METHOD FOR DETECTING AN ACCRETION OF ICE ON A WIND TURBINE

Abstract

A method for detecting an accretion of ice on a rotor blade of a rotor of a wind turbine is provided. The can be operated at a variable rotational speed. The method includes recording a wind speed of a wind acting upon the rotor, recording an operating variable that is dependent on the wind speed and comparing the recorded operating variable or the recorded wind speed with a reference variable of a characteristic wind-speed-dependent operating-variable curve of the wind turbine. The characteristic operating-variable curve indicates an operating variable assumed to be optimal in dependence on the wind speed. The method includes detecting an accretion of ice on the rotor blade if the recorded operating variable or the recorded wind speed, deviates from the reference variable by at least a predetermined minimum deviation specified in dependence on the wind speed.

Claims

1. A method for identifying an accretion of ice on a rotor blade of a rotor of a wind turbine, comprising: recording a wind speed of wind acting on the rotor, the rotor being operable at a variable rotational speed; recording an operating variable that is dependent on the wind speed; comparing the recorded operating variable or the recorded wind speed with a reference variable of a characteristic operating curve of the wind turbine, wherein the characteristic operating curve indicates an operating variable based on the wind speed; and identifying the accretion of ice on the rotor blade in response to the recorded operating variable or the recorded wind speed deviating from the reference variable by at least a minimum deviation, wherein the minimum deviation is specified based on the wind speed.

2. The method as claimed in claim 1, wherein: the recorded operating variable is a recorded output power generated by the wind turbine from the wind, a recorded rotor rotational speed of the wind turbine, and/or a recorded blade angle set to limit the output power, comparing the recorded operating variable or the recorded wind speed with the reference variable includes: comparing the recorded output power with a reference power of a characteristic operating power curve of the wind turbine, or comparing the recorded wind speed with a wind speed assigned to the recorded output power via the characteristic operating power curve, the characteristic operating power curve indicating a maximum output power based on the wind speed, comparing the recorded rotor rotational speed with a reference rotational speed of a characteristic rotational speed curve of the wind turbine, or comparing the recorded wind speed with a wind speed assigned to the recorded rotor rotational speed via the characteristic rotational speed curve, the characteristic rotational speed curve indicating a maximum rotor rotational speed based on the wind speed, and/or comparing the recorded blade angle with a reference angle of a characteristic blade angle curve of the wind turbine, or comparing the recorded wind speed with a wind speed assigned to the recorded blade angle via the characteristic blade angle curve, the characteristic blade angle curve indicating an optimal blade angle based on the wind speed, and identifying the accretion of ice on the rotor blade in response to the recorded operating variable or the recorded wind speed deviating from the reference variable by at least the minimum deviation includes identifying the accretion of ice on the rotor blade in response to: the recorded output power deviating from the reference power by at least a minimum deviation power, or the recorded wind speed deviating from the wind speed assigned to the recorded output power by minimum deviation speed, the minimum deviation power or the minimum deviation speed being specified based on the wind speed, the recorded rotor rotational speed deviating from the reference rotational speed by at least a minimum deviation rotational speed or the recorded wind speed deviating from the wind speed assigned to the recorded rotor rotational speed by the minimum deviation speed, the minimum deviation rotational speed or the minimum deviation speed being specified based on the wind speed, and/or the recorded blade angle deviating from the reference angle by at least a minimum deviation angle or the recorded blade angle deviating from the wind speed assigned to the recorded blade angle deviates by the minimum deviation speed, the minimum deviation angle or the minimum deviation speed being specified based on the wind speed.

3. The method as claimed in claim 2, wherein recording the operating variable includes: recording the output power generated by the wind turbine from the wind as the operating variable for wind speeds below a rated wind speed; and recording the blade angle set to limit the output power as the operating variable for wind speeds above the rated wind speed.

4. The method as claimed in claim 2, wherein the characteristic blade angle curve is specified depending on a reduced operating limit that is a reduced power limit and/or a reduced rotational speed limit, and the minimum deviation angle is specified based on the reduced operating limit.

5. The method as claimed in claim 1, wherein the minimum deviation is specified using a deviation curve, and the deviation curve indicates a behavior of the operating variable based on the wind speed, and wherein: the deviation curve deviates from the characteristic operating curve by the minimum deviation, and a varying distance is obtained between the characteristic operating curve and the deviation curve, the deviation curve is not shifted by a constant wind speed value relative to the characteristic operating curve, and/or the deviation curve deviates from the characteristic operating curve by wind speed values of differing magnitude.

6. The method as claimed in claim 1, wherein: in partial-load operation, the minimum deviation power, increases at least one of: from a first start-up wind speed, from a wind speed wind speed relevant threshold or continuously, and in full-load operation, a minimum deviation angle decreases with increasing wind speed.

7. The method as claimed in claim 1, wherein the wind turbine is operated using a rotational speed power characteristic having an output power of the wind turbine set based on a recorded rotational speed of the rotor.

8. The method as claimed in claim 1, wherein the accretion of ice on the rotor blade is detected only in response to a recorded outside temperature being below a maximum temperature.

9. The method as claimed in claim 1, comprising: setting the characteristic operating curve during operation of the wind turbine, and/or setting the characteristic operating curve during operation of the wind turbine includes recording operating deviations representative of deviations of the recorded operating variable from an operating variable according to the characteristic operating curve and that are not due to the accretion of ice, and wherein: the minimum deviation is selected such that it is greater in magnitude than a corresponding operating deviation.

10. The method as claimed in claim 21, comprising: recording the operating deviations as standard deviations; combining the operating deviations as a deviation band, and selecting or changing the deviation curve to be below the deviation band relative to the operating variable; and/or setting a respective minimum deviation or the deviation curve during operation of the wind turbine based on the characteristic operating curve and/or the recorded operating deviations.

11. The method as claimed in claim 10, wherein the respective minimum deviation or the deviation curve is set depending on an ambient temperature of the wind turbine.

12. The method as claimed in claim 10, comprising: setting the respective minimum deviation, or the deviation curve depending on the wind speed and at least one property of the wind acting on the rotor; and at least one of: selecting the at least one property of the wind from a list of properties including: a turbulence intensity of the wind, a vertical shear of the wind, a horizontal shear of the wind, a wind direction,. and a change in wind direction over height; and/or determining the minimum deviation depending on a classification of the wind, the classification of the wind depending on at least one wind field parameter formed by the at least one property of the wind.

13. The method as claimed in claim 10, wherein the respective minimum deviation or the deviation curve depends on a blade profile of the rotor blade.

14. The method as claimed in claim 1, comprising: determining multiple times in a predefined test period whether the recorded operating variable or the recorded wind speed deviates from the reference variable by at least the minimum deviation; and identifying the accretion of ice in response to a positive determination for a majority of the multiple times.

15. The method as claimed in claim 1, comprising: setting three wind speed sections for a lower wind-speed range, setting the three wind speed sections including: setting the three wind speed sections to be equally sized; setting a first wind speed section that extends from a start-up wind speed at which the wind turbine starts up to a first section wind speed; setting a second wind speed section that extends from the first section wind speed to a second section wind speed; and setting a third wind speed section that extends from the second section wind speed to a rated wind speed, wherein the minimum deviation in each of the three wind speed sections increases with increasing wind speed, and at least one of: the minimum deviation increases continuously from the first wind speed section to the third wind speed section with increasing wind speed, a mean minimum deviation of the third wind speed section is at least 50% greater than a mean minimum deviation of the second wind speed section, and/or the mean minimum deviation of the second wind speed section is at least 50% greater than a mean minimum deviation of the first wind speed section.

16. A wind turbine, comprising: at least one rotor operable at a variable rotational speed and having at least one rotor blade; a wind speed measuring device configured to record a wind speed of wind acting on the rotor; and a controller configured to: record an operating variable that is dependent on the wind speed; compare the recorded operating variable or the recorded wind speed with a reference variable of a characteristic operating curve of the wind turbine, wherein the characteristic operating curve indicates an operating variable based on the wind speed; and detect an accretion of ice on the rotor blade in response to the recorded operating variable or the recorded wind speed deviating from the reference variable by at least a minimum deviation, the minimum deviation being set depending on the wind speed.

17. (canceled)

18. The method as claimed in claim 3, wherein: recording the operating variable includes recording the generated output power and the blade angle; and identifying the accretion of ice on the rotor blade in response to the recorded operating variable or the recorded wind speed deviating from the reference variable by at least the minimum deviation includes identifying the accretion of ice on the rotor blade in response to: the recorded output power deviating from the reference power by at least a minimum deviation power, the recorded blade angle deviating from the reference angle by at least a minimum deviation angle, and/or the recorded wind speed deviating from a wind speed assigned to the recorded output power or from a wind speed assigned to the recorded blade angle by a minimum deviation speed.

