CN102945508B - Model correction based wind power forecasting method - Google Patents
Model correction based wind power forecasting method Download PDFInfo
- Publication number
- CN102945508B CN102945508B CN201210391408.7A CN201210391408A CN102945508B CN 102945508 B CN102945508 B CN 102945508B CN 201210391408 A CN201210391408 A CN 201210391408A CN 102945508 B CN102945508 B CN 102945508B
- Authority
- CN
- China
- Prior art keywords
- wind
- power
- prediction
- correct
- production capacity
- 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
- 238000012937 correction Methods 0.000 title abstract description 12
- 238000013277 forecasting method Methods 0.000 title abstract 2
- 238000000034 method Methods 0.000 claims abstract description 63
- 238000004088 simulation Methods 0.000 claims abstract description 17
- 238000012360 testing method Methods 0.000 claims abstract description 14
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000004519 manufacturing process Methods 0.000 claims description 58
- 230000005611 electricity Effects 0.000 claims description 22
- 230000009467 reduction Effects 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 13
- 238000006467 substitution reaction Methods 0.000 claims description 12
- 238000013461 design Methods 0.000 claims description 7
- 230000001915 proofreading effect Effects 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 description 6
- 238000005259 measurement Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000005855 radiation Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 238000013213 extrapolation Methods 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 208000033182 PLCG2-associated antibody deficiency and immune dysregulation Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 238000007598 dipping method Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- 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
- Y02A30/00—Adapting or protecting infrastructure or their operation
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to a model correction based wind power forecasting method, which comprises the following steps: firstly correcting a simulation wind speed forecasted by a numerical value weather forecast model to a forecast wind regime at an anemometer tower point, then correcting the forecast wind regime at the anemometer tower point to the forecast wind regime at a fan point, obtaining a wind power plant forecast theoretical power according to the forecast wind regime at the fan point and a corrected wind power forecast model, combining a result corrected by a wind power discounting coefficient to obtain an actual power forecasted by a wind power plant. A wind power forecast model correction flow comprises the following steps: firstly carrying out a fan cabin power curve test and power coefficient calculation through a fan theoretical power curve, then substituting the forecast wind regime at the fan point into the fan cabin power curve and combining the power coefficient to obtain the actual fan power. The method is adopted to reduce various model errors and ensure the accuracy rate of forecasting results.
Description
Technical field
The present invention relates to a kind of wind power prediction methods, relate in particular to a kind of wind power prediction methods based on model tuning.
Background technology
In the period of rising rapidly in the newly-increased wind energy turbine set installed capacity of China, the development of Wind Power Generation Industry has caused difficulty to power grid security and scheduling.In order to meet electrical network needs, improve wind-powered electricity generation utilization ratio, national correlation department has been put into effect series of standards and requirement, and wind power forecast system is imperative installing and using of each large-scale wind field.
At present, China's Wind Power Generation Industry development time is shorter, and the wind power prediction method that industry is main and forecast system are all in to be groped and elementary developing stage.According to the regulation of national correlation department, the examination of existing wind power forecast system mainly from accuracy rate, report 3 indexs of rate and qualification rate to weigh.Accuracy rate wherein and qualification rate depend on the error of forecast power data and actual power data.The flow process of industry main flow wind power forecast relates to numerical value weather simulation, wind field anemometer tower data prediction, several key links such as wind energy turbine set capability forecasting.The Output rusults of above-mentioned each link can be as the input parameter of its next link, and the model error of any one link will be brought in next link and go.This wherein several main model errors have numerical value weather simulation error, anemometer tower, blower fan wind speed error in dipping, anemometer tower data and numerical weather forecast data correlation analytical error, it is representative that anemometer tower is surveyed wind data, power of fan curve deviation, wind field production capacity reduction evaluated error etc.
Wind power prediction is comparatively complicated physical analogy process.The model parameter of the main flow wind power forecasting system of industry is generally nonadjustable, so it predicts the outcome, accuracy rate is difficult to guarantee with project change.
Several subject matter that exists of industry wind power forecast system has at present:
1, directly adopt representative anemometer tower prediction data to predict whole wind energy turbine set power: the measurement data such as wind speed, wind direction, temperature, air pressure of adopting hypotheses in this way and be anemometer tower will have suitable representativeness, can represent the data such as wind speed, wind direction of the point of each blower fan of wind field; In complex-terrain wind energy turbine set, anemometer tower wind regime can not replace the wind regime of all blower fan points, if still directly adopt in this case anemometer tower prediction data prediction wind energy turbine set power, its result will differ far away with actual conditions.
2, adopt the assurance wind-powered electricity generation unit power curve prediction wind energy turbine set power of producer: in actual wind field, actual power curve and the theoretical power (horse-power) curve of blower fan there are differences, this is mainly because the operating mode under the fan condition of actual wind field and theoretical test condition exists larger difference, in addition, the nacelle wind speed power curve (NACP) of blower fan and actual power curve be difference to some extent also, the power curve that in the industry cycle directly adopts producer to provide in part wind power forecast system is carried out prediction, and its result precision is difficult to guarantee.
3, the every reduction coefficient of wind energy turbine set is estimated to rely on engineering experience: in a specific wind field, objective reality reduction coefficient, mainly comprises: unit availability, power curve, envirment factor etc.
Summary of the invention
Technical problem to be solved by this invention is to provide a kind of wind power prediction methods based on model tuning, can effectively reduce above-mentioned data that anemometer tower prediction wind regime data, wind power curve, the every reduction coefficient of wind energy turbine set the cause error in various model application processes.
The technical scheme that the present invention solves the problems of the technologies described above is as follows: a kind of wind power prediction methods based on model tuning,
First the simulation wind speed of numerical weather prediction model forecast is corrected to the prediction wind regime of anemometer tower point;
Then the prediction wind regime of being put by anemometer tower is proofreaied and correct the prediction wind regime to blower fan point, the prediction wind regime of blower fan point and calibrated wind power forecast model obtain wind energy turbine set prediction theory power, then the result after proofreading and correct in conjunction with wind energy turbine set reduction coefficient draws the actual power of wind energy turbine set prediction;
The flow process that described wind power forecast model is proofreaied and correct is: first by blower fan theoretical power (horse-power) curve, carry out the test of fan engine room power curve and power coefficient calculates, then will in the prediction wind regime substitution fan engine room power curve of blower fan point, in conjunction with power coefficient, draw actual power of fan.
A kind of wind power prediction methods based on model tuning, first the numerical value Meteorological Forecast Model that utilizes meteorological department to provide, the weather condition of wind energy turbine set (mainly comprising the parameters such as wind speed, wind direction, temperature, air pressure, humidity) is predicted, set up real-time prediction model, the predicted value of numerical value Meteorological Forecast Model is changed into the power stage of wind energy turbine set.Particularly, by WRF(weather forecast pattern) the simulation wind speed MOS(MOS method of forecast) be corrected to the prediction wind regime of anemometer tower point, the prediction wind regime of anemometer tower point is predicted by anemometer tower, then the prediction wind regime of being put by anemometer tower is proofreaied and correct the prediction wind regime to blower fan point, the prediction wind regime of blower fan point and calibrated wind power forecast model are obtained to wind energy turbine set prediction theory power, then the result after proofreading and correct in conjunction with wind energy turbine set reduction coefficient draws the actual power of wind energy turbine set prediction.
The prediction wind regime of described blower fan point is proofreaied and correct and is utilized anemometer tower point prediction wind regime by MCP(, first to measure associated last prediction with blower fan historical data in conjunction with anemometer tower again) method foundation association, anemometer tower place prediction wind regime is corrected to blower fan site estimation wind regime, MCP method is: the supposition anemometer tower same period and the relation of blower fan data can be by mathematical simulations, for example linear and nonlinear regression, variance ratio method.
The wind power forecast model of described correction is actual wind power curve, the main flow process that described wind power forecast model is proofreaied and correct is: first by blower fan theoretical power (horse-power) curve, carry out the test of fan engine room power curve and power coefficient calculates, then will in the prediction wind regime substitution fan engine room power curve of blower fan point, in conjunction with power coefficient, draw actual power of fan, described power coefficient adopts back the production capacity of the historical fan engine room effective wind speed data substitution theoretical power (horse-power) curve of calculation and the difference of actual production capacity to calculate, wherein, the principle that in effective wind speed database, data are rejected mainly contains: 1, the external conditions such as wind speed exceed the working range of blower fan, 2, external condition exceeds the working range of testing equipment, 3, blower fan be could not get on to the Net, 4 blower fan power outputs are received as the restriction of the external conditions such as electrical network, 5, testing equipment lost efficacy or performance reduces, 6, 10min mean wind speed has exceeded beyond the sector of measuring, 7, wind speed has exceeded the zone of reasonableness of nacelle wind speed transfer function.
Described wind energy turbine set reduction coefficient is proofreaied and correct and is comprised that availability is proofreaied and correct, electrical loss is proofreaied and correct, environmental impact loss is proofreaied and correct;
Described availability is proofreaied and correct and is comprised that time availability is proofreaied and correct and production capacity availability is proofreaied and correct, it is by analyzing in blower fan data system historical logout information that described time availability is proofreaied and correct, the time availability of proofreading and correct each blower fan in conjunction with the shared time of different operating modes, it is defined as: the percentage of wind-powered electricity generation unit running time under nominal situation in one period; Described production capacity availability is proofreaied and correct the potential production capacity that comprises assessment wind-powered electricity generation unit, the actual production capacity of calculating wind-powered electricity generation unit, utilize fan engine room wind speed through being converted to after unconfined flow leeward wind speed, the potential production capacity of the actual power curve calculation wind-powered electricity generation unit in the test of substitution power curve, production capacity can be utilized to calculate and with equation expression be:
Wherein, P
availfor production capacity availability, P
potentialfor potential production capacity, P
actualfor actual production capacity, the fan operation state information providing in conjunction with blower fan data monitoring system, the theory based on information classification, carries out assessment to the blower fan potential production capacity under different conditions and actual production capacity, and then calculates the Seasonal fluctuation of availability;
Described electrical loss is proofreaied and correct and is adopted the electric weight that obtains wind energy turbine set port of export electric energy metering machine, the gross generation of calculating wind energy turbine set wind-powered electricity generation unit, the ratio of the gross generation of the electric weight of wind energy turbine set port of export electric energy metering machine and wind energy turbine set wind-powered electricity generation unit is the reduction of electrical loss;
Described environmental impact be lost in production capacity availability cannot the situation of accurate Calculation under as the supplementary item of the reduction factor, mainly due to external environment condition (temperature, wind speed Hysteresis) etc. the production capacity of losing when parameter exceeds fan design index, environmental impact loss is proofreaied and correct and is first supposed that blower fan calculates production capacity under the condition that does not exceed environmental limitations, then calculates the production capacity of losing when external environment condition parameter exceeds fan design index; The rule that described environmental impact loss is proofreaied and correct is:
1,, when wind speed is greater than cut-out wind speed, replacing all production capacity data is 0;
2, when observed temperature is greater than while cutting out high temperature, replacing all production capacity data is 0, and when data are before 0 and observed temperature data while being greater than incision high temperature, replacing all production capacity data is 0;
3, when observed temperature is when cutting out low temperature, replacing all production capacity data is 0, when data be before 0 and observed temperature data be less than while again cutting low temperature, replacing all production capacity data is 0.
The invention has the beneficial effects as follows: adopt a kind of wind power prediction system based on model tuning and method effectively to reduce the error in various model application processes that anemometer tower prediction wind regime data, wind power curve, the every reduction coefficient of wind energy turbine set cause, guaranteed the accuracy rate of forecast result.
Accompanying drawing explanation
Fig. 1 is the Physical architecture figure that the present invention is based on the wind power prediction system of model tuning;
Fig. 2 is the wind power prediction methods flow chart that the present invention is based on model tuning;
Fig. 3 is the blower fan point prediction wind regime method flow diagram of model tuning anemometer tower prediction wind regime of the present invention;
Fig. 4 is the flow chart of wind-powered electricity generation unit theoretical power (horse-power) curve prediction of the present invention model tuning;
Fig. 5 is the wind power prediction flow chart that the present invention is based on model tuning;
Fig. 6 is the autoregressive coefficient of the statistical extrapolation training sample of ultra-short term forecast of the present invention.
Embodiment
Below in conjunction with accompanying drawing, principle of the present invention and feature are described, example, only for explaining the present invention, is not intended to limit scope of the present invention.
As shown in Figure 1, Fig. 1 is a kind of topological structure of the wind power prediction system based on model tuning, comprise numerical weather forecast server, blower fan, anemometer tower, wind power server, database server, wind power forecast system client, described wind power forecast server by Internet or special line respectively with numerical weather forecast server, blower fan, anemometer tower is connected, described data server is connected with wind power forecast server, described wind power forecast system client is connected with described database server, wind power forecast server obtains the numerical weather forecast data of numerical weather forecast service system by Internet or private-line mode, the meteorological data of anemometer tower, the communication with dispatch instructions of rationing the power supply that the real time execution information of blower fan and electrical network are assigned is forecast wind power, wind power forecast server stores forecast result on database server into and in real time forecast result is adopted national grid E form to upload to Ⅱ/Ⅲ district, power grid security district.
Fig. 2 is the wind power prediction methods flow chart based on model tuning, first the simulation wind speed MOS of WRF forecast is corrected to the prediction wind regime of anemometer tower point, then the prediction wind regime of being put by anemometer tower is proofreaied and correct the prediction wind regime to blower fan point, the calibrated wind power forecast model of the prediction wind regime substitution of blower fan point is obtained to wind energy turbine set prediction theory power, result after proofreading and correct in conjunction with wind energy turbine set reduction coefficient again, calculates the actual power that wind energy turbine set is predicted.Wherein, the flow process that is evaluated as the design of this project innovation point of the prediction wind regime of blower fan point, actual power of fan curve, the every reduction coefficient of wind energy turbine set.
The prediction wind regime method of blower fan point as shown in Figure 3, this figure is the flow process of the wind speed and direction association of setting up by MCP method, according to landform, roughness of ground surface, the wind regime of wind field design phase, utilize anemometer tower point prediction wind regime to set up by MCP in conjunction with anemometer tower and blower fan historical data associated, anemometer tower place prediction wind regime is corrected to blower fan point prediction wind regime.
MCP method brief introduction: the relation of the supposition anemometer tower same period and blower fan data can be by mathematical simulation, such as linear and nonlinear regression, variance ratio method etc.
1. linearity and nonlinear regression: the data of the anemometer tower same period of take are independent variable, and the same period, the data of blower fan were dependent variable, set up the regression relation of the two, and linear regression is simple and ease for operation with it, and suitable precision and generally being adopted.
2, Variance ratio method just carried out a processing that variance is equal, and the method has directly provided the expression formula of slope and intercept, convenient and simple, reasonable.
Wherein, ux, the anemometer tower that uy is the same period and blower fan point data set
Fig. 4 is the main flow chart of wind-powered electricity generation unit theoretical power (horse-power) curve prediction model tuning, first by blower fan theoretical power (horse-power) curve, carry out fan engine room power curve test and power coefficient calculates, then by drawing actual power of fan performance in conjunction with power coefficient in the prediction wind regime substitution fan engine room power curve of blower fan point, finally draw wind energy turbine set predicted power.Power curve testing process can be with reference to IEC series standard, and power coefficient calculates and adopts back the production capacity of the historical cabin effective wind speed data substitution theoretical power (horse-power) curve of calculation and the difference of actual production capacity to calculate.
Wherein, the principle that in effective wind speed database, data are rejected mainly contains:
(1) external condition such as wind speed exceeds the working range of blower fan,
(2) external condition exceeds the working range of testing equipment,
(3) blower fan be could not get on to the Net,
(4) blower fan power output is received as the restriction of the external conditions such as electrical network,
(5) testing equipment lost efficacy or performance reduction,
(6) 10min mean wind speed has exceeded beyond the sector of measuring,
(7) wind speed has exceeded the zone of reasonableness of nacelle wind speed transfer function.
The correction of the every coefficient of wind energy turbine set: mainly comprised availability, electrical loss, environmental impact loss.
Availability estimation has comprised time availability and two evaluation methods of production capacity availability, wherein:
Time availability is calculated by analyzing in blower fan data system historical logout information, the time availability that can proofread and correct each blower fan in conjunction with the shared time of different operating modes.It is defined as: the percentage of wind-powered electricity generation unit running time under nominal situation in one period.The correction of production capacity availability mainly comprises: the potential production capacity of assessment blower motor group, the actual production capacity of calculating wind-powered electricity generation unit.Conventionally the method adopting is to utilize fan engine room wind speed through being converted to after unconfined flow leeward wind speed, the potential production capacity of the actual power curve calculation wind-powered electricity generation unit in the test of substitution power curve, and production capacity can be utilized to calculate and with equation expression be:
Wherein, P
potentialfor potential production capacity; P
actualfor actual production capacity.
The fan operation state information providing in conjunction with blower fan data monitoring system, the theory based on information classification, carries out assessment to the blower fan potential production capacity under different conditions and actual production capacity, and then calculates the Seasonal fluctuation of availability.
Electrical loss is proofreaied and correct and is adopted the electric weight that obtains wind energy turbine set port of export electric energy metering machine, calculates the gross generation of wind energy turbine set wind-powered electricity generation unit, and the ratio of the gross generation of the electric weight of wind energy turbine set port of export electric energy metering machine and wind energy turbine set wind-powered electricity generation unit is the reduction of electrical loss;
Environmental impact be lost in production capacity availability cannot the situation of accurate Calculation under as the supplementary item of the reduction factor, mainly due to external environment condition (temperature, wind speed) etc. the production capacity of losing when parameter exceeds fan design index, environmental impact loss is proofreaied and correct and is first supposed that blower fan calculates production capacity under the condition that does not exceed environmental limitations, then calculates the production capacity of losing when external environment condition parameter exceeds fan design index; The rule that described environmental impact loss is proofreaied and correct is:
(1), when wind speed is greater than cut-out wind speed, replacing all production capacity data is 0;
(2), when observed temperature is greater than while cutting out high temperature, replacing all production capacity data is 0, when data are before 0 and observed temperature data while being greater than incision high temperature, replacing all production capacity data is 0;
(3), when observed temperature is when cutting out low temperature, replacing all production capacity data is 0, when data be before 0 and observed temperature data be less than while again cutting low temperature, replacing all production capacity data is 0.
Wind power prediction methods based on model tuning, this project adopts WRF (Weather Research Forecast) pattern to simulate, this modular system has many characteristics such as portable, easy care, extendible, effective and convenient, and will have more advanced numerical computations and Data Assimilation technology, moving multi nested grid performance and more perfect physical process (especially convection current and mesoscale Precipitation Process).It will contribute to carry out for numerical Simulation of High Resolution dissimilar, different geographical synoptic process, improves resolution and the accuracy of weather forecast, makes new scientific achievement apply to limited area operational forecasting model more convenient.
What this research was selected is WRFV3 version, adopts one deck grid, and mode top is positioned at: 31.0 ° of N, 112.5 ° of E.It is 201 * 182 that HORIZONTAL PLAID is counted, and horizontal resolution is 15km; Vertical direction has 35 layers; Time step is 60s.(can for different survey region adjustment) main physical process: WRF Single-Moment6-class Microphysical scheme, the thermal diffusion scheme of Grell-Devenyi ensemble cumulus parameterization scheme, RRTM long-wave radiation scheme, Dudhia shortwave radiation scheme, ground layer scheme, four layers of soil, MRF boundary layer scheme etc.
In order to integrate with practical business, adopt NCEP forecast fields GFS data as WRF pattern initial fields and boundary condition, pattern is from 08 o'clock every day (during Beijing, lower same) started to report, integration 84 hours, by a hour output analog physical amount (as: wind speed, temperature, air pressure, specific humidity, meridional wind, zonal wind, cloud amount etc.).Other physical quantity of pattern output can be for the statistical correction of solar radiation model predictions output.
Fig. 5 is the wind power prediction system flow chart based on model tuning, in figure, the plan of rationing the power supply of blower fan data, numerical value weather data, real-time anemometer tower data and national grid in real time all deposits database in, through wind-powered electricity generation forecast process system, carry out short-period forecast and ultra-short term forecast, then forecast structure is uploaded to national grid, then short-period forecast and ultra-short term forecast result are fed back to database.
1, short-term wind-electricity power prediction
Short-term wind-electricity power is predicted as the wind power output power prediction in following 3 days, and temporal resolution is 15min.According to the difference of Forecasting Methodology and condition of compatibility, substantially contain the various situations that the forecast of wind energy turbine set wind power may run into.
1.1, former logos
Former logos is first wind field wind speed to be predicted, then through and historical data bring actual wind power curve into, finally obtain power prediction value.This method can not need a large amount of, long-term measured data, is more suitable for complex-terrain.
According to logarithm value simulation wind speed, whether proofread and correct and the difference of wind power forecasting model method for building up divides following methods to predict.
The actual power of fan curve of substitution matching after WRF forecast wind speed MOS proofreaies and correct
Adopt wind field historical actual measurement air speed data and wind power data to set up wind power forecast model, the simulation wind speed of WRF forecast is corrected to after anemometer tower, then is proofreaied and correct to every typhoon group of motors by anemometer tower wind speed.Bring prediction of wind speed into wind power forecast model, obtain prediction theory power, in the result in conjunction with after every reduction coefficient is proofreaied and correct, calculate the actual power of wind energy turbine set prediction.This method is applicable to wind energy turbine set anemometer tower data, also has the situation of historical wind power data, the main flow chart that Fig. 2 is this method.
The model tuning of fixed effect
The wind power model tuning that adopts the historical numerical simulation data of a year and a day and corresponding wind power data to divide moon or set up fixed effect season, and these correction parameters are preserved.
Wherein anemometer tower point and arbitrarily the MCP of blower fan point to be associated in certain period be comparatively fixing mapping relations.The power characteristic of blower fan and the correction coefficient of every reduction are also and have relation comparatively closely season.
(1) set up data and the method for simulation wind speed calibration model
(2) set up data and the method for wind power forecast model
Modeling data: wind energy turbine set actual measurement Wind Data, wind power data and the fan operation situation data of moving a year and a day.Actual unit wind power data need be with whole wind field 15min wind power divided by start number of units.
Wind-powered electricity generation unit actual power curve modeling method: adopt the bin method of IEC recommendation and the power characteristic measuring and calculation power curve based on nacelle wind speed meter.
1.2 power statistic laws
Power statistic law is exactly to set up a kind of mapping relations between input (measurement data of numerical value Meteorological Forecast Model, wind energy turbine set etc.) in system and the power of wind energy turbine set, comprises linear and non-linear method, specifically has autoregression technology, neural net etc.
The advantage of this method is to predict spontaneously to adapt to wind energy turbine set position, so systematic error has reduced automatically.Shortcoming is need long-term measurement data and extra training and ignored whole wind energy to the physical essence of electric energy conversion, under extreme weather conditions, system is difficult to Accurate Prediction, is very important, otherwise will causes very large predicated error to the correction prediction of these rare weather conditions.
Rolling modeling
Adopt forecast a few days ago certain period numerical simulation data/anemometer tower survey wind data and roll every day and to set up wind power forecast model with corresponding wind power data, bring data into wind power forecast model, just can obtain predicted power.This method is applicable to the situation that wind energy turbine set has nearly two months wind power data.
Modeling data: forecast ought push away the numerical simulation data of 30 days a few days ago, comprises wind speed (70m), wind direction sinusoidal (70m), wind direction cosine (70m), temperature (2m), surface pressure, humidity (2m); And corresponding wind power and fan operation situation data.
Mathematical Modeling Methods: multiple linear regression, neural net method
1.3 continue method
When numerical simulation weather forecast data lacks report, in the situation that above method all cannot normally be moved, for guaranteeing that the prediction of wind energy turbine set wind power reports rate, can adopt the method for continuing to carry out wind power forecast, use the live wind power result of proxima luce (prox. luc) as forecast result on the same day, this kind of situation can only reach on the basis that not late report fails to report as far as possible, and artificial reference weather forecast is carried out experiential modification to improve forecast accuracy.
2 ultra-short term predictions
2.1 real time corrections based on short-term forecast power
Based on short-term wind-electricity power forecast result, utilize the live wind power data of real-time update, short-term wind-electricity power forecast result is carried out to real time correction.
Set up model:
In formula, i is that ultra-short term gives the correct time time in advance, Y
itime ultra-short term power prediction during for i, F
itime short term power prediction during for i,
for live wind power mean value of the past period,
for the past period short term power predicted mean vote.
Model adopt by time time (15 minutes) roll and set up,
the arithmetic mean value of all getting in 2 hours.
When short-period forecast result and live power exist certain systematic bias, the method effect is more satisfactory.
2.2 statistical extrapolation
Utilize 200 left and right samples in the moment recently, set up respectively 16 auto-regressive equations, predictor is respectively the after-power factor lagging behind 15,30,45,240 minutes.
Auto-regressive equation: Y
t=a
iy
t-i+ b
i, in formula, Y is live power, inferior when i is hysteresis, by training sample Coefficient of determination a, b, as shown in Figure 6.For improving forecast accuracy, equation coefficient should obtain by each training of rolling.
The foregoing is only preferred embodiment of the present invention, in order to limit the present invention, within the spirit and principles in the present invention not all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (4)
1. the wind power prediction methods based on model tuning, is characterized in that:
First the simulation wind speed of numerical weather prediction model forecast is corrected to the prediction wind regime of anemometer tower point;
Then the prediction wind regime of being put by anemometer tower is proofreaied and correct the prediction wind regime to blower fan point, the prediction wind regime of blower fan point and calibrated wind power forecast model obtain wind energy turbine set prediction theory power, then the result after proofreading and correct in conjunction with wind energy turbine set reduction coefficient draws the actual power of wind energy turbine set prediction;
The flow process that described wind power forecast model is proofreaied and correct is: first by blower fan theoretical power (horse-power) curve, carry out the test of fan engine room power curve and power coefficient calculates, then will in the prediction wind regime substitution fan engine room power curve of blower fan point, in conjunction with power coefficient, draw actual power of fan.
2. a kind of wind power prediction methods based on model tuning according to claim 1, it is characterized in that: described anemometer tower point prediction wind regime is proofreaied and correct and utilized anemometer tower point prediction wind regime, anemometer tower historical data and blower fan historical data to set up the association between wind speed and direction by first measuring the method for associated last prediction again, and anemometer tower point prediction wind regime is corrected to blower fan point prediction wind regime.
3. a kind of wind power prediction methods based on model tuning according to claim 1 and 2, is characterized in that: described power coefficient adopts back the production capacity of the historical fan engine room effective wind speed data substitution theoretical power (horse-power) curve of calculation and the difference of actual production capacity to calculate.
