METHOD FOR OPERATING A WIND TURBINE

Abstract

A method for operating a wind turbine for generating a settable turbine power, where the wind turbine includes a rotor having rotor blades adjustable in their blade angle, is operable at a settable rotor speed, and is installed at an installation site at a distance to an obstacle, comprises the obstacle causing a wind disturbance, which, in dependence on current wind direction and wind velocity, can reach the wind turbine as a wind wake, and the wind turbine reducing its turbine operation by throttling down for protection against loads due to the wind wake, wherein the throttling down is controlled in dependence on the current wind direction and the current wind velocity, wherein a weather prediction is used in order to take into consideration at least one further weather property in addition to the wind direction and wind velocity, and wherein the throttling down is additionally controlled in dependence on the weather prediction, in particular on the further weather property.

Claims

1. A method for operating a wind turbine for generating a settable turbine power, wherein the wind turbine: includes a rotor having rotor blades adjustable in their blade angle, is operable at a settable rotor speed, and is installed at an installation site at a distance to an obstacle, wherein the method comprises: the obstacle causing a wind disturbance, which, in dependence on current wind direction and wind velocity, can reach the wind turbine as a wind wake, and the wind turbine reducing its turbine operation by throttling down for protection against loads due to the wind wake, wherein the throttling down is controlled in dependence on the current wind direction and the current wind velocity, wherein a weather prediction is used in order to take into consideration at least one further weather property in addition to the wind direction and wind velocity, and wherein the throttling down is additionally controlled in dependence on the weather prediction.

2. The method as claimed in claim 1, wherein: the further weather property is estimated in dependence on the weather prediction, and/or the further weather property is a thermal stratification of atmospheric boundary layers, and the throttling down is controlled in dependence on the thermal stratification.

3. The method as claimed in claim 1, wherein the throttling down is controlled in dependence on a thermal stratification so that an area surrounding the wind turbine is classified as flat land or mountainous land, and in the case of classification as flat land, throttling down is carried out more the weaker mixing is between boundary layers and/or the more stable the thermal stratification is, and in the case of classification as mountainous land, throttling down is carried out less the weaker mixing is between boundary layers and/or the more stable the thermal stratification is.

4. The method as claimed in claim 1, wherein for throttling down: the turbine power is reduced, the rotor speed is reduced, and/or the blade angle of each rotor blade is adjusted in the direction toward a vane position.

5. The method as claimed in claim 1, wherein: a length of the wind wake is estimated in dependence on the weather prediction; and the throttling down is controlled in dependence on the estimated length of the wind wake.

6. The method as claimed in claim 5, wherein the length of the wind wake is estimated in dependence on the weather property.

7. The method as claimed in claim 1, wherein the weather prediction is adapted using a regional weather model at the installation site, wherein: the regional weather model describes a relationship between the weather prediction, which is created for an area going beyond the installation site, and a local weather property at the installation site, and/or wherein: flow states at the wind turbine, which are influenced by the wind wake, are estimated in dependence on the weather prediction, and the throttling down is controlled in dependence on the estimated flow states.

8. The method as claimed in claim 7 wherein the flow states at the wind turbine are estimated in dependence on the weather property.

9. The method as claimed in claim 1, wherein flow states which are estimated in dependence on the weather prediction are additionally estimated in dependence on at least one local weather model, possibly with further consideration of a regional weather model, wherein: a local weather model describes a relationship between the weather prediction and flow states expected at the wind turbine.

10. The method as claimed in claim 1, wherein: a local weather model is trained and/or adapted in ongoing operation of the wind turbine by ongoing comparison of the weather prediction to flow states detected at the wind turbine, and/or is trained and/or adapted in dependence on historic data, and/or is determined by simulations and/or an estimation of flow states at the wind turbine, which are influenced by the wind wake, is improved by current measurements at or in the vicinity of the wind turbine and/or if the wind turbine is installed in a wind park, wakes within the wind park are taken into consideration.

11. The method as claimed in claim 1, wherein: the throttling down of the level is controlled continuously or in multiple steps in dependence on the weather prediction, and/or wherein: a further wind turbine forms the obstacle, and/or the wind turbine is one of multiple wind turbines of a wind park, and one of the other wind turbines, depending on the wind direction, forms the obstacle.

12. The method as claimed in claim 1, wherein the throttling down is additionally controlled in dependence on the further weather property.

13. A method for planning a wind park including multiple wind turbines, comprising: estimating an energy production to be expected of the wind park by simulating operation of the wind turbines for an estimation period of time, wherein for the simulation in each case: operation of a wind turbine for generating a settable turbine power is simulated, wherein, in the simulation, the wind turbine includes a rotor having rotor blades adjustable in their blade angle, is operable using a settable rotor speed, and is installed at an installation location at a distance to an obstacle, wherein simulating operation of a wind turbine includes: the obstacle causing a wind disturbance, which, in dependence on current wind direction and wind velocity, can reach the wind turbine as a wind wake, and the wind turbine reducing its turbine operation by throttling down for protection against loads due to the wind wake, wherein the throttling down is controlled in dependence on the current wind direction and the wind velocity, wherein a weather prediction is used in order to take into consideration at least one further weather property in addition to the wind direction and wind velocity, and the throttling down is additionally controlled in dependence on the weather prediction, wherein historic weather data of the installation site are used as the weather prediction.

14. The method as claimed in claim 13, wherein: the estimation period of time is one year; and the throttling down is additionally controlled in dependence on the further weather property.

15. The method as claimed in claim 13, wherein: for the simulation of an operation of a wind turbine, in each case: historic weather data are used for the weather prediction, a local weather model is trained or adapted in dependence on historic data, and/or is determined by simulations and/or in an estimation of flow states at the wind turbine, which are influenced by the wind wake, historic measured values are used instead of current measurements at or in the vicinity of the wind turbine, which are used for improvements.

16. A wind turbine for generating a settable turbine power, and the wind turbine comprising: a rotor having rotor blades adjustable in their blade angle, wherein the wind turbine is operable at a settable rotor speed, and wherein the wind turbine is installed at an installation site at a distance to an obstacle; wherein: the obstacle can cause a wind disturbance which, in dependence on current wind direction and wind speed, can reach the wind turbine as a wind wake, wherein the wind turbine is configured to be operated using a method in which: the wind turbine reduces its turbine operation by throttling down for protection against loads due to the wind wake, wherein the throttling down is controlled in dependence on the current wind direction and the current wind velocity, a weather prediction is used in order to take into consideration at least one further weather property in addition to the wind direction and wind velocity, and the throttling down is additionally controlled in dependence on the weather prediction, wherein: the control of the throttling down is implemented on a throttling down control unit and the wind turbine includes this throttling down control unit or is coupled thereto or includes an interface for coupling with the throttling down control unit.

17. The wind turbine as claimed in claim 16, wherein the throttling down is additionally controlled in dependence on the further weather property.

18. The wind turbine as claimed in claim 16, further comprising: a throttling down control unit, prepared to control the throttling down of the wind turbine.

19. A wind park having multiple wind turbines, the wind park comprising: at least one wind turbine as claimed in claim 16, and a park controller, which comprises a throttling down control unit prepared to control throttling down of the wind turbines.

20. The wind park as claimed in claim 19, wherein: the park controller is configured to communicate with the wind turbines, in order to control the throttling down of a first wind turbine in dependence on a second wind turbine, if the second wind turbine forms the obstacle for the first wind turbine.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0105] Embodiments of the invention are explained in more detail hereinafter by way of example on the basis of embodiments with reference to the appended figures.

[0106] FIG. 1 shows a wind turbine in a perspective view.

[0107] FIG. 2 shows a wind park in a schematic illustration.

[0108] FIG. 3 schematically shows an obstacle situation for two wind directions in a top view.

