CN102496926B - Method for judging and processing wind farm power prediction input data - Google Patents
Method for judging and processing wind farm power prediction input data Download PDFInfo
- Publication number
- CN102496926B CN102496926B CN 201110422235 CN201110422235A CN102496926B CN 102496926 B CN102496926 B CN 102496926B CN 201110422235 CN201110422235 CN 201110422235 CN 201110422235 A CN201110422235 A CN 201110422235A CN 102496926 B CN102496926 B CN 102496926B
- Authority
- CN
- China
- Prior art keywords
- wind
- energy turbine
- turbine set
- wind energy
- data
- 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
Images
Landscapes
- Wind Motors (AREA)
Abstract
The invention discloses a method for judging and processing wind farm power prediction input data. The wind farm power prediction input data is historical data of the wind farm, and contains normal data, damaged data and distorted data, wherein the damaged data and the distorted data has harmful effect on the reliability of a sample set of training data of a wind farm power prediction mode, and further influences the precision of the wind power prediction. The invention provides a preprocessing method for wind farm historical data, in the method, the normal data and the bad data in the data are distinguished, and the damaged data or the bad data in the data is increased or changed, so that the reliability of the sample set of the training data of the prediction model is enhanced, and the accuracy of wind farm power prediction is improved.
Description
Technical field
The application belongs to technical field of power systems, is specifically related to a kind of judgement and processing method of wind energy turbine set power prediction input data
Background technology
When carrying out wind power prediction, the direct impact prediction precision of two class data is arranged, the one, meteorological data, the 2nd, historical power data, the former directly obtains from weather bureau, is not carrying out relevant treatment herein, the preliminary treatment of the historical power data of our primary studies.Because blower fan instrument for wind measurement fault, blower fan are normal or disorderly closedown, communication failure, interchannel noise etc., following three class problems appear in historical power data regular meeting: 1) damaged data; 2) distortion data; 3) noise pollution.
The historical data of wind energy turbine set comprises measured data (comprising wind speed, wind direction, air pressure, temperature, historical power) and statistics (comprising maximum wind velocity, minimum windspeed, mean wind speed, wind speed, wind direction sine, wind direction cosine etc.), they split into a plurality of sample sets according to the different terrain height, these sample sets are used for the wind-powered electricity generation forecast model is trained, and form the wind energy turbine set short-period forecast model of exerting oneself in 24 hours.So the accuracy of wind energy turbine set historical data is directly connected to the accuracy of wind power prediction in the sample set, and the damaged data that exist in the historical data, distortion data can produce unpredictable influence to validity, the accuracy of predicting of sample set.At present also do not carrying out deep research and the processing method of unmatchful such data aspect the preliminary treatment of wind energy turbine set historical data at home.
Summary of the invention
At the problems referred to above that prior art exists, the application has proposed judgement and the processing method of the bad input data of wind energy turbine set power prediction.
The application specifically by the following technical solutions.
Judgement and the processing method of a kind of wind energy turbine set power prediction input data is characterized in that, judge normal data and bad data in the wind energy turbine set power prediction input data by described method, and bad data is wherein handled; Said method comprising the steps of:
(1) wind energy turbine set power is measured by the transformer station that is connect by the wind-powered electricity generation place, gathers the wind farm wind velocity value when measuring wind energy turbine set power;
(2) calculate the rate of change of the wind energy turbine set performance number that each wind energy turbine set performance number of constantly gathering gathers with respect to previous moment, with it as this wind energy turbine set power samples value rate of change, i.e. η constantly
Wind, h=(P
Wind, h-P
Wind, h-1)/P
Wind, h-1, η wherein
Wind, hFor wind energy turbine set at h constantly than h-1 power variation rate constantly, P
Wind, h-1Be wind energy turbine set h-1 performance number constantly, P
Wind, hBe wind energy turbine set h performance number constantly;
(3) calculating corresponds to the wind speed in the corresponding moment in the step (2) with respect to the wind speed rate of change of previous moment wind speed, i.e. ζ
Wind, h=(V
Wind, h-V
Wind, h-1)/V
Wind, h-1, ζ wherein
Wind, hFor wind energy turbine set at h constantly than the wind speed rate of change of h-1 period, V
Wind, h-1Be wind energy turbine set h-1 air speed value constantly, V
Wind, hBe wind energy turbine set h air speed value constantly;
(4) with step (2) resulting wind energy turbine set power samples value rate of change and step (3) constantly
Resulting synchronization wind speed rate of change is compared, if described electric field power samples value rate of change is greater than synchronization wind speed rate of change, i.e. η
Wind, h＞ζ
Wind, h-1, judge that then this moment wind energy turbine set power samples value is the historical power input of bad wind energy turbine set data;
(5) for the historical power input of the wind energy turbine set data of the historical power input data of bad wind energy turbine set and disappearance, uses the wind energy turbine set performance number of identical or close wind speed correspondence as the correction value of this moment wind energy turbine set power samples value:
At first, set up a database table with wind speed corresponding power with it according to wind energy turbine set historical wind speed and power data;
For the historical power input of the wind energy turbine set data of the historical power input data of bad wind energy turbine set or disappearance, seek the interval at its corresponding wind speed place in database table;
The historical power correction value of wind energy turbine set to the historical power input data of bad wind energy turbine set or disappearance has two kinds to choose mode, a kind of for getting the mean value of described interval corresponding all power, another kind of for getting the value of probability of occurrence maximum in interval corresponding all performance numbers of wind speed.
