Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
OPERATIONAL STATE BASED MULTI-ROTOR WIND TURBINE CONTROL STRATEGY
Document Type and Number:
WIPO Patent Application WO/2019/034218
Kind Code:
A1
Abstract:
A method is disclosed for controlling a multi-rotor wind turbine. The method comprises a step of receiving, from respective production controllers (115, 125, 145), operational data representative of a current power output of the respective rotors (110, 120, 130, 140). A rotor activity pattern is determined based on the operational data of each the rotors (110, 120, 130, 140), and for each rotor (110, 120, 130, 140), optimal control settings is determined and submitted to the respective production controllers (115, 125, 145). The optimal control settings are determined based on the rotor activity pattern and on aerodynamic performance data (210). The aerodynamic performance data (210) depends on the rotor activity pattern. When the respective rotor is operating at a constant tip speed ratio, the constant tip speed ratio is dependent on the rotor activity pattern.

Inventors:
NETO, Julio Xavier Vianna (Olof Palmes Allé 27A, 2.th, 8200 Århus N, 8200, DK)
Application Number:
DK2018/050190
Publication Date:
February 21, 2019
Filing Date:
August 03, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
VESTAS WIND SYSTEMS A/S (Hedeager 42, 8200 Aarhus N, 8200, DK)
International Classes:
F03D1/02; F03D7/02
Domestic Patent References:
WO2016150447A12016-09-29
WO2016128002A12016-08-18
WO2017092762A12017-06-08
Foreign References:
EP3032095A12016-06-15
Download PDF:
Claims:
CLAIMS:

1. A method for controlling a multi-rotor wind turbine (100) with at least two rotors (1 10, 120, 130, 140), the method comprising the steps of:

receiving, from respective production controllers (115, 125, 145) of the at least two rotors (110, 120, 130, 140), operational data representative of a current power output of the respective rotors (1 10, 120, 130, 140),

determining a rotor activity pattern based on the operational data of each one of the at least two rotors (1 10, 120, 130, 140),

determining, for each one of the at least two rotors (1 10, 120, 130, 140), optimal control settings, based on the rotor activity pattern and on aerodynamic performance data (210) of the respective rotor (1 10, 120, 130, 140), the aerodynamic performance data (210) being dependent on the rotor activity pattern, and

submitting the optimal control settings of each one of the at least two rotors (110, 120, 130, 140) to the respective production controllers (1 15, 125, 145), wherein

when the respective rotor is operating in partial load operation at a constant tip speed ratio, the constant tip speed ratio is dependent on the rotor activity pattern.

2. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the constant tip speed ratio is higher when the rotor activity pattern indicates that more rotors are active.

3. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein different optimal control settings are submitted to at least two of the at least two rotors (1 10, 120, 130, 140).

4. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the optimal control settings include at least:

an optimal pitch angle, and

an optimal rotational speed or tip speed ratio

5. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the operational data representative of the current power output of the respective rotors (1 10, 120, 130, 140) comprises at least two of:

a current power output

a current pitch angle, a current rotor speed,

a current wind speed,

6. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the aerodynamic performance data includes a lookup table (210) for providing power coefficients for different tip speed ratios, pitch angles and rotor activity patterns, the rotor activity patterns being derivable from the operational data of each one of the at least two rotors (1 10, 120, 130, 140). 7. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 6, wherein the rotor activity patterns are derivable from a combination of the operational data of each one of the at least two rotors (1 10, 120, 130, 140) and data concerning relative distances between the at least two rotors (110, 120, 130, 140). 8. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the optimal control settings are determined such as to maximize a power coefficient in view of the received operational data of all rotors (1 10-140) of the multi-rotor turbine (100). 9. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 6, wherein the optimal control settings are determined such as to select for at least one of the rotors (110-140) the control settings corresponding to a maximized power coefficient.

10. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 9, wherein the optimal control settings are determined such as to select for the at least one of the rotors (1 10-140) a combination of tip speed ratio and pitch angle, which combination corresponds to the maximized power coefficient.

11. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the optimal control settings are determined such as to maximize power production for the multi-rotor wind turbine (100) as a whole.

12. A method for controlling a multi-rotor wind turbine (100) as claimed in any of the preceding claims, wherein the steps of determining and/or submitting are only performed if the respective rotor is operating in partial load at a constant tip speed ratio for a continuous range of wind speeds.

13. A method for controlling a multi-rotor wind turbine (100) as claimed in claim 1 , wherein the steps of determining and/or submitting are only performed if the respective rotor is operating in region II. 14. A computer program product comprising computer code for performing, when executed on a computing means, a method as claimed in any one of the claims 1 to 13.

15. A multi-rotor wind turbine (100) comprising:

at least two rotors (110, 120, 130, 140), each rotor comprising a respective production controller (1 15, 125, 145), and

an optimal operation manager (200), operatively coupled to the respective production controllers (115, 125, 145) and configured to perform a method as claimed in any one of the claims 1 to 13.

Description:
OPERATIONAL STATE BASED MULTI-ROTOR

WIND TURBINE CONTROL STRATEGY

FIELD OF THE INVENTION

The invention relates to a method for controlling a multi-rotor wind turbine with at least two rotors. The method comprises steps of receiving operational data representative of a current operational state of the rotors, determining optimal control settings, based on at least the operational data and aerodynamic performance data, and submitting the optimal control settings to the production controllers. The invention further relates to a computer program product and a multi-rotor wind turbine operable to perform such a method.

BACKGROUND OF THE INVENTION

Depending on wind speed, wind turbines generally operate in four different modes, also called 'regions'. In region I, the rotor rotates at its minimum operational speed. Wth increasing wind speeds, the blade pitch (i.e. the orientation of the rotor blade surface, relative to the wind direction) is adapted to increase the power output of the wind turbine. In region I, the wind turbine does not run at its optimum efficiency. For every wind speed, there is an optimal combination of pitch and tip speed ratio (i.e. the rotational speed of the rotor times its radius, divided by the wind speed) at which the power coefficient (a measure for the rotor's efficiency) is at a maximum. At the end of region I, the pitch and the rotational speed of the rotor are such that the rotor performs at the top of its performance curve.

Wth further increase of the wind speed region II starts, the rotational speed of the rotor is changed proportionally while the pitch is kept constant. As a result the tip speed ratio is also kept constant and the rotor keeps on performing with a maximum power coefficient. When the rotor speed reaches its maximum (rated speed), the turbine enters region III. Like in region I, only a change of pitch can make the power output increase further with increasing wind speeds. Regions l-lll are also referred to as partial load operation, because the wind turbine has not yet reached its maximum power output. When the maximum power output is reached (also called rated power), the wind turbine operates in region IV. In the event that rated speed is not reached before rated power, there will be no region III and the operation switches to region IV directly from region II. In region IV, the blade pitch may be adapted to avoid damage to the wind turbine at even higher wind speeds. When the wind speed is too high, the turbine is brought to a halt for safety reasons. Region IV is also known as full load operation.

A typical wind turbine, under typical conditions, operates in region II for more than 50% of the time. Knowing the optimal tip speed ratio and blade pitch of a rotor thus is important for optimal operation of a wind turbine. The exact shape of the performance curve depends on the dimensions and geometry of the rotor and its blades. For single rotor wind turbines, determining and using the performance curve is a well-established practice. Computer models of the wind turbine and its rotor blades, together with an analysis of the air around the operating wind turbine (e.g. using blade Element

Momentum theory, also known as BEM), are used to establish the rotor-specific performance curves.

The inventor has found that simply using the separate performance curves of the individual rotors of a multi-rotor wind turbine does not always lead to optimal

performance. It is therefore an object of this invention, to improve the efficiency of multi- rotor wind turbines.

SUMMARY OF THE INVENTION

According to the invention this object is achieved by providing a method for controlling a multi-rotor wind turbine with at least two rotors, the method comprising the steps of receiving, from respective production controllers of the at least two rotors, operational data representative of a current power output of the respective rotors, determining a rotor activity pattern based on the operational data of each one of the at least two rotors, determining, for each one of the at least two rotors, optimal control settings, based on the rotor activity pattern and on aerodynamic performance data of the respective rotor and submitting the optimal control settings of each one of the at least two rotors to the respective production controllers. With the thus received optimal control settings, the respective production controller may then control the respective rotor. When the respective rotor is operating in partial load operation, the constant tip speed ratio is dependent on the rotor activity pattern. Preferably, this constant tip speed ratio is increased when more rotors are active or when the rotor speed of one or more of the rotors increases. In partial load, the constant tip speed ratio may be dependent on the activity pattern for a continuous range of wind speeds, the wind speeds being below the rated wind speed. This may e.g. be for the respective rotor operating in region II.