4. a kind of wind power prediction methods based on model tuning according to claim 1 and 2, is characterized in that: described wind energy turbine set reduction coefficient is proofreaied and correct and comprised that availability is proofreaied and correct, electrical loss is proofreaied and correct, environmental impact loss is proofreaied and correct;
Described availability is proofreaied and correct and is comprised that time availability is proofreaied and correct and production capacity availability is proofreaied and correct, it is by analyzing in blower fan data system historical logout information, proofreading and correct the time availability of each blower fan in conjunction with the shared time of different operating modes that described time availability is proofreaied and correct; Described production capacity availability is proofreaied and correct the potential production capacity that comprises assessment wind-powered electricity generation unit, the actual poor energy that calculates wind-powered electricity generation unit, utilize fan engine room wind speed through being converted to after unconfined flow leeward wind speed, the potential production capacity of the actual power curve calculation wind-powered electricity generation unit in the test of substitution power curve, production capacity availability is calculated and with equation expression is:
Wherein, P
availfor production capacity availability, P
potentialfor potential production capacity, P
actualfor actual production capacity, the fan operation state information providing in conjunction with fan monitoring system, the theory based on information classification, carries out assessment to the blower fan potential production capacity under different conditions and actual production capacity, and then calculates the Seasonal fluctuation of availability;
Described electrical loss is proofreaied and correct and is adopted the electric weight that obtains wind energy turbine set port of export electric energy metering machine, the gross generation of calculating wind energy turbine set wind-powered electricity generation unit, the ratio of the gross generation of the electric weight of wind energy turbine set port of export electric energy metering machine and wind energy turbine set wind-powered electricity generation unit is the reduction of electrical loss;
Described environmental impact loss is proofreaied and correct and is first supposed that blower fan calculates production capacity under the condition that does not exceed environmental limitations, then calculates the production capacity of losing when external environment condition parameter exceeds fan design index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210391408.7A CN102945508B (en) | 2012-10-15 | 2012-10-15 | Model correction based wind power forecasting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210391408.7A CN102945508B (en) | 2012-10-15 | 2012-10-15 | Model correction based wind power forecasting method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102945508A CN102945508A (en) | 2013-02-27 |
| CN102945508B true CN102945508B (en) | 2014-01-29 |
Family
ID=47728447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210391408.7A Active CN102945508B (en) | 2012-10-15 | 2012-10-15 | Model correction based wind power forecasting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102945508B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021424B (en) * | 2013-02-28 | 2018-12-07 | 乌托巴斯洞察公司 | Method and apparatus for predicting the output power of the blower in wind field |
| CN103984986B (en) * | 2014-05-06 | 2018-04-27 | 国家电网公司 | The self study arma modeling ultrashort-term wind power prediction method of real time correction |
| CN104008284A (en) * | 2014-05-20 | 2014-08-27 | 国家电网公司 | Correcting method for anemometer tower in numerical weather prediction |
| CN104200280B (en) * | 2014-08-22 | 2017-09-22 | 北方大贤风电科技（北京）有限公司 | A kind of method and system for wind power prediction |
| CN106875037B (en) * | 2016-12-30 | 2020-11-10 | 东软集团股份有限公司 | Wind power prediction method and device |
| CN106875038B (en) * | 2017-01-03 | 2021-09-28 | 国能日新科技股份有限公司 | Wind power prediction method and device based on different climate characteristics of multiple points in integrated local area |
| CN107153894B (en) | 2017-06-02 | 2018-11-27 | 北京金风科创风电设备有限公司 | Method and device for correcting predicted wind speed of wind power plant |
| CN108345996B (en) * | 2018-02-06 | 2021-07-20 | 北京天润新能投资有限公司 | System and method for reducing wind power assessment electric quantity |
| CN108197843B (en) * | 2018-02-26 | 2020-11-06 | 中国电建集团西北勘测设计研究院有限公司 | Wind power output characteristic evaluation method for flat terrain |
| CN111325376A (en) * | 2018-12-14 | 2020-06-23 | 北京金风科创风电设备有限公司 | Wind speed prediction method and device |
| CN110705769B (en) * | 2019-09-26 | 2022-07-05 | 国家电网公司华北分部 | New energy power generation power prediction optimization method and device |
| CN110705772B (en) * | 2019-09-26 | 2022-08-02 | 国家电网公司华北分部 | Regional power grid wind power generation power prediction optimization method and device |
| CN111639437B (en) * | 2020-06-08 | 2021-03-23 | 中国水利水电科学研究院 | Method for dynamically changing WRF mode parameterization scheme combination based on ground air pressure distribution situation |
| CN111950764B (en) * | 2020-07-03 | 2024-03-22 | 国网冀北电力有限公司 | Wind power prediction correction method for power grid under extreme weather conditions |
| CN112761896B (en) * | 2020-09-24 | 2024-05-14 | 国网内蒙古东部电力有限公司 | Calculation method, device and computer equipment for improving prediction accuracy of power generation amount of wind power station |