Wind energy turbine set disclosed by the invention prediction input data judge and processing method in, preferably before the judgement of wind energy turbine set power prediction input data, set up a database table of a wind speed and its corresponding power.
In step (5), when the performance number of identical or close wind speed correspondence had several, correction value was got the mean value of several performance numbers.
In step (5), when the performance number of identical or close wind speed correspondence had several, correction value was taken out the performance number of existing probability maximum.
By in this method to the preliminary treatment of wind energy turbine set historical data, can improve the reliability for wind energy turbine set power prediction model training data sample set, thereby effectively improve the precision of wind energy turbine set power prediction.
Description of drawings
Figure 1 shows that wind power prediction history data pretreatment process figure.
Embodiment
Also in conjunction with specific embodiments the execution mode of technical scheme of the present invention is described in further details according to Figure of description below.
Be illustrated in figure 1 as wind power prediction history data pretreatment process figure, said method comprising the steps of:
(1) wind energy turbine set power is measured by the transformer station that is connect by the wind-powered electricity generation place, gathers the wind farm wind velocity value when measuring wind energy turbine set power, sets up wind speed-power data storehouse table according to the wind speed of gathering and corresponding performance number, and is then within the rule for the data of power disappearance;
(2) calculate the rate of change of the wind energy turbine set performance number that each wind energy turbine set performance number of constantly gathering gathers with respect to previous moment, with it as this wind energy turbine set power samples value rate of change, i.e. Δ P constantly
Wind, h=P
Wind, h-P
Wind, h-1, Δ P wherein
Wind, hFor wind energy turbine set in the h period power variation rate than the h-1 period, P
Wind, h-1Be the performance number of wind energy turbine set h-1 period, P
Wind, hPerformance number for the wind energy turbine set h period.
(3) calculate and to correspond to the wind speed in the corresponding moment in the step (2) with respect to the wind speed rate of change of previous moment wind speed,, i.e. Δ V
Wind, h=V
Wind, h-V
Wind, h-1, Δ V wherein
Wind, hFor wind energy turbine set in the h period wind speed rate of change than the h-1 period, V
Wind, h-1Be the air speed value of wind energy turbine set h-1 period, V
Wind, hAir speed value for the wind energy turbine set h period.
(4) with step (2) resulting wind energy turbine set power samples value rate of change and step (3) constantly
Resulting synchronization wind speed rate of change is compared, if described electric field power samples value rate of change is greater than synchronization wind speed rate of change, i.e. Δ P
Wind, h＞Δ V
Wind, h-1, judge that then this moment wind energy turbine set power samples value is the historical power input of bad wind energy turbine set data, and in wind speed-power data storehouse table mark in addition, when the subsequent correction performance number not with reference to this performance number;
(5) for the historical power input of the wind energy turbine set data of the historical power input data of bad wind energy turbine set and disappearance, use the wind energy turbine set performance number of identical or close wind speed correspondence as the correction value of this moment wind energy turbine set power samples value, concrete steps are as follows:
Step 1: according to the database table of wind energy turbine set historical wind speed and power data foundation, for example wind speed is at V
Wind, k～V
Wind, k+1The time correspondence performance number P is arranged
Wind, 1, P
Wind, 2... P
Wind, n
Step 2: for the historical power input of the wind energy turbine set of bad wind energy turbine set historical input data and disappearance data, seek its wind speed corresponding interval in database table;
Step 3: according to the corresponding interval of wind speed, there are two kinds to choose mode to wind power sampling correction value, a kind of for getting the mean value of interval corresponding all performance numbers of wind speed, another kind of for getting the value of probability of occurrence maximum in interval corresponding all performance numbers of wind speed.When the performance number of identical or close wind speed correspondence had several, correction value was got the mean value of several performance numbers; When the performance number of identical or close wind speed correspondence had several, correction value was taken out the performance number of existing probability maximum.
Step 4: amended wind power value increased or replace missing data or bad data in the historical data base, finally obtain revised historical data base.The embodiment that more than provides is in order to illustrate the present invention and its practical application, be not that the present invention is done any pro forma restriction, any one professional and technical personnel is in the scope that does not depart from technical solution of the present invention, and the above technology of foundation and method do certain modification and the equivalent embodiment that is considered as equivalent variations is worked as in change.
Claims (4)
1. judgement and the processing method of wind energy turbine set power prediction input data is characterized in that, judge normal data and bad data in the wind energy turbine set power prediction input data by described method, and bad data is wherein handled; Said method comprising the steps of:
(1) wind energy turbine set power is measured by the transformer station that is connect by the wind-powered electricity generation place, gathers the wind farm wind velocity value when measuring wind energy turbine set power;
(2) calculate the rate of change of the wind energy turbine set performance number that each wind energy turbine set performance number of constantly gathering gathers with respect to previous moment, with it as this wind energy turbine set power samples value rate of change, i.e. η constantly
Wind, h=(P
Wind, h-P
Wind, h-1)/P
Wind, h-1, η wherein
Wind, hFor wind energy turbine set at h constantly than h-1 wind energy turbine set power samples value rate of change constantly, P
Wind, h-1Be wind energy turbine set h-1 performance number constantly, P
Wind, hBe wind energy turbine set h performance number constantly;
(3) calculating corresponds to the wind speed in the corresponding moment in the step (2) with respect to the wind speed rate of change of previous moment wind speed, i.e. ξ
Wind, h=(V
Wind, h-V
Wind, h-1)/V
Wind, h-1, ξ wherein
Wind, hFor wind energy turbine set at h constantly than the wind speed rate of change of h-1 period, V
Wind, h-1Be wind energy turbine set h-1 air speed value constantly, V
Wind, hBe wind energy turbine set h air speed value constantly;
(4) with step (2) resulting wind energy turbine set power samples value rate of change and step (3) constantly
Resulting synchronization wind speed rate of change is compared, if described wind energy turbine set power samples value rate of change is greater than synchronization wind speed rate of change, i.e. η
Wind, hξ
Wind, h, judge that then this moment wind energy turbine set power samples value is the historical power input of bad wind energy turbine set data;
(5) for the historical power input of the wind energy turbine set data of the historical power input data of bad wind energy turbine set and disappearance, uses the wind energy turbine set performance number of identical or close wind speed correspondence as the correction value of this moment wind energy turbine set power samples value:
At first, set up the database table of a wind speed and its corresponding power according to wind energy turbine set historical wind speed and power data;
For the historical power input of the wind energy turbine set data of the historical power input data of bad wind energy turbine set or disappearance, seek the wind speed interval at its corresponding wind speed place in database table;
The historical power correction value of wind energy turbine set to the historical power input data of bad wind energy turbine set or disappearance has two kinds to choose mode, a kind of for getting the mean value of interval corresponding all power of described wind speed, another kind of for getting the value of probability of occurrence maximum in interval corresponding all performance numbers of wind speed.
2. wind energy turbine set power prediction according to claim 1 is imported judgement and the processing method of data, it is characterized in that:
Before the judgement of wind energy turbine set power prediction input data, preferentially set up a database table of a wind speed and its corresponding power.
3. wind energy turbine set power prediction according to claim 2 is imported judgement and the processing method of data, it is characterized in that:
When the performance number of identical or close wind speed correspondence had several, described correction value was got the mean value of several performance numbers.
4. wind energy turbine set power prediction according to claim 2 is imported judgement and the processing method of data, it is characterized in that:
When the performance number of identical or close wind speed correspondence had several, described correction value was taken out the performance number of existing probability maximum.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110422235 CN102496926B (en) | 2011-12-16 | 2011-12-16 | Method for judging and processing wind farm power prediction input data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110422235 CN102496926B (en) | 2011-12-16 | 2011-12-16 | Method for judging and processing wind farm power prediction input data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102496926A CN102496926A (en) | 2012-06-13 |
| CN102496926B true CN102496926B (en) | 2013-09-04 |
Family
ID=46188725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110422235 Active CN102496926B (en) | 2011-12-16 | 2011-12-16 | Method for judging and processing wind farm power prediction input data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102496926B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102855296A (en) * | 2012-08-14 | 2013-01-02 | 中国电力科学研究院 | Renewable energy power generation history data preprocessing method and system |
| CN103207948B (en) * | 2013-04-08 | 2016-01-20 | 同济大学 | Based on the wind energy turbine set anemometer wind speed missing data interpolating method of wind speed correlativity |
| CN103530508B (en) * | 2013-09-30 | 2017-01-11 | 国家电网公司 | Method for establishing wind speed-power conversion probability model |
| CN105205544B (en) * | 2014-06-24 | 2018-07-24 | 华北电力大学（保定） | A kind of wind power forecasting method based on dual random theory |
| CN104200067A (en) * | 2014-08-11 | 2014-12-10 | 国家电网公司 | Method and device for determining wind speed probability distribution and method for evaluating power of wind power system |
| CN105528735B (en) * | 2015-12-03 | 2019-05-31 | 甘肃省电力公司风电技术中心 | Bearing calibration based on the exceptional data point for measuring wind speed and spatial coherence |
| CN105512766A (en) * | 2015-12-11 | 2016-04-20 | 中能电力科技开发有限公司 | Wind power plant power predication method |
| CN107288673B (en) * | 2017-07-11 | 2019-07-23 | 广东工业大学 | A kind of vcehicular tunnel energy-saving ventilating air control method based on wind speed |
| CN108593968B (en) * | 2017-12-08 | 2020-09-15 | 北京金风科创风电设备有限公司 | Method and device for determining correction coefficient of anemometer |
| CN109101659B (en) * | 2018-09-03 | 2021-08-27 | 贵州电网有限责任公司 | Method for power data abnormity in small hydropower station data acquisition system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3668051B2 (en) * | 1999-06-04 | 2005-07-06 | ティーエム・ティーアンドディー株式会社 | System recognition method in charge / failure judgment and tidal current calculation |
| ES2428390T3 (en) * | 2005-05-13 | 2013-11-07 | Siemens Aktiengesellschaft | Power control system of a wind farm |
| CN102044872A (en) * | 2010-11-18 | 2011-05-04 | 西北电网有限公司 | Medium-long term forecasting method for wind power |
| CN102170130B (en) * | 2011-04-26 | 2013-03-06 | 华北电力大学 | Short-term wind power prediction method |
| CN102269124B (en) * | 2011-06-30 | 2013-04-03 | 内蒙古电力勘测设计院 | Ultra-short term wind power station generated power forecasting system |
-
2011
- 2011-12-16 CN CN 201110422235 patent/CN102496926B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN102496926A (en) | 2012-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102496926B (en) | Method for judging and processing wind farm power prediction input data | |
| CN103235981B (en) | A kind of wind power quality trend forecasting method | |
| CN102914969B (en) | Comprehensive error correction method of short-period wind power prediction system | |
| CN103631681B (en) | A kind of method of online reparation abnormal data of wind power plant | |
| CN101592538B (en) | Method for computing steady-state output power of wind power station based on actual measured data | |
| CN103679298A (en) | Method for evaluating wind power plant short-period power prediction precision | |
| CN103020743B (en) | Wind energy turbine set ultra-short term wind speed forecasting method | |
| CN103208037B (en) | A kind of power forecasting method based on on-line correction being applied to new energy power station | |
| CN101414317A (en) | Equivalent wind speed method for processing wind electric field static power equivalence dispersion problem | |
| CN113205210B (en) | Wind power plant wind speed and power prediction method, system and equipment for complex terrain and storage medium | |
| CN109167387A (en) | Wind field wind power forecasting method | |
| CN102254092B (en) | Dynamic equivalent method for large-scale wind power station with double-feed wind power set | |
| CN103996079A (en) | Wind power weighting predication method based on conditional probability | |
| CN113379142B (en) | Short-term wind power prediction method based on wind speed correction and fusion model | |
| CN103500370B (en) | Dynamic wind power plant wind direction coordinate precomputation system method for building up | |
| CN106650977A (en) | Short-term power prediction method used for newly-built wind farm | |
| CN102738792B (en) | Wind power predicting method | |
| CN102797631A (en) | Method, system and device for carrying out online self-correcting on optimal gain of wind generating set | |
| CN106611243A (en) | Residual correction method for wind speed prediction based on GARCH (Generalized ARCH) model | |
| CN105046349A (en) | Wind power prediction method considering wake effect | |
| CN105205564A (en) | Wind power plant wind curtailment electric quantity statistical system and method based on anemometer tower neural network | |
| CN109190845B (en) | Two-stage dynamic optimization short-term wind power prediction method | |
| CN105741192B (en) | Short-term wind speed combined forecasting method for wind turbine engine room of wind power plant | |
| CN103699804A (en) | Method for monitoring power output deviation in real time during running of wind turbines | |
| CN105046383A (en) | Real-time wind power predicting method based on ensemble empirical mode decomposition and relevant vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |