is assessed and is abandoned the estimation of wind-powered electricity generation amount and provides key message.
Recent decades, in order to solve the accurate forecasting problem of wind-powered electricity generation, scientific research personnel is mainly by based on numerical weather forecast
Forecasting Methodology, statistical method (persistence forecasting method, autoregressive moving average method, artificial neural network method), hybrid predicting side
Method, explores the method for improving wind power plant wind power prediction accuracy, and achieves certain achievement.But existing wind work(
The precision of prediction of rate Forecasting Methodology is not still high, and international advanced wind power prediction systematic error is 15% or so, it is impossible to meets work
Journey needs.
The content of the invention
Goal of the invention:The object of the present invention is to provide a kind of precision of prediction height, abandon the few wind power prediction method of wind-powered electricity generation amount.
Technical solution:To reach this purpose, the present invention uses following technical scheme:
Wind power time series combination forecasting method of the present invention, comprises the following steps:
S1:The wind power generation power data of wind power plant to be predicted, i.e. history wind power data are per diem gathered, and will collection
The history wind power data arrived generates a corresponding time series amount X, as training sample set;
S2:Using training sample set X, using time series predicting model as nuclear model, using direct multi-step prediction method, calculate
Generate wind power prediction sequence XDMS;
S3:Using training sample set X, using time series predicting model as nuclear model, using multi-step prediction method is rolled, calculate
Generate wind power prediction sequence XIMS;
S4:Per diem to wind power prediction sequence XDMS、XIMSStatistical analysis is carried out with training sample set X, in confidence level factor alpha
Under respective high precision predicted time scope T is calculatedDMSAnd TIMS, wherein ∑ TDMS+∑TIMS=24 it is small when;
S5:Using time series predicting model as nuclear model, in time range TDMSIt is interior to use direct multistep forecasting method
To day to be predicted in time range TDMSInterior wind power is predicted, and obtains time range TDMSInterior prediction result;In the time
Scope TIMSInterior use rolls multistep forecasting method to day to be predicted in time range TIMSInterior wind power is predicted, when obtaining
Between scope TIMSInterior prediction result;
S6:By time range TDMSInterior prediction result and time range TIMSInterior prediction result is combined, and has been formed
Whole day wind power prediction result to be measured;
When " day " described in above step S1, S4, S5 and S6 is 24 small.
Further, the step S4 comprises the following steps:
S4.1:Per diem to wind power prediction sequence XDMSError calculation is predicted with training sample set X, in confidence level system
The average error curve E of direct multistep forecasting method prediction error band is calculated under number αDMS;
S4.2:Per diem to wind power prediction sequence XIMSError calculation is predicted with training sample set X, in confidence level system
The average error curve E for rolling multistep forecasting method prediction error band is calculated under number αIMS;
S4.3:Compare EDMSWith EIMSNumerical values recited in when 24 is small, chooses EDMS<EIMSCorresponding time interval is the time
Scope TDMS, choose EIMS<EDMSCorresponding time interval is as the time domain scale T for rolling multistep processes predictionIMS;
When " day " described in above step S4.1 and S4.2 is 24 small.
Further, the time series predicting model includes appointing in arma modeling, AR models, MA models and ANN model
Meaning is a kind of.
Beneficial effect:The present invention can effectively improve the precision of prediction a few days ago of wind power prediction method, solve current work
Journey wind power short-term forecast technology there are wind power prediction effect it is poor the problem of, can be that power grid enterprises save huge spare hair
Capacitance buying expenses, lifting power grid enterprises operation benefits, while wind-powered electricity generation amount of abandoning can be effectively reduced, lifting wind-powered electricity generation is effectively grid-connected
Capacity, brings enormous profits for electricity power enterprise, promotes energy-saving and emission-reduction.
Brief description of the drawings
Fig. 1 is the flow chart of the present invention;
Fig. 2 is multistep processes predicted time scope T of the present inventionDMSWith the time domain scale T for rolling multistep processes predictionIMS's
Choosing method;
Fig. 3 is the contrast of the invention with the prediction effect of traditional persistence model.
Embodiment
Below in conjunction with the accompanying drawings, technical scheme is further illustrated by embodiment.
Fig. 1 is the flow chart of the present invention.A kind of wind power time series combination forecasting method provided by the invention, its feature
It is:Comprise the following steps:
S1:The wind power generation power data of wind power plant to be predicted, i.e. history wind power data are per diem gathered, and will collection
The history wind power data arrived generates a corresponding time series amount X, as training sample set;
S2:Using training sample set X, using time series predicting model as nuclear model, using direct multi-step prediction method, calculate
Generate wind power prediction sequence XDMS;
S3:Using training sample set X, using time series predicting model as nuclear model, using multi-step prediction method is rolled, calculate
Generate wind power prediction sequence XIMS;
S4:Per diem to wind power prediction sequence XDMS、XIMSStatistical analysis is carried out with training sample set X, in confidence level factor alpha
Under respective high precision predicted time scope T is calculatedDMSAnd TIMS, wherein ∑ TDMS+∑TIMS=24 it is small when;
S5:Using time series predicting model as nuclear model, in time range TDMSIt is interior to use direct multistep forecasting method
To day to be predicted in time range TDMSInterior wind power is predicted, and obtains time range TDMSInterior prediction result;In the time
Scope TIMSInterior use rolls multistep forecasting method to day to be predicted in time range TIMSInterior wind power is predicted, when obtaining
Between scope TIMSInterior prediction result;
S6:By time range TDMSInterior prediction result and time range TIMSInterior prediction result is combined, and has been formed
Whole day wind power prediction result to be measured.
When " day " described in above step S1, S4, S5 and S6 is 24 small.
Wherein, step S4 comprises the following steps:
S4.1:Per diem to wind power prediction sequence XDMSError calculation is predicted with training sample set X, in confidence level system
The average error curve E of direct multistep forecasting method prediction error band is calculated under number αDMS;
S4.2:Per diem to wind power prediction sequence XIMSError calculation is predicted with training sample set X, in confidence level system
The average error curve E for rolling multistep forecasting method prediction error band is calculated under number αIMS;
S4.3:Compare EDMSWith EIMSNumerical values recited in when 24 is small, chooses EDMS<EIMSCorresponding time interval is the time
Scope TDMS, II areas as shown in Figure 2;Choose EIMS<EDMSCorresponding time interval is as the time domain scale for rolling multistep processes prediction
TIMS, I areas and III areas as shown in Figure 2.
When " day " described in above step S4.1 and S4.2 is 24 small.
For the prediction effect of further verification the method for the present invention, Fig. 3 provides one, in July, 2012 Jiangsu wind power plant point
Do not apply it is predicting after the method for the present invention and traditional persistence model (persistence model) as a result, and point
Do not contrasted with actual measurement wind power, it can be clearly seen that the prediction result of the method for the present invention is coincide with actual measurement wind power curve
Degree is more preferable.Also, according to measuring and calculating, the prediction effect of the method for the present invention is than traditional persistence model (persistence
Model prediction effect) improves 12.74%.
Claims (2)
- A kind of 1. wind power time series combination forecasting method, it is characterised in that:Comprise the following steps:S1:Per diem gather the wind power generation power data of wind power plant to be predicted, i.e. history wind power data, and will collect History wind power data generates a corresponding time series amount X, as training sample set;S2:Using training sample set X, using time series predicting model as nuclear model, using direct multi-step prediction method, generation is calculated Wind power prediction sequence XDMS;S3:Using training sample set X, using time series predicting model as nuclear model, using multi-step prediction method is rolled, generation is calculated Wind power prediction sequence XIMS;S4:Per diem to wind power prediction sequence XDMS、XIMSStatistical analysis is carried out with training sample set X, is pressed under confidence level factor alpha Respective high precision predicted time scope T is calculated according to the following stepsDMSAnd TIMS, wherein ∑ TDMS+∑TIMS=24 it is small when,S4.1:Per diem to wind power prediction sequence XDMSError calculation is predicted with training sample set X, under confidence level factor alpha The average error curve E of direct multistep forecasting method prediction error band is calculatedDMS;S4.2:Per diem to wind power prediction sequence XIMSError calculation is predicted with training sample set X, under confidence level factor alpha The average error curve E for rolling multistep forecasting method prediction error band is calculatedIMS;S4.3:Compare EDMSWith EIMSNumerical values recited in when 24 is small, chooses EDMS<EIMSCorresponding time interval is time range TDMS, choose EIMS<EDMSCorresponding time interval is as the time domain scale T for rolling multistep processes predictionIMS;S5:Using time series predicting model as nuclear model, in time range TDMSIt is interior to be treated using direct multistep forecasting method Predict day in time range TDMSInterior wind power is predicted, and obtains time range TDMSInterior prediction result;In time range TIMSInterior use rolls multistep forecasting method to day to be predicted in time range TIMSInterior wind power is predicted, and obtains time model Enclose TIMSInterior prediction result;S6:By time range TDMSInterior prediction result and time range TIMSInterior prediction result is combined, and is formed complete Day wind power prediction result to be measured;When " day " described in above step S1, S4, S5 and S6 is 24 small.
- 2. wind power time series combination forecasting method according to claim 1, it is characterised in that:The time series is pre- Surveying model includes any one in arma modeling, AR models, MA models and ANN model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510056776.XA CN104657787B (en) | 2015-02-03 | 2015-02-03 | A kind of wind power time series combination forecasting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510056776.XA CN104657787B (en) | 2015-02-03 | 2015-02-03 | A kind of wind power time series combination forecasting method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104657787A CN104657787A (en) | 2015-05-27 |
CN104657787B true CN104657787B (en) | 2018-05-04 |
Family
ID=53248884
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510056776.XA Active CN104657787B (en) | 2015-02-03 | 2015-02-03 | A kind of wind power time series combination forecasting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104657787B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104915736A (en) * | 2015-06-29 | 2015-09-16 | 东北电力大学 | Method for improving accuracy of wind power combined prediction based on improved entropy weight 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 |
CN108536652A (en) * | 2018-03-15 | 2018-09-14 | 浙江大学 | A kind of short-term vehicle usage amount prediction technique based on arma modeling |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102226903A (en) * | 2011-05-11 | 2011-10-26 | 北方工业大学 | Rolling forecasting combined model method |
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 |
CN102682207A (en) * | 2012-04-28 | 2012-09-19 | 中国科学院电工研究所 | Ultrashort combined predicting method for wind speed of wind power plant |
CN104298859A (en) * | 2014-09-22 | 2015-01-21 | 国电南瑞科技股份有限公司 | Multi-interval-prediction-result weighting integration method |
-
2015
- 2015-02-03 CN CN201510056776.XA patent/CN104657787B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102226903A (en) * | 2011-05-11 | 2011-10-26 | 北方工业大学 | Rolling forecasting combined model method |
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 |
CN102682207A (en) * | 2012-04-28 | 2012-09-19 | 中国科学院电工研究所 | Ultrashort combined predicting method for wind speed of wind power plant |
CN104298859A (en) * | 2014-09-22 | 2015-01-21 | 国电南瑞科技股份有限公司 | Multi-interval-prediction-result weighting integration method |
Non-Patent Citations (1)
Title |
---|
风电场功率短期预测方法研究;韩爽;《中国博士学位论文全文数据库》;20090215(第2期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN104657787A (en) | 2015-05-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jiang et al. | Runoff forecast uncertainty considered load adjustment model of cascade hydropower stations and its application | |
CN103545832B (en) | A kind of photovoltaic system energy accumulation capacity configuration based on generating predicated error | |
CN102269124B (en) | Ultra-short term wind power station generated power forecasting system | |
CN104361406B (en) | A kind of photovoltaic plant can utilize solar power generation amount Forecasting Methodology | |
CN103296701B (en) | Active power control method in wind power plant | |
CN104268659B (en) | A kind of photovoltaic power station power generation power ultra-short term prediction method | |
CN103679282B (en) | The Forecasting Methodology of wind power climbing | |
CN102479347B (en) | Method and system for forecasting short-term wind speed of wind farm based on data driving | |
CN104463356A (en) | Photovoltaic power generation power prediction method based on multi-dimension information artificial neural network algorithm | |
CN103683326A (en) | Method for calculating optimal admitting ability for wind power multipoint access of regional power grid | |
CN103065202A (en) | Wind power plant ultrashort term wind speed prediction method based on combination kernel function | |
CN105243502A (en) | Hydropower station scheduling risk assessment method and system based on runoff interval prediction | |
CN104657787B (en) | A kind of wind power time series combination forecasting method | |
CN106611243A (en) | Residual correction method for wind speed prediction based on GARCH (Generalized ARCH) model | |
CN103440529A (en) | Lateral error translation modification method of wind electricity power forecast system | |
CN103345585A (en) | Wind power prediction correction method and system based on support vector machine | |
CN105046349B (en) | It is a kind of meter and wake effect wind power forecasting method | |
CN105262148B (en) | The planning year power balance method of meter and wind power output characteristic | |
CN104239979B (en) | A kind of wind power plant generated output ultra-short term prediction method | |
CN104901341A (en) | Thermal power generation coal consumption calculating method based on power unit output and changing load-acceleration | |
Roy | Impact of short duration wind variations on output of a pitch angle controlled turbine | |
CN108268963A (en) | A kind of photovoltaic power short term prediction method for reviewing one's lessons by oneself positive test | |
CN106447132A (en) | A medium-and-long term generating capacity prediction method for a regional photovoltaic power station group | |
CN104657786A (en) | Short-term wind power mixed predicting method based on Boosting algorithm | |
CN103065049B (en) | A kind of wind power real-time estimate computing method based on coif5 small echo real-time decomposition |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |