WO2011124226A1 - Procédé et système de prévision d'énergie éolienne - Google Patents

Procédé et système de prévision d'énergie éolienne Download PDF

Info

Publication number
WO2011124226A1
WO2011124226A1 PCT/DK2011/050113 DK2011050113W WO2011124226A1 WO 2011124226 A1 WO2011124226 A1 WO 2011124226A1 DK 2011050113 W DK2011050113 W DK 2011050113W WO 2011124226 A1 WO2011124226 A1 WO 2011124226A1
Authority
WO
WIPO (PCT)
Prior art keywords
wind energy
power
forecasting
site
wind
Prior art date
Application number
PCT/DK2011/050113
Other languages
English (en)
Inventor
Carsten Hein Westergaard
Original Assignee
Vestas Wind Systems A/S
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Vestas Wind Systems A/S filed Critical Vestas Wind Systems A/S
Priority to US13/639,824 priority Critical patent/US9337656B2/en
Publication of WO2011124226A1 publication Critical patent/WO2011124226A1/fr

Links

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
    • 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
    • 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"
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
    • 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
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Definitions

  • the present invention relates to wind-energy-conversion technology and, for example, to a method and a system for forecasting wind energy production..
  • the known methods of wind energy forecasting comprise meteorological weather forecasting and converting this weather forecast into wind forecast that is converted into wind energy forecast.
  • Conventionally expected wind energy production is based on meteorological wind speed data; the expected production is then derived by using a model of the wind energy installation concerned, e.g. in the form of wind-speed dependent power curves.
  • a method for forecasting wind energy production comprises collecting power data indicative of power output from a set of wind energy installations comprising a first wind energy installation at a first site and a second wind energy installation at a second site, and estimating the available power output at a forecasting site at a geographic position.
  • the estimating is based on power data from the set of wind energy installations by projecting the power data towards the future and/or the geographic position of the forecasting site.
  • a method for forecasting wind energy comprising collecting power data indicative of power output from a set of wind energy installations comprising a first wind energy installation at a first site and a second wind energy installation at a second site, and estimating the available power output at a forecasting site based on power data from the set of wind energy installations and power gradients derived from the power data.
  • a wind energy production forecasting system comprising a processor and a memory.
  • the processor is programmed for estimating the available power output at the forecasting site based on power data indicative of power output from a set of wind energy installations comprising a first wind energy installation and a second wind energy installation by the method according to the first or the second aspect.
  • a method for scheduling wind energy production for an electricity grid which is fed by at least one wind energy installation and by at least one power plant with schedulable production, wherein wind energy production is forecast by the method according to the first or second aspect, and the forecast wind energy production is taken into account in the production scheduling of the at least one power plant.
  • wind energy is sensitive to the cube of the wind speed which means that the inaccuracy of power forecast as a first order approximation is 3 times higher than the wind speed forecast. For example, if the uncertainty of the wind speed is 5%, the power uncertainty is of the order 15%.
  • Another source of errors in the conventional methods may be due to the fact that a wind turbine (or wind energy installation) feels a wind speed averaged over its rotor area while typical wind speed measurements relate to a smaller area.
  • a further source of errors in conventional methods is an uncertainty in the knowledge of the power curves.
  • the methods and system described herein utilize measured power data of wind energy installations for forecasting power performance of one or more wind energy
  • the methods and system reduce or eliminate the need for calculating or estimating wind energy based on power curves that implies inaccuracy in wind energy forecasting.
  • the "forecasting" includes projecting current wind power production into the future in order to obtain a prediction.
  • forecasting includes projecting current (or past) wind power production data towards a geographical site which is different from the site were the wind power production is (or was) measured.
  • a combined projection in time and space is made; i.e. future production at a different geographical site is predicted.
  • the methods and system described herein utilize that data relating to wind power production may be achieved from several thousands of wind turbines.
  • the data sampling time needed may depend on the distance and energy transport speed. It is suggested to use the power produced by the wind energy installations for forecasting power performance. This is done by determining the transport of energy density. Wind energy forecasting may also be obtained for possible future wind energy installation sites, e.g. by interpolation or extrapolation.
  • the forecasting methods are carried out computer-based, i.e. the activities of collecting power data and estimating the available power output at a forecasting site by projecting the power data towards the future and/or the geographic position of the forecasting site are carried out by an accordingly programmed computer when the program is executing on the computer.
  • Fig, 1 a schematically illustrates an embodiment of the method
  • Fig. 1 b illustrates an exemplary projection method based on anoptical flow method in one dimension
  • Fig. 2 schematically illustrates a forecasting network including a set of wind energy installations connected to a wind energy forecasting system
  • Fig. 3 is a block diagram schematically illustrating an exemplary forecasting system
  • Fig. 4 illustrates a power density map at time t
  • Fig. 5 illustrates a forecasted power density map at time t + At .
  • Fig. 6 illustrates a power gradient map
  • the present system and methods are based on power data indicative of power output from a set of wind energy installations.
  • the methods and the system are based on direct information about electrical energy produced rather than wind speed.
  • a conventional weather forecast predicting wind speed and/or wind direction is used as an additional input to perform or improve the projection into the future the basis of the forecast is the measured wind power production.
  • Some embodiments do not use any weather forecast predicting wind speed and/or wind direction at all.
  • the wind energy production forecast may be one of the input parameters for energy production scheduling in an electricity grid, e.g. by providing this forecast information to the grid operator.
  • the production by these types of producers can therefore be scheduled in advance to meet the expected electricity demand, e.g. by the grid operator. These producers are therefore also referred to as "schedulable producers".
  • the prediction is an input parameter in the scheduling process performed e.g. by the grid operator of the schedulable producers.
  • the scheduling is, for example, does not only counteract predictable load variations but also the forecast variations of wind power production. For example, if the forecast predicts a reduction of wind power production the schedule can counteract by instructing schedulable producers to compensate for the reduction e.g. in a reciprocal manner, and vice versa.
  • the scheduling is performed automatically by a suitably programmed scheduling computer.
  • the scheduling can be carried out at different time scales. For example, in some embodiments, a short-term forecast of wind energy production (e.g. with a prediction for the next minutes) is provided. In such embodiments the grid operator's scheduling computer instructs schedulable short-term producers to produce electric power reciprocally to the wind energy forecast. In some embodiments, a medium-term (e.g. hours) forecast of power output is provided. In such embodiments the grid operator's scheduling computer instructs schedulable medium-term producers to produce electric power reciprocally to the wind energy forecast.
  • a short-term forecast of wind energy production e.g. with a prediction for the next minutes
  • the grid operator's scheduling computer instructs schedulable short-term producers to produce electric power reciprocally to the wind energy forecast.
  • a medium-term (e.g. hours) forecast of power output is provided. In such embodiments the grid operator's scheduling computer instructs schedulable medium-term producers to produce electric power reciprocally to the wind energy forecast.
  • the set of wind energy installations comprises a number of wind installations at different sites including a first wind energy installation at a first site and a second wind energy installation at a second site.
  • the set of wind energy installations may comprise at least 500 wind energy installations.
  • a wind energy installation may comprise one single wind turbine or a plurality of wind turbines or sets of wind turbines, the later often called wind parks or wind farms.
  • a single wind turbine or a wind turbine park including a number of wind turbines may be referred to as a wind energy installation.
  • wind turbines operating at nominal power output are excluded from the collected power data, or their power output is included but corrected, e.g. based on a local wind speed measurement so as to better reflect the power that would be producible without a design power limit.
  • These embodiments aim at getting rid of the saturation effect due to design power limits.
  • the saturation effect due to design power limits is intentionally included, because saturation may also happen to take place, e.g. at a forecasting site in the future.
  • wind turbines operating at nominal power output are included in the collected power data so that the set of collected power data reflects the saturation effect and represents what is actually produced (not what could be produced theoretically).
  • the power data may be projected towards the future and/or the geographic position of the forecasting site for estimating the available power output at the forecasting site.
  • the method comprises estimating the available power output, i.e. wind energy production, at the forecasting site.
  • the forecasting site may be identical to one of the first and the second site, i.e. the forecasting site may be the first site with the first wind energy installation or the second site with the second wind energy installation.
  • the available future power output from an existing wind energy installation may be forecast.
  • the methods may be used for planning future wind energy installation sites by enabling precise forecast of power output for possible new wind energy installations at new sites. This is useful for analysing potential sites and selecting the best site among candidates, and for selecting e.g. a suitable tower height and/or wind-turbine type (belonging to a suitable wind class) for a possible site.
  • collecting power data comprises collecting at least a first and a second set of power data at a first and a second time, respectively. Thus, at least two time-shifted sets of data are collected.
  • Each set of power data may be indicative of power output from at least the first and the second wind energy installation.
  • the power data of each wind energy installation may be obtained by measuring the effective electrical power output being generated by this wind energy installation. For example, this is done with the aid of measurement instruments that are (usually already) installed for each wind turbine or by measurement instruments in common for each wind energy installation.
  • power data indicative of power output from wind energy installations may be collected at selected interval or intervals, such as at least every 30 minutes, at least every 15 minutes, e.g. every 10 minutes or every 5 minutes. Intervals down to e.g. 10 seconds may be employed for selected wind energy installations, e.g. in order to obtain local power output forecasting.
  • the power data indicative of power output from a set of wind energy installations may comprise scalar information about the wind energy production, e.g. electrical power or electrical energy (produced in a specific time windows).
  • the power data may be a map of wind energy production for some or all sites.
  • one set of power data indicative of power output of some or all sites is measured substantially at the same point of time.
  • power data may be a scalar field with respect to the geographic positions of the sites representing a snap-shot of wind energy production.
  • estimating the available power at the forecasting site by projecting the power data towards the future is based on what is called an "optical flow method".
  • the optical flow is a motion, or a pattern of motions, deduced from
  • Optical flow methods are computer-based methods normally used to determine a motion of an object in an image.
  • gradient based optical flow methods gradient based methods
  • Optical Flow Estimation in "Mathematical Models in Computer Vision: The Handbook", N. Paragios, Y. Chen, and O. Faugeras (editors), Chapter 15, Springer, 2005, pp. 239-258. Since a (digital) image is e.g.
  • a motion in an image results in changes of lightness values at the different pixels.
  • a change of specific power outputs at the different wind turbine sites is analogue to a motion in an image.
  • the motion of the "image" represented by the power outputs at the different sites is calculated from at least two sets of power data, representing the power data at a first point of time and at a second point of time with a known time difference.
  • a velocity, i.e. speed and direction, of the motion of the image is calculated from the distance of the motion and the time difference.
  • the motion thus determined is then projected towards a future point in time.
  • the resulting moved "power image" (power outputs at different sites) is the forecast of the power output.
  • certain assumptions are made, e.g.
  • acceleration of the "power image” is also determined, and the acceleration thus determined is taken into account in the forecast (e.g. the acceleration can be assumed to remain constant rather than the velocity).
  • the forecast of wind energy production towards the future is estimated by calculating the time- dependent motion of (the scalar) wind energy production relative to the geographical positions of at least the first site and the second site.
  • estimating the available power output comprises projecting the power data towards the geographic position of the forecasting site.
  • the projection may comprise at least one of an extrapolation method and an interpolation method.
  • the projecting is based on the power data indicative of power output from the wind energy installations at least at the first and second site.
  • Collection of power data from the large number of wind energy installations improves accuracy of estimation of available power output for sites where no power data are available, e.g. by interpolation. This is useful in modeling new wind energy installations and predicting their performance at new sites, etc.
  • the forecast in time is combined with geographical extrapolation (or interpolation). For example, current power measurements at sites A, B, and C are used to predict wind power production at a different site D in one hour from now. Further, in some embodiments forecasting is based on data received from external meteorological data sources, such as weather databases, satellites, etc., e.g. as provided by National Oceanic and Atmospheric Administration (NOAA).
  • NOAA National Oceanic and Atmospheric Administration
  • estimating the available power output at the forecasting site is based on weather data of geographic positions of the wind energy installations.
  • the weather data may comprise at least one of wind speeds and wind directions.
  • estimating the available power output at the forecasting site is based on at least one of a direction and a magnitude of a power gradient derived from the power data.
  • the power gradient allows estimating power output by projecting known power output from the first or second site towards the forecasting site along the gradient.
  • the present methods and system map and forecast power performance by utilizing power data from wind energy installations as these are online, connected and can transmit performance or power data as often as requested. Offline data may also be considered, where this proves to be practical. Collecting power data from wind parks, individual wind turbines and wind turbines from around the world will allow this to take place.
  • the method comprises collecting power data for a set of wind energy installations.
  • Power data is collected, e.g. at 10-minute intervals, from a set of wind energy installations including at least the first and second wind energy installation.
  • Power data P(t,h,X n ) may be collected.
  • Power data for a wind energy installation may comprise power data for one or more wind turbines.
  • the power data may be normalized, such as to standard air density and pressure.
  • the method may comprise calculating one or more power densities from the power data indicative of power output from the set of wind energy installations.
  • Power densities may be calculated by the wind energy forecasting system and/or calculated and fed to the forecasting system as power data from the wind energy installation(s).
  • the method may comprise calculating unified power densities, e.g. at a predetermined height h 0 , for one or more wind energy installations in the set of wind energy installations.
  • the height h 0 may be freely chosen, wherein a height gradient of the wind speed, especially at ground level, may be taken into account.
  • the unified power densities (t,h 0 , X may be computed by applying a transfer function
  • the transfer function T(X n ,%) is based on the rotor area A Rotor and the hub height h .
  • the transfer function T(X n ,%) may be based on wind turbine data (type, operational status, operating parameters, etc).
  • the transfer function r( B ,...) may be based on local area information related to the site including but not limited to landscape profile that locally influences wind speed or turbine location inside park where shadowing effects between neighbored wind turbines occur.
  • local meteorological data and other factors may be considered by the transfer function T(X n ,...) . In such embodiments at least one of the power density and the unified power density depend on alternative or additional parameters.
  • the method may allow for power data P(t,h,X n ) or p(t,h,X n )to be used in the forecasting.
  • weighted significance may be used for forecasting the available power output. For example, a wind energy installation with a bigger number of wind turbines may be given more weight than a wind energy installation with less wind turbines.
  • Estimating the available power output at a forecasting site may comprise estimating one or more unified power densities at a forecasting site. Estimating the available power output at a forecasting site may be based on measurements of one or more meteorological parameters, e.g. wind velocities, i.e. wind speeds and/or wind directions, at the wind energy installations.
  • meteorological parameters e.g. wind velocities, i.e. wind speeds and/or wind directions
  • Estimating the available power output at a forecasting site may be based on cross correlation of power data indicative of power output from the set of wind energy installations, such as from the first wind energy installation and the second wind energy installation. Estimating the available power output may be based on power data from at least 5 different wind energy installations at different sites, such as from at least 10, at least 100, at least 1000 different wind energy installations at different sites.
  • the projection of the scalar can be directly determined from cross correlation between wind energy installations.
  • n(t, h 0 ,X n ) is replaced with P(t, h, X n ) or p(t, h,X n ) may be employed.
  • Wind energy installations are embedded into the earth boundary layer and the weather propagation is mainly driven by geotropic mechanisms outside the earth boundary layer. Accordingly, the method may in addition to the power data collect data related to these mechanisms, e.g. temperature and/or pressure data, and compute propagation velocity (speed and/or direction) from these two quantities, as these (in particular the pressure) are strongly correlated with the outer part of the part of the boundary layer.
  • the method for forecasting wind energy may be based on one or more scalars. It is believed that incorporation of one or more scalars may improve the forecasting of available power. Examples of scalars include but are not limited to one or more of temperature, pressure and gases released in the airflow.
  • the transport velocity, especially speed and direction, of the scalars can be determined, for example by cross- correlation methods. Accordingly, estimating or forecasting the available power output at a forecasting site may be based on transport of other scalars, such as one or more of temperature, pressure and gases released in the airflow. For example, this may be found for the wind energy installation n at site X n by finding the maximum value of
  • / is a suitable scalar, for example pressure or temperature.
  • a correlation function where n(t, h 0 ,X n ) is replaced with P(t, h, X n ) or p(t, h,X n ) may be employed.
  • the method may comprise sending data indicative of available power output at a forecasting site to a wind energy installation operator.
  • Estimating the available power output at a forecasting site is based on power gradients derived from the power data.
  • Power gradients may be derived by use of known methods ranging from correlation methods to neural network pattern recognition algorithms.
  • the method enables that transport of the power density can be projected over large distances and thus over long time, e.g. up to 24 hours or longer. Combining the power density transport function with other scalars it is contemplated that several days of good forecasting may be achieved.
  • Fig. 1 a schematically illustrates an embodiment.
  • the method 2 comprises collecting 4 power data indicative of power output from a set of wind energy installations comprising a first wind energy installation at a first site and a second wind energy installation at a second site.
  • Power data P(t,h, X n ) may be collected for a large number of wind energy installations at different sites around the world.
  • the distance between the first site and the second site may be at least 10 km, such as at least 50 km. It is an advantage of the present invention that the method is scalable, and can be deployed locally, such that the distance between the first site and the second site may be less than 10 km.
  • a sub-scale system could produce fast correlation between wind energy installations closely spaced, if the data infrastructure allows for fast
  • the method 2 comprises estimating 6 the available power output at a forecasting site located at a geographic position.
  • the estimating is based on power data from the set of wind energy installations and projection of the power data towards the future and/or the geographic position of the forecasting site.
  • Estimating 6 the available power output optionally comprises determining or calculating power densities p(t, h,X n ) of each of the wind energy installations in the set of wind energy installations.
  • the power densities may be calculated at the wind energy installation or in a SCADA system, thus collecting 4 power data may comprise collecting power density data p ⁇ t,h,X n ) from the set of wind energy installations.
  • Estimating 6 the available power output comprises calculating unified power densities n(t,h 0 ,X n ) b applying a transfer function T(X n ,...) to the power densities p(t,h,X n ) .
  • the method does not require, although it may be advantageous, that the power data is collected synchronously from the wind energy installations. However, the time of the power data collection must be known. The power data may be time-shifted in order to synchronize the power data.
  • Each set of power data represents a snap-shot of wind energy production of the set of wind energy installations. This enables a movement, or velocity, of the power field
  • the future power output ⁇ ( ⁇ + At,h 0 ,X n ) is, for example, calculated by an optical flow method based on the two snap-shots.
  • the optical flow method determines differences between wind energy productions of each site and calculates a movement of the wind energy production relative to the sites. To estimate the future wind energy productions of the sites this movement is projected into the future.
  • Fig. 1 b illustrates an exemplary method to determine the movement, or velocity, of a measured power field.
  • the example is a gradient-based optical flow method.
  • the "optical flow” is shown for only one dimension in Fig. 1 b (this would correspond to wind turbines arranged along a line and a motion of the "power image" produced by the wind turbines along the line).
  • fi(x) and f 2 (x) are one- dimensional sets of power data (with the line of wind turbines assumed to be nearly continuous) measured with a known time difference.
  • Fig. 1 b it is assumed that the "power image" moved a distance d between the two measurements.
  • f(x) needs to be measured only at one point on x 0 .
  • the gradient which, in one dimension, is the derivative f of the power distribution f(x) at the position x 0 ) at the measurement point x 0 the distance d can be inferred from the measured difference fi( x o) ⁇ ( ⁇ ) °f tne power at the two different points in time by
  • the power is measured at at least two different points, and the two measurements are combined, in order to uniquely determine not only the displacement d, but also the direction of the displacement of the "power image".
  • the power is measured at more than two points/sites; the resulting system of equations is then over-determined, and an approximate solution is estimated by one of the common estimation methods, e.g. the least-squares method.
  • the common estimation methods e.g. the least-squares method.
  • the velocity of the thus determined movement of the "power image” is then calculated from the displacement d and the known time difference (e.g. by dividing the
  • a "power image” for a future point of time is then obtained.
  • the extrapolation, or projection is in some embodiments based on an inertia assumption (i.e. that the velocity thus determined remains constant).
  • no prediction in time but only a geographic prediction of wind energy production is desired.
  • measurement of only one set of power data indicative of power output from the set of wind energy installations is sufficient.
  • the available power output is estimated by projecting the power data towards the geographic positions of the forecasting sites by an interpolation method for the forecasting sites between sites of the installations or by an extrapolation method for the other forecasting sites.
  • the available future power output n(t + At, h 0 , X n ) may be forecasted for a suitable time frame At , e.g. ten minutes, one hour, three hours, six hours, twelve hours, eighteen hours, one day, two days or more.
  • the forecast unified power densities are transformed to power densities by application of the inverse 7 7_1 ( B ,%)of the transfer function T(X n ,%) ⁇ n order to obtain
  • the forecast can be issued continuously or at selected intervals and may be updated as new power data is acquired for the wind installations.
  • the method 2 comprises sending 8 data indicative of available power output at the forecasting site to a wind energy installation operator.
  • the data may comprise power data of a number of wind energy installations enabling an operator to plot power densities or unified power densities as an animation over time, e.g. a color coded map showing the (unified) power densities for a selected geographical area with wind energy installations n.
  • the (future) wind energy production is forecast to provide input for scheduling power production of schedulable power producers.
  • the schedulable power producers are scheduled to vary their power in a manner reciprocal to the forecast wind power production.
  • Fig. 2 schematically illustrates a forecasting network including a set of wind energy installations comprising wind energy installations 102, 104, 106, the forecasting network further including a wind energy forecasting system 108 and optionally a Supervisory Control And Data Acquisition (SCADA) System 1 10 connected via data network 1 12.
  • the wind energy forecasting system 108 is configured for estimating the available power output at a forecasting site, e.g. the first wind energy installation 102, based on power data indicative of power output from a set of wind energy installations comprising the first wind energy installation 102 and the second wind energy installation 104.
  • the power data of wind energy installations 102, 104, 106 are collected via data network 1 12, optionally via SCADA system 1 10.
  • the wind energy forecasting system 108 may be directly connected to the SCADA system 1 10 via connection 1 14.
  • the wind energy forecasting system 108 may be configured for sending data indicative of available power output at a forecasting site to an operator of a wind energy installation at the forecasting site, e.g. via data network 1 12 or other network.
  • Fig. 3 is a block diagram schematically illustrating an exemplary forecasting system.
  • the forecasting system 108 comprises a processor 120 and memory 122 connected to the processor 120 via data connection 124, and an interface 126 comprising a data interface and a user interface connected to the processor via data connection 128.
  • the user interface may comprise a display and/or a keyboard and/or connectors for a display/keyboard.
  • the processor 120 is adapted to collect power data via the interface 126 and estimating the future unified power density (t + At, h 0 , X n ) at wind energy installation site X n based on power data from the set of wind energy installations.
  • FIG. 4 illustrates a geographical map of power densities at time t
  • Fig. 5 illustrates a map of forecasted power densities at time t +
  • the forecasted power densities are based on power data from wind energy installations 102, 104, 106, 130.
  • the sites of the wind energy installation are identical with the forecasting sites. Iso-power density curves are plotted on the map.
  • the wind energy installations are distributed over a large geographical area.
  • the forecast is based on power gradients which are illustrated in Fig. 6 showing a mapping of the power gradients based on power data of the wind energy installations 102, 104, 106, 130.
  • the forecast wind energy production may be obtained by projecting the wind energy production measured at time ⁇ along the power gradients.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Development Economics (AREA)
  • Game Theory and Decision Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Wind Motors (AREA)

Abstract

L'invention concerne un procédé permettant de prévoir la production d'énergie éolienne. Le procédé consiste notamment à recueillir des données d'énergie indiquant la sortie de puissance d'un ensemble d'installations d'énergie éolienne comprenant une première installation d'énergie éolienne sur un premier site et une seconde installation d'énergie éolienne sur un second site, et à estimer la sortie de puissance disponible au niveau d'un site de prévision situé en une position géographique. L'estimation s'appuie sur des données de puissance provenant de l'ensemble des installations d'énergie éolienne et sur une projection des données de puissance vers le futur et/ou sur la position géographique du site de prévision. L'invention concerne en outre un procédé de programmation de la production d'énergie éolienne pour un réseau d'alimentation électrique et un système de prévision de production d'énergie éolienne destiné à prévoir une sortie de puissance d'une installation d'énergie éolienne.
PCT/DK2011/050113 2010-04-08 2011-04-06 Procédé et système de prévision d'énergie éolienne WO2011124226A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US13/639,824 US9337656B2 (en) 2010-04-08 2011-04-06 Method and system for forecasting wind energy

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US32196110P 2010-04-08 2010-04-08
US61/321,961 2010-04-08
DKPA201070139 2010-04-08
DKPA201070139 2010-04-08

Publications (1)

Publication Number Publication Date
WO2011124226A1 true WO2011124226A1 (fr) 2011-10-13

Family

ID=44146804

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/DK2011/050113 WO2011124226A1 (fr) 2010-04-08 2011-04-06 Procédé et système de prévision d'énergie éolienne

Country Status (1)

Country Link
WO (1) WO2011124226A1 (fr)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542162A (zh) * 2011-12-26 2012-07-04 成都阜特科技有限公司 一种基于有效风速的风能评估方法
CN102609611A (zh) * 2011-12-26 2012-07-25 成都阜特科技有限公司 一种基于有效风速的风能评估方法
WO2013106941A1 (fr) * 2012-01-17 2013-07-25 Energiebüro AG Procédé de prévision de production d'énergie future prévue d'une centrale éolienne ou solaire
CN104504463A (zh) * 2014-12-12 2015-04-08 华南理工大学 基于趋势探测器和数学形态学算子的风能预测方法
CN108898251A (zh) * 2018-06-29 2018-11-27 上海电力学院 考虑气象相似性和功率波动的海上风电场功率预测方法
WO2019194734A1 (fr) * 2018-04-03 2019-10-10 Absolicon Solar Collector Ab Procédé de prédiction de puissance thermique de capteurs solaires
CN111080003A (zh) * 2019-12-10 2020-04-28 华能威宁风力发电有限公司 一种适应复杂山地风电场的测风塔规划选址方法
CN111260139A (zh) * 2020-01-17 2020-06-09 石化盈科信息技术有限责任公司 一种工业循环水系统的优选方法
CN117495434A (zh) * 2023-12-25 2024-02-02 天津大学 电能需求预测方法、模型训练方法、装置及电子设备

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
DAVID J. FLEET ET AL.: "Mathematical Models in Computer Vision: The Handbook", 2005, SPRINGER, article "Optical Flow Estimation", pages: 239 - 258
EPO: "Mitteilung des Europäischen Patentamts vom 1. Oktober 2007 über Geschäftsmethoden = Notice from the European Patent Office dated 1 October 2007 concerning business methods = Communiqué de l'Office européen des brevets,en date du 1er octobre 2007, concernant les méthodes dans le domaine des activités", JOURNAL OFFICIEL DE L'OFFICE EUROPEEN DES BREVETS.OFFICIAL JOURNAL OF THE EUROPEAN PATENT OFFICE.AMTSBLATTT DES EUROPAEISCHEN PATENTAMTS, OEB, MUNCHEN, DE, vol. 30, no. 11, 1 November 2007 (2007-11-01), pages 592 - 593, XP007905525, ISSN: 0170-9291 *