The aerodynamic performance data used for determining the optimal control settings is dependent on the rotor activity pattern. Aerodynamic performance data describes a relation between obtainable power output (how well does the rotor perform) and aerodynamic conditions (wind speed and direction relative to rotor blade alignment). The aerodynamic performance of a rotor depends on a lot of different internal and external factors. The more factors are taken into account, the more reliably the aerodynamic performance data can be used to predict what control settings will lead to what power output.

In its simplest form, the method according to the invention takes the operational state of the whole turbine into account for determining one optimal set of control settings (e.g. for tip speed ratio and blade pitch) that is applied to all active rotors. In a more advanced version, each individual rotor receives its own separate set of control settings.

Alternatively, some of the rotors or groups of rotors of the same multi-rotor wind turbine may receive identical settings, while other rotors or groups of rotors get different settings. With a multi-rotor wind turbine a so-called wake mixing effect changes the axial pressure profile and consequently the performance of the wind turbine. A multi-rotor wind turbine has to run with a higher thrust than a single rotor wind turbine, and so the optimal tip speed ratio is also higher. As a result, the optimal control settings for each one of the rotors do not only depend on the wind speed and the properties of the rotor itself, but also on the activity of the nearby rotors of the same multi-rotor wind turbine. When, for example, one rotor in a two rotor wind turbine is down for maintenance, the other rotor will have a performance curve that is the same as it would have been in a single rotor wind turbine. But when the first rotor is activated, the wake mixing effect will change the air flow and a higher tip speed ratio will be needed for operating at optimum efficiency. If, for whatever reason, one of the rotors only operates in a derated mode (e.g. on 50% of its capacity), the optimal tip speed ratio will be different again. By taking into account the rotor activity pattern of the multi-rotor wind turbine, it is assured that the multi-rotor wind turbine is operated with optimal control settings under many different circumstances. When operating with the tip speed ratio of a rotor is kept constant, such as in region II. Preferably, this constant tip speed ratio is increased when more rotors are active or when the rotor speed of one or more of the rotors increases. Exemplary control settings to be determined and submitted to the production controllers of the respective rotors are a pitch angle and an optimal tip speed ratio. Instead of the optimal tip speed ratio, an optimal rotational speed of the rotor may be determined and submitted. When the wind speed is known the tip speed ratio and the rotational speed of the rotor can easily be converted to one another. In the following, the term 'speed of rotation' may be used and should be interpreted as referring to the tip speed ratio, the rotational speed of the rotor, or equivalent measures for the speed of rotation. Wth Operational data representative of a current power output', we both mean data from which the current power output is directly derivable as well as parameters that can be used, together with other parameters, to calculate and/or estimate the current power output. Preferably, the operational data received from the respective production controllers comprises at least two of a current power output, an optimal pitch angle, a current rotor speed and a current wind speed. Again, rotor speed may be substituted by tip speed or tip speed ratio. Based on this operational data and the rotor specific aerodynamic performance data, an optimal pitch angle and speed of rotation can be determined and submitted to the respective production controllers. According to the invention, not only the operational data of the rotor itself, but also that of its neighbouring rotors is taken into account.

The aerodynamic performance data may be stored in the form of lookup tables, providing power coefficients for different rotation speeds, pitch angles and rotor activity patterns, the rotor activity patterns being derivable from the operational data of each one of the at least two rotors. Of course, the lookup tables may be based on different useful parameters, such as wind speeds, power outputs, thrust or thrust coefficients.

The rotor activity patterns indicate which rotors are active, not running at full capacity or down for maintenance, safety or other reasons. Different activity patterns lead to different optimal blade pitch, tip speed ratios and/or other operational parameters. The activity patterns are derivable from the combined operational parameters of the individual rotors and may, e.g., be expressed as a simple list of 0s and 1 s for inactive and active rotors or a list of percentages indicating at what percentage of its capacity each rotor is working. 'Capacity', herein, may stand for, e.g., rated power output, maximum rotational speed, or optimal power output or rotational speed given the current wind speed. Other ways of expressing the actual activity of each rotor will be foreseeable for the skilled person. In an advantageous embodiment of the method according to the invention, the optimal control settings are determined such as to maximize a power coefficient in view of the received operational data of all rotors of the multi-rotor turbine. For example, the optimal control settings are determined such as to select for at least one of the rotors the control settings corresponding to a maximized power coefficient. This may, e.g., be implemented by selecting for at least one of the rotors a combination of tip speed ratio and pitch angle, which combination corresponds to the maximized power coefficient. According to another aspect of the invention, a multi-rotor wind turbine is provided. The multi-rotor wind turbine comprises at least two rotors and an optimal operation manager, operatively coupled to the production of the respective rotors controllers and configured to perform a method as described above. As an example, a multi-rotor wind turbine may comprise four rotors, but wind turbines with less or more rotors may be operated in the same way. Furthermore a computer program product is provided, comprising computer code for performing such a method.

It will be appreciated that preferred and/or optional features of the first aspect of the invention may be combined with the other aspects of the invention. The invention in its various aspects is defined in the independent claims below and advantageous features are defined in the dependent claims below.

BRIEF DESCRIPTION OF THE DRAWINGS For a better understanding of the invention, some embodiments of the invention will now be described with reference to the following drawings, in which:

Figures 1 A and 1 B schematically show two examples of a multi-rotor wind turbine in which the method according to the invention may be implemented.

Figure 2 shows a schematic representation of the most relevant functional parts of the multi-rotor wind turbine.

Figure 3 shows a diagram for illustrating different operation modes of a wind turbine rotor. Figure 4 shows a diagram for illustrating the nature of some of the aerodynamic performance data that may be used in the method according to the invention.

Figure 5 shows a flow diagram of a method according to the invention

DETAILED DESCRIPTION

Figure 1A schematically shows a multi-rotor wind turbine 100 in which the method according to the invention may be implemented. The currently most common type of wind turbine is the horizontal axis wind turbine (HAWT). It usually has a nacelle placed on top of a high vertical pole, with the rotor blades attached to a horizontal low speed shaft that extends from the nacelle. The nacelle may comprise a gear box for coupling the low speed shaft to an also horizontal high speed shaft that is connected to the generator. Power generated by the generator is transported to the ground by a power line running through the core of the pole, where it can be used or stored immediately or be coupled to a larger power grid. Where, in the past, wind turbines and their rotor blades have grown bigger and bigger to satisfy the increasing demand for wind powered electricity, recently also another strategy has been introduced; the multi-rotor wind turbine 100. Instead of one nacelle with one rotor on the top of the pole, this multi-rotor wind turbine 100 comprises two or more nacelles, here shown with four nacelles 111 , 121 , 131 , 141 , each carrying their own rotor 110, 120, 130, 140. In order to avoid the rotor blades of different rotors 110-140 running into each other, the nacelles 1 11-141 are spaced from each other by attaching them to arms 105, originating from the pole. In this example, all four arms 105 originate from the same top part of the wind turbine 100, but one or more arms may also be attached to a lower part of the pole. Different

constructions for installing four rotors 1 10-140 on one pole are, of course, foreseeable. An example of a multi-rotor wind turbine different from the one shown in Fig. 1A is shown in Fig. 1 B, where the four rotors are arranged in two layers, and each layer can be yawed independently. While in the current examples all four rotors 110-140 rotate in the same vertical plane, it is also possible to put one or more rotors in different planes. Alternative configurations are also possible, where multiple poles are situated close enough to each other for a wake mixing effect to occur. Twin rotor wind turbines have been designed comprising two poles, organized in a V-shape. In the following exemplary embodiments, the multi-rotor wind turbine has four rotors 1 10-140. It is, however, to be noted that a multi-rotor wind turbine, may alternatively comprise 2, 3, 5, 6 or more rotors. Figure 2 shows a schematic representation of some functional parts of the multi-rotor wind turbine 100. For conciseness only, the third rotor 130 is omitted. Parameters being received from and submitted to the fourth rotor 140 are denoted by the subscript n for indicating that the invention also works for wind turbines with more than four rotors. In a similar manner, the wind turbine also works with two, three, five or more rotors. Each rotor 110, 120, 140 is electronically coupled to a respective production controller 115, 125, 145. The production controller 1 15-145 is operable to receive sensor readings from all types of sensors useful for the optimized control of the wind turbine 100. Such sensor readings may represent (and are not limited to) wind speed, speed of rotation, gear box settings, pitch angle, yaw angle and power output. Depending on what they are actually measuring, the sensors may, e.g., be installed on the rotor blades, in the rotor hub, in the gearbox or the generator or on a brake or rotor shaft. Wind speed, for example, may be measured centrally with only one wind sensor or at each rotor separately using one or more wind speed sensors installed on each rotor. Alternatively wind speed can be estimated based on a measured power output and the measured values of other relevant parameters. It is to be noted that, also for wind speed estimation, it may be

advantageous to take the wake mixing effect into account. Nearby rotors may influence the relation between wind speed, power output, blade pitch and rotational speed of a rotor.

The production controller 115-145 processes, and optionally stores, all the incoming information and adjusts control settings like desired pitch angle, yaw angle and speed of rotation in such a way to control and optimize the power output of the rotor 1 10-140. Specific examples of control strategies are described below with reference to figures 3 and 4.

According to the invention, the control settings of one rotor 1 10-140 do not only depend on the operational data originating from its own sensors, but also on the operational data of all (or some of) the other rotors 110-140. The inventors have found out that just controlling the four rotors 1 10-140 as if they belong to four separate single-rotor wind turbines, does not lead to the best possible power output. As it turns out, with a multi- rotor wind turbine a so-called wake mixing effect changes the axial pressure profile for the neighbouring rotors and consequently the performance of the wind turbine 100. It has been realized that a multi-rotor wind turbine 100 has to run with a higher thrust than a single rotor wind turbine, and so the optimal tip speed ratio is also higher. As a result, the optimal control settings for each one of the rotors 110-140 does not only depend on the wind speed and the properties of the rotor 1 10-140 itself, but also on the activity of the nearby rotors 1 10-140 of the same multi-rotor wind turbine 100.

When, for example, one rotor in a two rotor wind turbine is down for maintenance, the other rotor will have a performance curve that is the same as it would have been in a single rotor wind turbine. But when the first rotor is activated, the wake mixing effect will change the air flow and a higher tip speed ratio will be needed for operating at optimum efficiency. If, for whatever reason, one of the rotors only operates in a derated mode (e.g. at 50% of its capacity), the optimal tip speed ratio will be different again. By taking into account the operational parameters of each one of the rotors in the multi-rotor wind turbine, it is assured that the multi-rotor wind turbine is operated with optimal control settings under many different circumstances.

In this schematic representation, the production controllers 1 15-145 are situated inside the respective nacelles 1 1 1-141 of their rotors, but alternative setups are foreseeable. For example, a central control unit may be provided for controlling the power production of each one of the rotors 1 10-140, or all data may be communicated wirelessly to a cloud server that process the incoming data and returns control instructions via the same or a similar communication signal. Consequently, the optimal operation manager 200 may be implemented in software code running on the same computer as is used for the four production controllers.

Figure 3 and 4 show diagrams for illustrating different operation modes of a wind turbine rotor. First this is done for wind turbine rotors in general, and then for a multi-rotor wind turbine 100 according to the invention. It is to be noted that the curves shown in this diagram only serve to illustrate a typical relation between the rotational speed, blade pitch and power output of one rotor. In practice, different rotors and different

circumstances may lead to different curves. For example, as already explained when discussing the background of the invention, a region I I I may not be part of the curve at all,

Depending on wind speed (v) , wind turbines generally operate in four different modes, also called 'regions'. In region I , the rotor rotates at its minimum operational speed (ώ) . With increasing wind speeds (v) , the blade pitch (Θ ) is adapted to increase the power output (P) of the wind turbine. In region I , the wind turbine does not run at its optimum efficiency. For every wind speed (v), there is an optimal combination of pitch (Θ ) and tip speed ratio (λ, i.e. the rotational speed ω of the rotor times its radius R, divided by the wind speed v) at which the power coefficient (C p , a measure for the rotor's efficiency, see figure 4) is at a maximum. At the end of region I, the pitch (Θ ) and the rotational speed (ώ) of the rotor are such that the rotor performs at the top of its performance curve.

There region I I starts. With further increase of the wind speed (v) , the rotational speed (ώ) of the rotor is changed proportionally while the pitch (Θ ) is kept constant. As a result the tip speed ratio (X) is also kept constant and the rotor keeps on performing with a maximum power coefficient (C p ). When the rotor speed (ώ) reaches its maximum, the turbine enters region I II . Like in region I, only a change of pitch (Θ ) can make the power output (P) increase further with increasing wind speeds (v). When a maximum power output (P) is reached (also called rated power, P 0 ), the wind turbine operates in region IV. In region IV, the blade pitch (Θ ) may be adapted to avoid damage to the wind turbine at even higher wind speeds (v) . When the wind speed (v) is too high, the turbine is brought to a halt for safety reasons.

A typical wind turbine, under typical conditions, operates in region II for more than 50% of the time. Knowing the optimal tip speed ratio (X) and blade pitch (θ ') of a rotor thus is important for optimal operation of a wind turbine. The exact shape of the performance curve depends on the dimensions and geometry of the rotor and its blades. For single rotor wind turbines, determining and using the performance curve is a well-established practice. Computer models of the wind turbine and its rotor blades, together with an analysis of the air around the operating wind turbine (e.g. using Blade Element

Momentum theory - BEM), are used to establish the rotor-specific performance curves.

While the advantages of the current invention may be most notable for a multi-rotor wind turbine operating in region I I , it will also be beneficial in the other partial load regions I and I I I. Although the rotational speed in these regions may be fixed (rated speed in region I I I, minimal speed in region I), the wake mixing caused by the neighbouring rotors may change the optimal pitch angle and/or the optimal values for other operational parameters of the rotor.

Figure 4 shows an example of such a performance curve. The performance curve, or a collection of multiple performance curves, may be used as the aerodynamic performance data for use in the method according to the invention. The performance curves of Figure 4 shows the power coefficient (C p ) of one rotor for different tip speed ratios (A) and at a fixed pitch (Θ). For different pitch angles, the curves change. So for every wind speed, there is not only an optimal tip speed ratio (X), but also an optimal blade pitch (0 * ), which can be derived from a 3-dimensional Cp-λ-θ plot. Lookup tables or modelling functions may be used to determine the optimal control settings (X, Θ * ).

Curves are shown for three different situations. The dotted line indicates the C p -A curve for a single rotor system, or for a multi-rotor wind turbine in which only one rotor is active. The dashed line shows that the C p -X curve changes when the rotor is generating power near another active rotor. An active rotor changes the local air stream and pressure profiles. When four rotors are in full operation, the C p -A curve is shifted even more, as is shown by the solid line in Figure 4.

What can also be observed in Figure 4 is that the wake-mixing effect causing the changes in the C p -A curve shifts the optimal power coefficient (C p ) to a higher X This means that when more rotors of a multi-rotor wind turbine are active, each rotor has to operate with a higher rotational speed to achieve optimum efficiency. Other factors that influence the C p -X curve are the number of and distance to the additional active rotors, the shape and dimensions of such rotors and operational parameters like its rotational speed, blade pitch or power output (e.g. as a percentage of P 0 ).

It is therefore that, according to the invention, the control settings of each rotor 1 10-140 are not determined by its own production controller 1 15-145 and its own operational parameters. The relevant operational parameters of each rotor 1 10-140, are sent to a central optimal operation manager 200, which is then able to determine the optimal control settings for each individual rotor 1 10-140 based on the combined operational data of all the rotors 1 10-140. For this purpose, the optimal operation manager 200 has access to a large set of lookup tables 210 and/or modelling functions 210 that describe the relations between the aerodynamic performance and operational parameters of the rotors. Relevant operational parameters of a rotor are its blade pitch, yaw angle, rotational speed, thrust and power output. Wind speed can be seen as an operational parameter because it defines the context in which the rotor is operating and relations between different operational parameters, such as rotational speed and tip speed ratio. Historical data of the operational parameters can optionally be stored and used by the optimal operation manager, in order to account for dynamic effects. One of the operational parameters to take into account may be a rotor activity pattern, indicating which rotors are active, not running at full capacity or down for maintenance, safety or other reasons. As shown in Figure 4, different activity patterns lead to different optimal blade pitch and tip speed ratios. The activity patterns are derivable from the combined operational parameters of the individual rotors.

Figure 5 shows a flow diagram of a method according to the invention. The method comprises an input step 51 , a processing step 52 and a control step 53. In the input step, the optimal operation manager 200 receives, from the respective production controllers 115-145 of the at least two rotors 110-140, operational data representative of a current power output of the rotors 110-140. In the processing step, the optimal operation manager 200 determines optimal control settings for each rotor 110-140, based on the received operational data and the aerodynamic performance data 210 that is stored in or accessible by the optimal operation manager.

It is to be noted that in a similar way, the same or similar aerodynamic performance data may be used for improving the accuracy of, e.g., a wind speed estimator or other .