| CN112700134A (en) * | 2020-12-30 | 2021-04-23 | 华润电力技术研究院有限公司 | Method, system and equipment for wind measuring tower representativeness analysis |
| CN113869604A (en) * | 2021-10-25 | 2021-12-31 | 山东大学 | Wind power prediction method and system based on WRF wind speed prediction |
| CN114791749B (en) * | 2021-12-27 | 2023-05-05 | 中国船舶工业综合技术经济研究院 | Isolation airtight capsule cabin for simulating and experiencing local weather conditions |
| CN117394306B (en) * | 2023-09-19 | 2024-06-14 | 华中科技大学 | Wind power prediction model establishment method based on new energy grid connection and application thereof |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
-
2012
- 2012-10-15 CN CN201210391408.7A patent/CN102945508B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102945508A (en) | 2013-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102945508B (en) | Model correction based wind power forecasting method | |
| Stathopoulos et al. | Wind power prediction based on numerical and statistical models | |
| Kubik et al. | Exploring the role of reanalysis data in simulating regional wind generation variability over Northern Ireland | |
| CN109726867B (en) | High-resolution electric power weather forecasting method based on multi-mode set | |
| CN109784563B (en) | Ultra-short-term power prediction method based on virtual anemometer tower technology | |
| CN108388962A (en) | A kind of wind power forecasting system and method | |
| CN113496311A (en) | Photovoltaic power station generated power prediction method and system | |
| CN110705771B (en) | Method and device for predicting and optimizing generation power of new energy of regional power grid | |
| CN109978242A (en) | The photovoltaic power generation cluster power forecasting method and device of scale are risen based on statistics | |
| CN101794996A (en) | Real-time predicting method for output of wind electric field | |
| CN111458769B (en) | Method and system for predicting environmental meteorological data of power transmission line | |
| De Araujo | WRF wind speed simulation and SAM wind energy estimation: A case study in Dili Timor Leste | |
| Tan et al. | Short term wind energy resource prediction using WRF model for a location in western part of Turkey | |
| Lopes et al. | Improved ECMWF forecasts of direct normal irradiance: A tool for better operational strategies in concentrating solar power plants | |
| CN104008284A (en) | Correcting method for anemometer tower in numerical weather prediction | |
| CN117117819A (en) | Photovoltaic power generation short-term power prediction method, system, equipment and medium | |
| Cabezon et al. | Comparison of methods for power curve modelling | |
| CN110416996A (en) | A kind of distributed photovoltaic cluster power prediction system based on cloud platform | |
| CN106355264B (en) | Reference crop evapotranspiration combined prediction method | |
| Cheggaga et al. | A neural network solution for extrapolation of wind speeds at heights ranging for improving the estimation of wind producible | |
| CN116976518A (en) | Wind power prediction method and device for centralized wind power plant | |
| Cao et al. | Development of short-term wind power forecasting methods | |
| CN115912334A (en) | Method for establishing prediction model of output guarantee rate of wind power plant and prediction method | |
| CN112761896B (en) | Calculation method, device and computer equipment for improving prediction accuracy of power generation amount of wind power station | |
| CN108345996A (en) | A kind of system and method reducing wind power checking energy |
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 | ||
| C56 | Change in the name or address of the patentee | ||
| CP03 | Change of name, title or address |
Address after: 15, building 430206, building A, overseas talent building, future science and Technology City, East Lake Development Zone, Wuhan, Hubei Patentee after: WINDMAGICS (WUHAN) CO.,LTD. Address before: 430073 Hubei city of Wuhan province East Lake high tech Development Zone, Optics Valley Software Park Avenue, block E3 room 1503 Patentee before: WINDMAGICS (WUHAN) RENEWABLE ENERGY TECHNOLOGY CO.,LTD. |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A kind of wind power prediction methods based on model tuning Effective date of registration: 20170217 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: 2017420000005 |
|
| PLDC | Enforcement, change and cancellation of contracts on pledge of patent right or utility model | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20180413 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: 2017420000005 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: Model correction based wind power forecasting method Effective date of registration: 20180423 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: 2018420000019 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20190412 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: 2018420000019 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A wind power forecasting method based on model correction Effective date of registration: 20210329 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) Co.,Ltd. Registration number: Y2021420000014 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20220328 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: Y2021420000014 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A wind power prediction method based on model correction Effective date of registration: 20220329 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: Y2022420000088 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20230410 Granted publication date: 20140129 Pledgee: Guanggu Branch of Wuhan Rural Commercial Bank Co.,Ltd. Pledgor: WINDMAGICS (WUHAN) CO.,LTD. Registration number: Y2022420000088 |