Monitoring of a blade moment of a rotor blade

Abstract

A method of monitoring a blade moment of a rotor blade of a wind turbine rotor is provided. The wind turbine rotor (120) comprises at least three rotor blades (121, 122, 123). The method comprises obtaining measurements of a blade moment of each of the rotor blades (121, 122, 123) and determining a moment measure (MA, MB, MC) for each of the rotor blades (121, 122, 123) from the obtained blade moment measurements. Further, a moment imbalance measure (MimbA, MimbB, MimbC; MimbO) is determined for at least one of the rotor blades. The moment imbalance measure (MimbA, MimbB, MimbC; MimbO) is employed for detecting if a blade moment imbalance condition exists for at least one of the rotor blades (121, 122, 123).

Description

[0001] 2024PF00540 Subsequent Filing

[0002] Description

[0003] Monitoring of a blade moment of a rotor blade

[0004] FIELD OF THE INVENTION

[0005] The present invention relates to a method of monitoring a blade moment of a rotor blade of a wind turbine. It further relates to a respective monitoring system and to a computer program for monitoring a blade moment of a rotor blade of a wind turbine rotor.

[0006] BACKGROUND

[0007] To make energy generation by wind turbines more efficient, larger wind turbines having longer rotor blades are employed. The rotor blades are exposed to the wind, which applies respective forces to the blades. Drag and lift as well as changes in the rotational speed of the wind turbine rotor generate moments that act on the blades. The blades are flexible to a certain degree and thus bend in response to these moments. The moments experienced by the blades result in a loading of and strain in the blades.

[0008] To avoid damage to the blades and to evaluate blade loading, it is often desirable to measure the moments (in particular bending moments) applied to each blade. The rotor blades of the wind turbine may thus be equipped with respective sensors to detect the moments that are applied to the blades.

[0009] For example, the document CN 106704102 A describes a method for determining the balance status of blades of a wind turbine, wherein a ratio of a difference between two blade bending moments and one of these two blade bending moments is employed for detecting a blade balance condition.

[0010] Document WO 2023 / 078520 Al describes a method for reducing rotor imbalance in wind turbines, wherein blade connecting 2024PF00540 Subsequent Filing

[0011] 2

[0012] wires and pre-tension wires are employed between rotor blades.

[0013] It is desirable to make the detection of such blade moments reliable. In particular, the estimation of the blade loading or bending will be faulty if a respective sensor for measuring the blade moments fails.

[0014] The document US 2013 / 110414 Al describes a method of detecting a failure of a bending moment sensor. The method compares data output by the sensor with reference data to detect if the sensor has entered a fault mode. The method described in the document US 2022 / 178352 Al describes likewise a method for detecting a faulty blade load sensor. In plane moments of the rotor blades of the wind turbine rotor are compared to theoretical in plane moments to detect if a load sensor has a reduced reliability.

[0015] It is desirable to allow a more versatile detection of a fault of a respective moment sensor and in particular to make the detection more reliable. It is further desirable to also detect further conditions that may be detrimental to the operation of the wind turbine.

[0016] SUMMARY

[0017] Accordingly, there is a need to mitigate at least some of the drawbacks mentioned above and to provide an improved detection of a problem associated with the wind turbine operation, such as a fault of a sensor employed for measuring blade moments or a fault condition of wind turbine operation.

[0018] This need is met by the features of the independent claims. The dependent claims describe embodiments of the invention.

[0019] According to an aspect of the invention, a method of monitoring a blade moment of a rotor blade of a wind turbine rotor is provided. The wind turbine rotor comprises at least 2024PF00540 Subsequent Filing

[0020] three rotor blades. The method comprises obtaining measurements of a blade moment for each of the rotor blades, and determining a moment measure for each of the rotor blades from the obtained blade moment measurements. It further comprises determining a moment imbalance measure for at least one of the rotor blades. Determining the moment imbalance measure comprises comparing the moment measure determined for one of the rotor blades to the moment measure determined for the other rotor blades (the measure of the other rotor blades may thus be used as a reference). The moment imbalance measure may be employed to detect if a blade moment imbalance condition exists for at least one of the rotor blades.

[0021] By such method, problems associated with the wind turbine operation may thus be detected without the need for an external reference, such as theoretical values for in-plane moments, or for reference data indicating an expected sensor output, as in the prior art. Rather, the moment measure determined for the blade under investigation may be compared to the moment measure determined for the other rotor blades, which thus serve as a reference. A moment imbalance condition can thus be detected irrespective of the operating conditions or environmental conditions in which the wind turbine operates. For example, theoretical or modeled references are generally generated for a particular set of operating conditions of the wind turbine. If the wind turbine operates in a different operating mode, such as curtailed operation, noise-reduced operation, or the like, or if the environmental conditions change, such as changes in the air density, these theoretical values may no longer be valid. In contrast, since all three rotor blades operate in the same environment and at the same operating conditions, the present method employs a reliable reference that may be valid in any operating or environmental conditions and that may thus allow a reliable detection of any moment imbalance condition. The detection may thus be more versatile, as it can be used in any operating and environmental conditions. 2024PF00540 Subsequent Filing

[0022] 4

[0023] Detecting if a blade moment imbalance condition exists may for example comprise detecting if the measurement of the blade moment is faulty for at least one of the rotor blades and / or detecting if a fault condition of wind turbine operation exists. A fault condition of the wind turbine operation that may be detected based on the moment imbalance measure may for example comprise a pitch offset of a rotor blade, or a problem associated with the physical structure of a rotor blade, or the like.

[0024] Detecting a moment imbalance condition may allow the initiating of a mitigation action, such as activating a safety function (e. g. wind turbine curtailment or shut down) and / or issuing of an alarm and / or notification, upon which service personnel may for example determine the cause of the moment imbalance condition and remove it. If it is detected that the measurement of the blade moment is faulty, a mitigation action may also be taken automatically by replacing the faulty measurement, as indicated further below.

[0025] Optionally, the type of fault causing a detected moment imbalance condition may be determined based on the blade moment measurements and / or by measurements obtained from one or more further sensors, such as an inertial or vibration sensor that may detect a rotational imbalance of the wind turbine rotor. Based on the blade moment measurements and / or by the measurements obtained from the one or more ( further) sensors (in particular sensors different from the sensors from which the blade moment measurements are obtained, e. g. different from any such sensor), it may be determined whether the detected moment imbalance condition is caused by the presence of a faulty blade moment measurement. This may for example allow a distinction of a fault condition of the wind turbine operation (e. g. pitch offset, structural blade damage, etc. ) from a faulty moment measurement.

[0026] The detection that a moment imbalance condition exists may indicate that a fault condition exists, such as a moment 2024PF00540 Subsequent Filing

[0027] 5

[0028] measurement fault or fault condition of wind turbine operation. The existence of a moment imbalance condition may mean that the (overall) moment imbalance measure approaches or exceeds a certain value, e. g. exceeds a threshold, for example that a moment imbalance of a certain size is present. The method may further comprise determining the type of the fault condition, e. g. whether it is a measurement fault or a fault condition of the wind turbine operation.

[0029] For example, determining if the blade moment imbalance condition is caused by the presence of a faulty blade moment measurement based on the blade moment measurements may comprise statistically processing the measurements of blade moment of the respective rotor blade. Preferably, this may comprise determining a moment measure in form of an edgewise moment standard deviation and employing the edgewise moment standard deviation to determine if the blade moment imbalance condition is caused by the presence of a faulty blade moment measurement. The change in the standard deviation in case of a faulty blade moment measurement may differ significantly from a change in the standard deviation resulting from an actual physical imbalance. Other properties of the blade moment measurements, such as mean values and the like, may likewise provide a suitable distinction between imbalances caused by measurement fault and imbalances caused by a fault condition of the wind turbine operation.

[0030] In some optional embodiments, the method may be employed for detecting at least or solely a faulty blade moment measurement.

[0031] In an embodiment, detecting that the blade moment measurement is faulty may for example comprise detecting a sensor fault, a calibration fault, and / or the like. The detection step may form part of the method, and the method may optionally comprise taking one or more mitigation actions if a faulty measurement is detected. 2024PF00540 Subsequent Filing

[0032] In an embodiment, the moment measure of a blade is determined by statistically processing the measurements of blade moment of the respective rotor blade. For example, to determine a value of the moment measure, measurements of the blade moment obtained during a predetermined period of time, for example during at least one revolution of the wind turbine rotor, may be processed. Statistical processing may for example include averaging and / or determination of a standard deviation.

[0033] The blade moments may be bending moments. The blade moments may be measured by measuring a loading or strain of the blade using a respective sensor and processing the sensor data to derive the respective blade moment. A respective strain gauge may be employed. For example, a sensor may comprise an optical fiber running along the blade, preferably a temperature-compensated optical fiber, by means of which strain may be measured. Respective sensors and blade moment measurements are known in the art, and any of these may be employed here.

[0034] The moment measure may comprise at least one, preferably two or three of the following types of moment measures: a mean flapwise moment; a mean edgewise; and an edgewise moment standard deviation. Other moment measures can likewise be used, such as a flapwise moment standard deviation, a mean in-plane moment, a mean out-of-plane moment (both determined with respect to the rotor plane), respective standard deviations, and the like. The flapwise moment and the edgewise moment may be defined with respect to a zero pitch reference line of the blade. The method may be implemented with any or with any combination of the respective moment measures. The type of moment measures used may only need to be considered when setting a respective threshold, which is explained in more detail further below.

[0035] The moment imbalance measure may be determined for each of the employed types of moment measures. By monitoring more 2024PF00540 Subsequent Filing

[0036] than one moment imbalance measure, the reliability of detecting a blade moment imbalance condition may be improved.

[0037] Determining a moment imbalance measure may comprise determining a blade moment imbalance measure for each blade. The blade for which a blade moment imbalance condition exists, e. g. for which the moment measurement suffers from a fault, may thus be identified reliably.

[0038] The blade moment imbalance measure may be determined for a specific rotor blade by averaging the moment measure of the other two rotor blades and comparing the moment measure of the specific rotor blade to the averaged moment measure. The comparing may be performed by a subtraction. A reliable measure of the imbalance may thus be determined in a fast and efficient manner. In particular, the blade moment imbalance measure for each blade may be determined in accordance with equation 1 indicated further below.

[0039] Determining a moment imbalance measure may comprise determining an overall imbalance measure from the moment measure of each rotor blade. Employing such overall imbalance measure may make the detection of a blade moment imbalance condition more efficient.

[0040] For example, the overall imbalance measure may be determined from each of the blade moment imbalance measures of the blades. Such implementation may facilitate the fault detection while making available imbalance information for each rotor blade. For example, the overall imbalance measure may be determined by equation 2 indicated below, or may be determined by averaging the absolute values of the blade moment imbalance measures of the at least three rotor blades. In other implementations, the overall imbalance measure may be determined differently, e. g. directly from the moment measures ( for example by substituting the equations for the blade moment imbalance measures into the equation for the overall imbalance measure further below). 2024PF00540 Subsequent Filing

[0041] 8

[0042] Such blade moment imbalance measure and / or such overall imbalance measure may be determined for each type of moment measure that is employed by the method, e. g. for the mean flapwise moment and / or the mean edgewise moment and / or the edgewise moment standard deviation. This may improve the reliability of the detection, since a measurement fault or operation problem may show predominantly in one of these measures while not being detectable in another measure. The blade moment imbalance measure and the overall imbalance measure may be determined in essentially the same way for each type of moment measure.

[0043] In an embodiment, employing the moment imbalance measure to detect if blade moment imbalance condition exists may comprise comparing the moment imbalance measure to a threshold and detecting that the blade moment imbalance condition exists for at least one of the rotor blades if the moment imbalance measure exceeds the threshold. Preferably, the overall imbalance measure is compared to a respective threshold. In other implementations, one or more, e. g. each, of the blade moment imbalance measures may be compared to a respective threshold. Detecting a faulty measurement may thus include detecting if MimbA> threshold1; MimbB> threshold1, Mlmbc> thresholds, wherein MimbA, MlmbB, Mlmbcis the blade moment imbalance measure for three rotor blades A, B and C, and thresholdl, threshold!, and thresholds are respective thresholds, which may be the same or may be different.

[0044] Further, the comparison to a threshold may be performed for each type of moment measure. Accordingly, a dedicated threshold may be provided for each type of moment measure and the respective overall imbalance measure may be compared to the corresponding dedicated threshold. In other words, a first threshold for the mean flapwise moment, a second threshold for the mean edgewise moment, and / or a third threshold for the edgewise moment standard deviation may be 2024PF00540 Subsequent Filing

[0045] 9

[0046] provided for comparison to the respective overall imbalance measure.

[0047] If the moment imbalance, in particular the overall moment imbalance, exceeds the threshold, the method may further comprise comparing the blade moment imbalance measures determined for the rotor blades to each other or to respective thresholds. Thereby, the rotor blade for which the blade moment imbalance condition exists, e. g. for which the measurement of the blade moment is faulty, may be detected. For example, the blade having the largest magnitude of the blade moment imbalance measure, in particular the largest absolute value, may be determined to be the blade having the faulty moment measurement. Other possibilities for identifying the blade for which the measurement is faulty or for which another fault condition is present exist. For example, if several types of moment measure are monitored, the comparison may be made for each type, which may facilitate identification of the blade that suffers from the imbalance condition, e. g. from the faulty moment measurement.

[0048] The method may comprise measuring an operating parameter indicative of a wind turbine operating condition when measuring the blade moment of the rotor blades, and the threshold may be dependent on the operating parameter. For example, the operating parameter may be indicative of wind speed to which the wind turbine is exposed. Accordingly, when the comparison is made, the moment imbalance measure may be compared to the threshold that corresponds to the value of the operating parameter, e. g. wind speed, that was measured when the blade moments were measured from which the moment imbalance measure was determined. Making the threshold dependent on the operating parameter may on the one hand increase the sensitivity of the detection and on the other hand reduce false detections. The threshold may in particular be set closer to the values observed during normal operation of the blade moment measurements. The operating parameter may be wind speed, wind turbine rotor speed, generator speed, 2024PF00540 Subsequent Filing

[0049] 10

[0050] wind turbine output power, or any other parameter indicative of the current operating condition of the wind turbine.

[0051] If the blade moments and / or moment measures are averaged, the operating parameter may correspondingly be averaged over the same or a corresponding period of time. This may ensure that correspondence between the moment imbalance measure and the wind turbine operating condition for which the threshold is retrieved is conserved.

[0052] The threshold may be determined in different ways. For example, the blade moments may be modeled and respective moment imbalance measures may be determined for different wind speeds, and the threshold may be set above the modeled moment imbalance measures (which correspond to faultless blade moment measurements). A certain margin may be observed between the modeled measures and the threshold to avoid accidental triggering of the detection of the imbalance condition, e. g. of the fault detection.

[0053] The threshold may be stored in a lookup table, for example for different wind speed bins. It may also be provided as parameters of a mathematical function that returns a threshold value for a given wind speed ( for the respective type of moment measure). A combination of linear functions, a polynomial function or the like may be used.

[0054] Determining the moment imbalance measure may comprise averaging the moment measure of each blade for at least two revolutions of the wind turbine rotor and determining the moment imbalance measure for the blade from the averaged moment measures of each blade. The occurrence of outliers and incorrect detection of a fault may be reduced by averaging over more revolutions.

[0055] In an embodiment, the method comprises determining a first moment imbalance measure from an averaged moment measure of each blade that is averaged over a first number of 2024PF00540 Subsequent Filing

[0056] 11

[0057] revolutions of the wind turbine rotor, and determining a second moment imbalance measure from an averaged moment measure of each blade that is averaged over a second number of revolutions of the wind turbine rotor. The second number may be larger than the first number. Preferably, detecting if a moment imbalance condition exists for at least one of the rotor blades may comprise comparing the first moment imbalance measure to a first threshold and comparing the second moment imbalance measure to a second (different) threshold. The second threshold may for example be lower than the first threshold. Such method may allow increasing the speed with which larger moment imbalances are detected while at the same time ensuring that smaller moment imbalances are likewise reliably detected. Averaging over only few revolutions allows a faster detection. A higher threshold may be used to prevent false detections, thus allowing a fast detection of a larger error. Averaging over a larger number of revolutions reduces the deviations (outliers). By comparing such longer average to a lower threshold, smaller imbalances may be detected reliably while at the same time avoiding false detection of a measurement fault. The first number may for example be selected from the range of 1 to 5, and the second number may be selected from the range of 4 to 20.

[0058] In an embodiment, detecting if a blade moment imbalance condition exists comprises detecting if the measurement of the blade moment is faulty for at least one of the rotor blades. The method further comprises, if it is detected that the measurement of blade moment is faulty for one of the rotor blades, substituting the measurement of the blade moment for this rotor blade by the measurement of the blade moment for a rotationally preceding rotor blade. The substitute measurements for the rotor blade may be delayed by one third of the rotation period of the wind turbine rotor. Such delay may account for the 120° azimuth difference between the rotor blade and the preceding rotor blade. A preceding rotor blade may be a rotor blade that is offset by 2024PF00540 Subsequent Filing

[0059] 12

[0060] 120° in the direction of rotation of the wind turbine rotor. Such substitution of the measurements of the blade moment, which are faulty, by the measurements of blade moment for another blade may be termed "moment replication mode" herein. It may allow continued operation with relatively reliable blade moment data, since the blade moments experienced by the different rotor blades are generally rather similar.

[0061] Furthermore, performance and safety of the operation may be improved, as it may be avoided that the wind turbine operates with faulty blade moment measurements.

[0062] It is also conceivable to change the operation of the wind turbine upon detecting a faulty blade moment measurement. The wind turbine may change the control behavior, reduce the turbine performance (e. g. curtailed operation), and the like; it may for example enter into a safe operating mode.