19. The method as claimed in claim 6, wherein the minimum deviation angle or a minimum deviation speed increases in response to the wind turbine working in a reduced mode in which an output power is reduced to a power value that is at most 50% of a rated power of the wind turbine.

20. The method as claimed in claim 8, wherein the accretion of ice on the rotor blade is detected only in response to the recorded outside temperature being below the maximum temperature and above a minimum temperature that is set depending on the wind speed.

21. The method as claimed in claim 9, comprising: selecting or changing a deviation curve such that the characteristic operating curve having the operating deviations is not below the deviation curve.

22. The method as claimed in claim 1, wherein the characteristic operating-variable curve indicates the operating variable assumed to be optimal based on the wind speed.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWINGS

[0087] The invention is now explained in more detail below on the basis of exemplary embodiments, with reference to the accompanying figures.

[0088] FIG. 1 shows a wind turbine in a perspective representation.

[0089] FIG. 2 shows a diagram of a deviation of an operating variable between normal operation and operation with accretion of ice.

[0090] FIG. 3 shows a diagram similar to FIG. 2, but for limited-power operation.

[0091] FIG. 4 shows a diagram with wind-speed-dependent behaviors of operating variables together with proposed minimum deviations, or assigned deviation curves.

[0092] FIG. 5 shows a schematic block diagram for determining deviation curves.

[0093] FIG. 6 shows a block diagram for a variant for detecting an accretion of ice based on a wind classification.

DETAILED DESCRIPTION

[0094] FIG. 1 shows a wind turbine 100 having a tower 102 and a nacelle 104. Arranged on the nacelle 104 is a rotor 106 that has three rotor blades 108 and a spinner 110 is. During operation, the wind causes the rotor 106 to rotate, thereby driving a generator in the nacelle 104.

[0095] FIG. 2 shows a diagram of the behavior of a deviation Δ of a recorded wind speed from a wind speed assigned to a recorded operating variable, and the deviation, which is thus called the wind speed deviation, is represented as a function of the wind speed V. The wind speed V is plotted on the abscissa with the unit m/s. The deviation Δ is normalized to a deviation measure. The deviation at rated wind speed V.sub.N may be used as a deviation measure for normalization. It was recognized that such wind speed deviations between normal operation and operation with accretion of ice depend on the wind speed and therefore further investigations were carried out. In simulations, operating variables in an operation without accretion of ice were compared with respectively the same operating variable in an operation with accretion of ice, and the deviations that resulted for the assigned wind speeds were considered in dependence on wind speed. Such a result is represented in FIG. 2. In a partial-load operation from a start-up wind speed V.sub.A to the rated wind speed V.sub.N, FIG. 2 shows the deviation of a wind speed assigned to the recorded output power, which is denoted to here as the wind speed deviation Δ.sub.P assigned to the output power, between undisturbed operation and operation with accretion of ice. The representation in FIG. 2, and the same applies to FIG. 3, shows the behavior of the respective deviation in terms of magnitude.

[0096] It can be seen that at the start-up wind speed V.sub.A, which here is about 2 m/s, the wind speed deviation Δ.sub.P has a comparatively low value of about 0.1. Up to the rated wind speed V.sub.N, this wind speed deviation Δ.sub.P assigned to the output power increases to about the value 1. From this was derived the concept that a minimum deviation should likewise increase from the start-up wind speed V.sub.A to the rated wind speed V.sub.N, and the minimum deviation should in each case be somewhat less than the shown wind speed deviation Δ.sub.P assigned to the output power.

[0097] At rated wind speed V.sub.N, for instance, the wind turbine is operated at rated power and the rotor blades are then gradually turned further out of the wind as wind speed increases. Therefore, from rated wind speed V.sub.N onwards, a blade angle is considered as an operating variable for detecting an accretion of ice. FIG. 2 in this connection shows the deviation of the wind speed assigned to the blade angle, which is referred to here as the wind speed deviation Δ.sub.α assigned to the blade angle, for operation without accretion of ice and with accretion of ice. Here, too, the deviation is shown according to magnitude. Here, too, it was found that the wind speed deviation Δ.sub.α assigned to the blade angle between normal operation and operation with accretion of ice depends on the wind speed. It was found in this case that the dependence is such that the wind speed deviation Δ.sub.α assigned to the blade angle substantially decreases with increasing wind speed. From this, the rule was derived that a minimum deviation for the wind speed deviation assigned to the blade angle for detecting an accretion of ice is also to be specified depending on the wind speed, but in such a way that the minimum deviation is of a somewhat lesser magnitude than the examined wind speed deviation assigned to the blade angle for operation with and without accretion of ice.

[0098] FIG. 3 shows basically a very similar diagram to FIG. 2, but with the difference that here the wind turbine is strongly curtailed. It is in fact curtailed to a power that it would attain in partial-load operation at about 8 meters per second (m/s). In the figure, this wind speed is plotted on the abscissa as reduced wind speed V.sub.R. It is thus also only expedient to consider the output power up to this reduced wind speed V.sub.R, and therefore the diagram of FIG. 3 includes a wind speed deviation Δ.sub.p′ assigned to the output power only from the start-up wind speed V.sub.A up to this reduced wind speed V.sub.R. It should be noted that the representation of FIG. 3, in contrast to the representation of FIG. 2, does not start at 0, but at the start-up wind speed V.sub.A, which is technically irrelevant. Preferably, it is proposed that the wind speed deviation assigned to the output power be increased with increasing wind speed only from a wind-speed-relevant threshold, which in this case describes a wind speed of approximately 3 m/s.

[0099] The behavior of the wind speed deviation Δ.sub.p′ assigned to the output power in FIG. 3 corresponds in principle, also quantitatively, to the behavior of the wind speed deviation Δ.sub.p assigned to the output power in FIG. 2, except that in FIG. 3 the behavior of the wind speed deviation Δ.sub.p′ assigned to the output power ends at the reduced wind speed V.sub.R.

[0100] Thus, the blade pitch control takes over the control of the wind turbine from the reduced wind speed V.sub.R onwards. Accordingly, the behavior of a wind speed deviation Δ.sub.α′ assigned to the blade angle is investigated and represented from the reduced wind speed V.sub.R onwards. However, the behavior of the wind speed deviation Δ.sub.α′ assigned to the blade angle in the case of curtailed operation according to FIG. 3 differs significantly from the behavior of the wind speed deviation Δ.sub.α assigned to the blade angle in FIG. 2. In FIG. 3, it can be seen that the behavior of the wind speed deviation Δ.sub.α′ assigned to the blade angle increases, starting from the reduced wind speed V.sub.R, with increasing wind speed, even beyond the rated wind speed V.sub.N. Only at high wind speeds of over 22 meters per second (m/s) is there a slight decrease.

[0101] From this, it was derived as a control rule to set the minimum deviation for the wind speed deviation assigned to the blade angle lower with increasing wind speed in the case of non-curtailed operation of the wind turbine, whereas, in the case of a significantly curtailed wind turbine it is set higher with increasing wind speed.

[0102] Shown schematically in FIG. 4 is a diagram with a power curve P and a blade angle curve α. In particular, the linear behaviors of the power P and of the blade angle a are highly simplified and serve as illustration.

[0103] FIG. 4 is based on a closed-loop control concept in which the wind turbine is started when the wind speed V attains the start-up wind speed V.sub.A. This start-up wind speed V.sub.A is selected in such a way that a small amount of power can already be generated. In partial-load operation, which extends from the start-up wind speed V.sub.A to the rated wind speed V.sub.N, a higher power P can be generated with increasing wind speed. The closed-loop control system may be designed in such a way that a rotational-speed power characteristic is used as a basis and, based on this rotational-speed power characteristic, a power P is set depending on a recorded rotational speed. Ideally, this results in a value on the power curve P for each wind speed V. This dependency is known, or should be known, as precisely as possible.

[0104] When the wind speed attains the rated wind speed V.sub.N and continues to increase, the power P attains the value of the rated power P.sub.N. Until then, the blade angle α has remained constant, in the example at 5°. From the rated wind speed V.sub.N onwards, however, the rotor blades are then turned out of the wind as the wind speed increases, and the blade angle α increases accordingly as the wind speed continues to increase. A blade angle a that is dependent on the wind speed can also be derived from this. This is shown by the correspondingly solid blade-angle curve for the blade angle α.

[0105] Both the power P and the blade angle α each form an operating variable of the wind turbine. This operating variable is now monitored to ascertain whether it falls below the deviation curve, which is shown as a dotted line in FIG. 4. For partial-load operation, the power P is the relevant operating variable, and the associated deviation curve accordingly shows the behavior of a power P′. If the power P′ falls below this deviation curve, an accretion of ice is assumed.

[0106] Correspondingly, a deviation curve for the blade angle α′ for the wind-speed range from the rated wind speed V.sub.N is also plotted with a dotted line. If the blade angle α falls below the deviation curve of the blade angle α′ in this range, an accretion of ice is assumed.

[0107] It is now proposed that the minimum deviation for both the operating variable of the power P and the operating variable of the blade angle a be specified in dependence on the wind speed. This means that the respective deviation curve deviates from the assigned operating-variable curve to a different extent depending on the wind speed. For partial-load operation, the minimum deviation in this case increases with increasing wind speed. Accordingly, the distance between the power curve P and the deviation curve of the power P′ increases with increasing wind speed.

[0108] Alternatively, here too the assigned wind speed may be considered, i.e., in illustrative terms, the deviation between the power curve P and the deviation curve of the power P′ in the horizontal direction. This also increases with increasing wind speed.

[0109] For the blade angle as an operating variable in full-load operation, i.e., from rated wind speed V.sub.N, it is proposed that the minimum deviation become smaller with increasing wind speed. The distance between the operating-variable curve of the blade angle a and the deviation curve of the blade angle α′ therefore decreases with increasing wind speed. Both the deviation curve of the P′ and the deviation curve of the blade angle α′ have indicated steps at which the distance between the deviation curve and the assigned operating-variable curve then changes. However, this is only one possibility for changing the respective underlying minimum deviation and a straight behavior of each of the deviation curves may also be considered.

[0110] Alternatively, here too the assigned wind speed may be considered, i.e., in illustrative terms, the deviation between the operating-variable curve of the blade angle a and the deviation curve of the blade angle α′ in the horizontal direction. This also decreases with increasing wind speed.

[0111] FIG. 4 additionally shows a variation, namely that a transition range from partial-load operation to full-load operation is provided. According to such a transition range, the rotor blades are already adjusted before attaining the rated wind speed V.sub.N, this being represented by a dashed line and marked as blade angle α*. As a result, the power P is also reduced somewhat earlier, and accordingly a power P* is also plotted with a dashed line for the transition range. Accordingly, the respective deviation curve also changes for the range and this changed section is plotted as a dotted line section for both the deviation curve of the power P′* and for the deviation curve of the blade angle α′*. In the transition range, it is thus proposed to monitor, both for the deviation curve of the power P′* and for the deviation curve of the blade angle α′*, whether the corresponding operating variable falls below the corresponding deviation curve.

[0112] FIG. 5 shows an embodiment of a first specification unit (e.g., controller) 500, which receives a wind speed V, an outside temperature T and wind field data W as input variables. In addition, the specification unit 500 takes into account whether the wind turbine is curtailed, and if so, to what extent. For this purpose, it receives the curtailment power P.sub.R. If this curtailment power P.sub.R corresponds to the rated power of the wind turbine, it is not curtailed. However, if it is significantly lower, for example, and only has a value of, for example, 40% of the rated power P.sub.N, the specification unit 500 detects that reduced, or curtailed, operation is in effect.

[0113] Depending on these input variables, a minimum deviation for the power Δ.sub.P and a minimum deviation for the blade angle Δ.sub.α are then determined. These may be used directly as criteria for detecting an accretion of ice. Alternatively, a minimum deviation for an assigned wind speed may be determined in each case in order to detect an accretion of ice in dependence thereon.

[0114] Alternatively or additionally, a deviation curve for the power P.sub.V′ and a deviation curve for the blade angle α.sub.V′ may be determined in dependence thereon in a second specification unit (e.g., controller) 502. For this purpose, the second specification unit 502 requires, in addition to the respective minimum deviations, the operating-variable curves, namely for the power P.sub.V and the blade angle α.sub.V. In addition, the assignment to the respective wind speed V is required and for this purpose the wind speed V may be transferred from the first specification unit 500 to the second specification unit 502. Alternatively, a minimum deviation for an assigned wind speed may be determined in each case and transferred to the specification unit 502.