[0109] FIG. 4 schematically illustrates the effect of a thermal air stratification on an obstacle situation for two examples.

[0110] FIG. 5 shows a schematic flow chart for implementing a proposed method.

DETAILED DESCRIPTION

[0111] FIG. 1 shows a schematic illustration of a wind turbine according to some embodiments. The wind turbine 100 includes a tower 102 and a nacelle 104 on the tower 102. An aerodynamic rotor 106 having three rotor blades 108 and a spinner 110 is provided on the nacelle 104. The aerodynamic rotor 106 is set into a rotational movement in operation of the wind turbine by the wind and thus also rotates an electrodynamic rotor or impeller of a generator, which is coupled directly or indirectly with the aerodynamic rotor 106. The electrical generator is arranged in the nacelle 104 and generates electrical energy. The pitch angle of the rotor blades 108 can be changed by pitch motors at the rotor blade roots 109 of the respective rotor blades 108.

[0112] Moreover, a turbine controller 140 is schematically shown in FIG. 1, which is shown outside the tower 102 only for reasons of illustration. The turbine controller 140 can be arranged in the tower 102 and/or in the nacelle 104. It also comes into consideration that a division is provided, so that one part of the turbine controller is located in the tower 102 and another part is located in the nacelle.

[0113] The method for operating the wind turbine and in particular for throttling down the operation of the wind turbine can be implemented in the turbine controller 140. For this purpose, the turbine controller can receive a weather prediction from an external weather model 142. This is schematically indicated in FIG. 1 and such weather data of the weather prediction can be transmitted to the turbine controller via radio and/or by a SCADA.

[0114] The turbine controller additionally receives sensor data from sensors 144, for which an anemometer having a wind vane is shown here by way of example and as an illustration. The turbine controller insofar actually uses information regarding wind velocity and wind direction for throttling down. However, this information can also be ascertained in another way. For the wind direction, it comes into consideration in particular that it is provided to the turbine controller in any case and/or the current azimuth orientation of the wind turbine is used instead of the wind direction. The current azimuth orientation of the wind turbine can in turn be dependent on the wind direction detected by the wind vane.

[0115] The current wind velocity can be used as a measured value by the anemometer or another sensor, for which the anemometer also stands as a representation, or it can be estimated from turbine data such as speed, power, and blade angle.

[0116] It is additionally proposed that specific flow conditions at the wind turbine 100 be detected, which also provides sufficient information for detecting turbulence. Further sensors can be provided for this purpose, for which the sensors 144 are also to stand as a representation.

[0117] Regional and/or local weather models can also be implemented in the turbine controller, which can improve the data of the weather model 142. The regional weather model can particularly be provided for the purpose of giving the data of the weather model 142 a higher resolution. The local weather model can particularly be provided for depicting a specific situation at the wind turbine 100.

[0118] Moreover, the turbine controller 140 can contain the information about the obstacle. Particularly in the case of fixed obstacles, corresponding information can already be stored as one-time information during the erection of the wind turbine. However, transmitting such information again regularly also comes into consideration.

[0119] FIG. 2 shows a wind park 112 having three wind turbines 100 by way of example, which can be identical or different. The three wind turbines 100 are thus representative for a fundamentally arbitrary number of wind turbines of a wind park 112. The wind turbines 100 provide their power, namely in particular the generated current, via an electrical park grid 114. The respectively generated currents or powers of the individual wind turbines 100 are added up here and a transformer 116 is usually provided, which steps up the voltage in the park in order to then feed it into the supply grid 120 at the feed point 118, which is also designated in general as a PCC. FIG. 2 is only a simplified illustration of a wind park 112. For example, the park grid 114 can be designed differently in that, for example, a transformer is also provided at the output of each wind turbine 100, to name only one other embodiment.

[0120] The wind park 112 moreover includes a central park computer 122, which can also be designated synonymously as a central park controller. This can be connected via data lines 124, or wirelessly, to the wind turbines 100, in order to exchange data with the wind turbines via this and in particular to receive measured values from the wind turbines 100 and transmit control values to the wind turbines 100.

[0121] According to one variant, the throttling down of one of the wind turbines 100 can be implemented on the central park computer 122. Throttling down for several of the wind turbines can also be implemented here, wherein individual throttling down is carried out in each case for each wind turbine and insofar an individual method is at least partially implemented in each case. The methods can take into consideration common features, however, in particular the ascertainment of the at least one further weather property can be identical for all wind turbines, since such a further weather property applies in the same way to the entire wind park, depending on the park size.

[0122] The park computer 122 can insofar comprise a throttling down control unit. The throttling down control unit can form a separate physical unit or can be implemented as program code on the park computer. The park computer or the throttling down control unit also includes interfaces for this purpose in order to be able to receive weather data of a weather model 142.

[0123] The central park computer 122, which can also be designated in simplified and synonymous form as a park computer, moreover receives sensor data from a sensor arrangement 145. The sensor arrangement 145 can be a measuring mast in the wind park. However, it is also conceivable that sensor data are received from the wind turbines. The park computer 122 can receive sensor data from the wind turbines, and also other data from the wind turbines, via the data lines 124. The park computer can also control the wind turbines via the data lines 124.

[0124] The park computer 122 can contain a park controller or the park computer 122 can also be designated as a park controller.

[0125] If the throttling down is controlled by the park computer 122, it ascertains the at least one further weather property. This possibly takes place individually for each affected wind turbine, namely in each case via the knowledge of a relevant obstacle. A weather prediction is thus also used here in order to take into consideration at least one further weather property in addition to the wind direction and wind velocity. This is insofar to be understood generally so that the weather prediction provides the further weather property as information, or this further weather property is derived from the information which the weather prediction provides. The park computer 122 derives a provided throttling down therefrom and can transmit this as a throttling down signal via the data lines 124 to each of the affected wind turbines, which then implement the throttling down for themselves.

[0126] FIG. 3 illustratively explains the problem occurring due to an obstacle for a wind turbine. FIG. 3 shows two top views A and B, each of a wind turbine 300 and an obstacle 302, which is only schematically shown here and can be representative for various obstacles. The obstacle can be a wind turbine, a tree or a building or also a hill, or a forest, in order to only mention several examples.

[0127] The two illustrations A and B differ in the prevailing wind direction 304 A and 304 B, respectively.

[0128] In illustration A, the wind direction 304 A has the result that a wind wake 306 A induced by the obstacle 302 reaches the wind turbine 300. The wind wake 306 A is insofar schematically shown by two delimiting dashed lines.

[0129] In illustration A, the wind wake 306 A thus reaches the wind turbine 300 and in this way turbulence can especially occur at the wind turbine 300, which can damage it.

[0130] With the somewhat different wind direction 304 B of illustration B, the obstacle 302 also induces a wind wake 306 B, which does not reach the wind turbine 300, however, and therefore does not result in damaging turbulence at the wind turbine 300. The wind turbine 300 therefore does not need to be throttled down.

[0131] The illustration in FIG. 3 explains a fundamentally known situation having known solution. It has the result that for the situation of illustration A, thus whenever the wind direction 304 A is present, the wind turbine 300 is throttled down. Of course, this also takes place in dependence on the wind velocity, because no relevant damage to the wind turbine 300 is to be expected at low wind velocities.

[0132] However, it has now been recognized that the illustrated wind direction 304 A, even with sufficiently strong wind, does not have to result in a particularly large load due to the wind wake 306 A at the wind turbine 300. It has been recognized that at least one further weather property can have an influence and should therefore be taken into consideration.

[0133] One such weather property is a thermal stratification, which is illustrated in FIG. 4. The thermal stratification between air layers is illustrated by a horizontal thermal boundary layer 408. FIG. 4 therefore shows in each of illustrations A and B a side view of a wind turbine 400 and an obstacle 402. The wind turbine 400 can correspond to the wind turbine 300 of FIG. 3, and moreover also a wind turbine 100 of FIG. 1 or FIG. 2. The obstacle 402 can correspond to the obstacle 302 of FIG. 3. Wind having a wind direction 404 is also shown. The wind direction 404 can correspond to the wind direction 304 A according to FIG. 3. The wind direction 404 is thus such that a wind wake 406 A or 406 B originating from the obstacle 402 is initially directed in the direction toward the wind turbine 400.

[0134] Particularly with a flat landscape and the thermal boundary layer 408 shown in illustration A, the wind wake 406 A can reach the wind turbine 400. This is especially a situation which is to be expected with flat terrain. If the region is thus flat, therefore not mountainous, and if strong thermal stratification is present, which is symbolized here by the thermal boundary layer 408, it is to be expected that the wind wake 406 A will reach the wind turbine 400.

[0135] In the illustration according to Figure B, no thermal boundary layer is shown and a case is therefore illustrated here in which no or only minor thermal stratification of the air masses of the atmosphere is present. This can have the result that the wind wake 406 B does not reach the wind turbine 400 or reaches it more weakly in comparison to situation A. Only for illustration, the wind wake 406 B is shown for this purpose so that originating from the obstacle 402 toward the wind turbine 400 it is resolved more and more strongly, for example, itself has turbulence, so that a clear or a long clear wind wake does not result.

[0136] The illustration of FIG. 4 of situation B is insofar only to be understood as illustrative, however, and is not supposed to reflect the actually occurring weather phenomena. It should be made clear that in spite of the same wind direction 404 and same wind strength assumed in FIG. 4, in one case the wind wake 406 A reaches the wind turbine 400, but in the other case the wind wake 406 B does not reach the wind turbine 400. Of course, it also comes into consideration that the wind wake 406 B only reaches the wind turbine 400 more weakly. In any case, it reaches the wind turbine 400 so much more weakly that this turbine has to be throttled down less or not at all in comparison to the situation according to FIG. 4, illustration A.

[0137] FIG. 5 explains a sequence for throttling down a wind turbine. The sequence 500 basically begins with the weather prediction step 502. Accordingly, a numeric weather prediction is provided. The physical variables which influence the occurrence of damaging flow states can be obtained in time and space thereby. Weather predictions can be provided in the form of current numerical weather predictions (NWP).

[0138] For improvement, such weather predictions can be coupled with a regional weather model, which is carried out in refining step 504. In the refining step, a regional weather model can therefore be present in order to also achieve a higher resolution. Future weather models can possibly provide a higher resolution, so that the regional weather model or a local weather model could be superfluous.

[0139] It is therefore especially proposed that the weather model, which has a low spatial resolution, be upgraded in refining step 504. It is especially proposed that an additional method be applied in order to obtain a prediction of the flow at the turbine position. Such methods establish a relationship between the results of the weather models and the damaging flow states to be expected for the turbines. The relationship between the results of the weather models and the flow state at the turbines can be ascertained by prefinished simulations.

[0140] Such simulations can be designated as RANS or LES. Statistical models can also be used.

[0141] In addition to a regional weather model, a local weather model, which stands for the mentioned methods which result in more accurate flow conditions at the turbine, can insofar also be used in the refining step.

[0142] Finally, the results are passed on at throttling down step 506, which derives the specific throttling down from the data. The throttling down step therefore stands for the application of a weather-related sector management at the turbine. The prediction quality of the models which are described in weather prediction step 502 or refining step 504 can additionally be improved by measurements, for example, from measuring masts located in the vicinity of the turbine or LIDAR systems installed on the turbine.

[0143] The recorded data can be incorporated via a SCADA system of the turbine into the regulation strategy. SCADA data can therefore also be discussed here. An optimization of the prediction quality can thus be carried out by training of the base model using historic data. In addition, a live optimization, thus an online optimization in ongoing operation of the prediction quality by fusing the current prediction with live data comes into consideration. In these two cases, thus the optimization of the prediction quality and the live optimization, methods of machine learning are used for the optimization.

[0144] The mentioned sensor data can therefore be incorporated by sensor step 508, which only stands as an example here for the recording of measurement data. Other measurement data such as wind direction and wind velocity can also be recorded in the measuring step and can each be incorporated.

[0145] According to some embodiments, particularly the following was recognized or the following is proposed.

[0146] Some embodiments can be used for ongoing operation and for park planning. For an application in ongoing operation of existing turbines/parks, the proposed algorithm especially includes three steps:

[0147] Step A1: providing the numeric weather prediction: The physical variables which influence the occurrence of harmful or damaging flow states in time and space can be obtained thereby. Weather predictions can be used in the form of: [0148] a) current numerical weather predictions (NWP), and/or [0149] b) current NWP models, coupled with a regional weather model (in English mesoscale coupling).

[0150] Future weather models, which are expected to be resolved in scale, thus have a high resolution, can then also be used.

[0151] Step A2: Since the weather models described in step A1 under a) and b) have a low spatial resolution, it is proposed that additional methods be applied to obtain a prediction of the flow at the turbine position. These methods represent a relationship between the results of the weather models and the harmful or damaging flow states to be expected for the turbines. The relationship between the results of the weather models and the flow state at the turbines can be established by prefinished RANS or LES simulations. Statistical models can also be used.

[0152] Step A3: Applying the weather-related sector management at the turbine. The prediction quality of the models which are described in step A1 and step A2 can additionally be improved by measurements such as i) from measuring masts located in the vicinity of the turbine or ii) LIDAR systems installed on the turbine or iii) can be incorporated into the SCADA data of the turbine itself and into the regulation strategy, by the following measures: [0153] a) optimization of the prediction quality by training the base model using historic data [0154] b) live optimization of the prediction quality by fusing the current prediction with live data

[0155] In both cases A3 a) and A3 b), methods of machine learning can be used for the optimization.

[0156] For the application in park planning, historic weather predictions for the site in question are used instead of current weather predictions. Step A1 and step A2 remain essentially unchanged here.

[0157] In a step B3 deviating from step A3, the effect of the sector management can then be simulated in order to quantify the increase of the AEP or the service life.

[0158] The regulation algorithms can be integrated into the turbine controllers or into the park controller.

[0159] After integration into a controller, the method can be used to achieve more power in ongoing operation or lengthening of the service life of the turbines.

[0160] The adaptations of the controller can also already be taken into consideration in the park planning phase. A quantification of the AEP thus increased or the service life thus increased can be used to reduce the electricity production costs.

[0161] A maximum current production or service life can be achieved by an adaptation of the controller to the conditions prevailing at the respective site. Some embodiments provide throttling down in dependence on the wind direction and is insofar sectorial throttling down and the throttling down can therefore be designated as sector management. However, in addition to the wind velocity, a further weather property is taken into consideration, so that an additional condition is present, and the method can therefore be designated as conditional sector management. This conditional sector management results in lower costs due to the consideration of this additional condition.

[0162] Due to the prediction of harmful or damaging flow states for the turbine, the turbines can be throttled down at points in time at which these harmful states or states damaging the turbine occur with high probability. In contrast, the turbines can be moved into a mode of increased power production when these harmful states or states damaging the turbine occur with a low probability. Such a regulation strategy or control strategy is reasonable against the background of the fact that loads with high amplitude are incorporated exponentially in the consumption of the service life. The power or the service life can be maximized by such a regulation strategy and the electricity production costs can thus be reduced.

[0163] Aspects of the various embodiments described above can be combined to provide further embodiments. In general, in the following claims, the terms used should not be construed to limit the claims to the specific embodiments disclosed in the specification and the claims, but should be construed to include all possible embodiments along with the full scope of equivalents to which such claims are entitled.