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102542162A (zh) * 2011-12-26 2012-07-04 成都阜特科技有限公司 一种基于有效风速的风能评估方法
CN102609611A (zh) * 2011-12-26 2012-07-25 成都阜特科技有限公司 一种基于有效风速的风能评估方法
WO2013106941A1 (fr) * 2012-01-17 2013-07-25 Energiebüro AG Procédé de prévision de production d'énergie future prévue d'une centrale éolienne ou solaire
CN104504463A (zh) * 2014-12-12 2015-04-08 华南理工大学 基于趋势探测器和数学形态学算子的风能预测方法
CN104504463B (zh) * 2014-12-12 2017-12-01 华南理工大学 基于趋势探测器和数学形态学算子的风能预测方法
WO2019194734A1 (fr) * 2018-04-03 2019-10-10 Absolicon Solar Collector Ab Procédé de prédiction de puissance thermique de capteurs solaires
US11486606B2 (en) 2018-04-03 2022-11-01 Absolicon Solar Collector Ab Method of forecasting heat output of solar collectors
CN108898251A (zh) * 2018-06-29 2018-11-27 上海电力学院 考虑气象相似性和功率波动的海上风电场功率预测方法
CN111080003A (zh) * 2019-12-10 2020-04-28 华能威宁风力发电有限公司 一种适应复杂山地风电场的测风塔规划选址方法
CN111080003B (zh) * 2019-12-10 2023-05-02 华能威宁风力发电有限公司 一种适应复杂山地风电场的测风塔规划选址方法
CN111260139A (zh) * 2020-01-17 2020-06-09 石化盈科信息技术有限责任公司 一种工业循环水系统的优选方法
CN111260139B (zh) * 2020-01-17 2023-09-19 石化盈科信息技术有限责任公司 一种工业循环水系统的优选方法
CN117495434A (zh) * 2023-12-25 2024-02-02 天津大学 电能需求预测方法、模型训练方法、装置及电子设备
CN117495434B (zh) * 2023-12-25 2024-04-05 天津大学 电能需求预测方法、模型训练方法、装置及电子设备

Similar Documents

Publication Publication Date Title
US9337656B2 (en) Method and system for forecasting wind energy
WO2011124226A1 (fr) Procédé et système de prévision d'énergie éolienne
US10288038B2 (en) Reducing curtailment of wind power generation
US20220275788A1 (en) Forecasting output power of wind turbine in wind farm
US10598157B2 (en) Reducing curtailment of wind power generation
EP3464895B1 (fr) Système et procédé de prévision de la puissance de sortie d'un parc éolien
US7523001B2 (en) Method and apparatus for operating wind turbine generators
JP2016136001A (ja) 予測装置
KR101476522B1 (ko) 에너지 발전량 예측 시스템
CN104299044A (zh) 基于聚类分析的风功率短期预测系统及预测方法
CN112329977A (zh) 一种面向极端场景的风电功率预测系统
WO2016160697A1 (fr) Système et procédé pour réaliser une prévision du vent
JPWO2014155482A1 (ja) 発電量予測装置、発電量予測方法、及びプログラム
Özgönenel et al. Short-term wind speed estimation based on weather data
CN110737876B (zh) 区域电网光伏功率预测优化方法和装置
CN105046349B (zh) 一种计及尾流效应的风电功率预测方法
CN104008284A (zh) 测风塔在数值天气预报中的校正方法
US11365719B2 (en) Power production forecasting for a wind turbine
Rohrig et al. Improvement of the power system reliability by prediction of wind power generation
EP3271875A1 (fr) Système et procédé de prédiction de génération solaire d'électricité
CN111950780A (zh) 一种风电场短期功率预测方法
Agabus et al. ESTIMATION OF WIND POWER PRODUCTION THROUGH SHORT-TERM FORECAST.
KR102491085B1 (ko) 대기모형 예보 기반 풍력 발전량 예측 장치 및 방법
JP2022041390A (ja) 電力管理装置および複数の発電設備の発電量予測管理方法
CN117043793A (zh) 预测可再生能源设施的发电

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11719759

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 13639824

Country of ref document: US

122 Ep: pct application non-entry in european phase

Ref document number: 11719759

Country of ref document: EP

Kind code of ref document: A1