[0115] FIG. 6 shows a variant that takes at least the wind speed V and wind-field parameters W into account for the detection of an accretion of ice. For this purpose, these parameters are transferred to a classification block (e.g., controller). From these parameters the classification block 600 derives a wind class. This wind class is then transferred to a specification block (e.g., controller) 602, which additionally receives the wind speed V as an input variable. The specification block 602 then calculates a minimum deviation ΔP.sub.m from the wind speed and the determined wind class WK. FIG. 6 to that extent describes an example in which the considered operating variable represents the power of the wind turbine. This may also be applied equally and simultaneously to other operating variables such as, for example, the blade angle. Alternatively, the assigned wind speed may likewise be considered in each case instead of the operating variables.

[0116] The specification block 602 may make use of a stored characteristics map, which is indicated in the block 602 in FIG. 6. Such a characteristics map may have various wind-speed-dependent minimum deviations, and an example of this is shown in specification block 602, according to which three behaviors of the minimum deviation Δ, which are dependent on the wind speed V, are plotted and are marked as W.sub.k1, W.sub.k2 or W.sub.k3 according to wind class.

[0117] The specification block 602 therefore selects a corresponding characteristic depending on the wind class W.sub.k, which it has received from the classification block 600. Depending on the wind speed V, which it also receives, the specification block 602 then selects the specific minimum deviation A according to the specifically selected characteristic, namely depending on the wind speed V. The result is output and is shown at the output of the specification block 602 as a predetermined minimum deviation ΔP.sub.m.

[0118] For the purpose of evaluation, a comparison is first made with the measured power P.sub.i and the specified power Ps. The predetermined power P.sub.S may be taken from a specified wind-speed power curve, which is not shown further in FIG. 6. The comparison may be made by a difference in the summing element (e.g., controller) 604. Alternatively, wind speeds may also be compared here by using a measured, i.e., recorded, wind speed instead of the measured power P.sub.i. In addition, the assigned wind speed is determined, thus for example, read off, with the measured power P.sub.i in a specified wind-speed power curve, which may be referred to as a characteristic operating-power curve. This assigned wind speed is then input into the summing element 604 as the assigned wind speed for difference formation with the measured wind speed, instead of the specified power P.sub.S.

[0119] The result of this comparison, be it the comparison of the operating variables or of the wind speeds, is marked as deviation e in FIG. 6. This deviation e and the predetermined minimum deviation ΔP.sub.m are then input into the evaluation block (e.g., controller) 606. If the deviation of the wind speeds is used, the predetermined minimum deviation ΔP.sub.m is of course adjusted thereto accordingly.

[0120] In the evaluation block 606, it is checked whether the deviation e is greater than or equal to the predetermined minimum deviation ΔP.sub.m. If this is the case, an accretion of ice is assumed and the wind turbine can then be stopped and, in addition or alternatively, a heating process may be started in order to thaw the ice from the rotor blades. This is marked as S/H at the output of evaluation block 606.

[0121] Otherwise, the wind turbine will continue to operate normally, which is marked NORM on the output of evaluation block 606.

[0122] It was thus recognized that the use of a merely constant threshold value can have several serious disadvantages, and a proposed solution is presented which represents an improvement on the prior art. It is known that ice impairs the aerodynamic properties of the rotor blade profile. Typically, the lift coefficient and stall angle, i.e., the angle of attack at which flow separation occurs on the blade profile around which there is flow, are reduced. Furthermore, the drag coefficient increases, such that the lift-to-drag ratio also deteriorates significantly due to the reduced lift. It was recognized in this case that all of these influences on the aerodynamic properties of the rotor blade profile can result in a wind turbine reacting very differently to ice formation in the operating range over the wind speeds that are encountered, and that this wind-speed-dependent behavior can also vary greatly from one wind turbine type to another.

[0123] One of the reasons for this is that the properties of the rotor blade profiles used can vary from one wind turbine type to another, and the design philosophies chosen for rotor blade design can differ greatly due to different wind class requirements, which was also recognized. It is therefore proposed to depart from the constant threshold value for the detection of the accretion of ice in the power-curve method of the prior art, and to further develop the threshold value towards a wind-speed-dependent threshold value. Such a proposed wind-speed-dependent threshold value in this case is wind-turbine or rotor-blade specific, such that it is proposed to define the function of the threshold value separately for each turbine type. This is regarded as a significant further development compared to the prior art, which has only one universal threshold value.