[0063] If it is detected that the measurement of the blade moment is faulty for one of the rotor blades (i. e., for a further rotor blade) when operating with such substitute measurements for a rotor blade (i. e. when operating in the moment replication mode), the method may further comprise changing an operation mode of the wind turbine to a safe mode or shut down of the wind turbine. In such a situation, it may not be possible to reliably determine the blade with the sensor error and no further replication of measurements may be possible. Entering such safe mode operation may thus ensure the operational safety of the wind turbine. Safe mode operation may include a reduced power generation, e. g. output power curtailment, choosing a less aggressive operating mode, pitching out rotor blades, and the like. A service may then need to be performed to re-calibrate and / or repair the respective monitoring system to restore reliable blade moment measurements.

[0064] The blade moment measurements may be performed by a monitoring system comprising at least one load sensor on each rotor blade. As mentioned above, such load sensor may include a strain gauge, in particular optical fibers or the like. 2024PF00540 Subsequent Filing

[0065] 13

[0066] According to a further aspect of the invention, a monitoring system configured to monitor a blade moment of a rotor blade of a wind turbine rotor is provided. The wind turbine rotor comprises at least three rotor blades, and the monitoring system is configured to perform any of the methods disclosed herein. The monitoring system may for example comprise a processing unit and a memory, and the memory may store control instructions which, when executed by the processing unit, cause the monitoring system to perform any of the disclosed methods. Such monitoring system may comprise at least one load sensor on each blade to measure a quantity, such as strain, from which the blade moment is derived.

[0067] According to a further aspect, a wind turbine comprising such monitoring system is provided. The monitoring system may be provided as part of a control system of the wind turbine, or may be coupled to such control system of the wind turbine.

[0068] According to a further aspect of the invention, a computer program for monitoring a blade moment of a rotor blade of a wind turbine rotor by a monitoring system is provided. The wind turbine rotor comprises at least three rotor blades. The computer program comprises control instructions which, when executed by a processing unit of the monitoring system, cause the processing unit to perform any of the methods disclosed herein. The computer program may be provided on a volatile or non-volatile data carrier or storage medium and / or may be provided via a communication connection, such as a wired or wireless network connection.

[0069] By such monitoring system, wind turbine or computer program, advantages similar to those outlined further above with respect to the method may be achieved.

[0070] It is to be understood that the features mentioned above and those yet to be explained below can be used not only in the respective combinations indicated, but also in other 2024PF00540 Subsequent Filing

[0071] 14

[0072] combinations or in isolation, without leaving the scope of the present invention. In particular, the features of the different aspects and examples of the invention can be combined with each other unless noted to the contrary.

[0073] BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The forgoing and other features and advantages of the invention will become further apparent from the following detailed description read in conjunction with the accompanying drawings. In the drawings, like reference numerals refer to like elements.

[0075] Fig. 1 is a schematic drawing showing a wind turbine including a monitoring system according to an embodiment.

[0076] Fig. 2 is a flow diagram illustrating a method of monitoring a blade moment of a rotor blade according to an embodiment.

[0077] Fig. 3 is a schematic diagram illustrating an overall imbalance measure for mean flapwise moment and a respective threshold according to an embodiment.

[0078] Fig. 4 is a schematic diagram illustrating an overall imbalance measure for mean edgewise moment and a respective threshold according to an embodiment.

[0079] Fig. 5 is a schematic diagram illustrating an overall imbalance measure for edgewise moment standard deviation and a respective threshold according to an embodiment.

[0080] DETAILED DESCRIPTION

[0081] In the following, embodiments and / or examples of the invention will be described in detail with reference to the accompanying drawings. It is to be understood that the 2024PF00540 Subsequent Filing

[0082] 15

[0083] following description of the embodiments is given only for the purpose of illustration and is not to be taken in a limiting sense. It should be noted that the drawings are to be regarded as being schematic representations only, and elements in the drawings are not necessarily to scale with each other. Rather, the representation of the various elements is chosen such that their function and general purpose become apparent to a person skilled in the art. As used herein, the singular forms "a, " "an, " and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprising, " "having, " "including, " and "containing" are to be construed as open-ended terms (i. e., meaning "including, but not limited to, ") unless otherwise noted.

[0084] Herein below, embodiments are described in which detecting if a blade moment imbalance condition exists comprises detecting if the measurement of the blade moment is faulty for at least one of the rotor blades. It should be clear that the disclosed teachings can likewise be applied for detecting if a fault condition of wind turbine operation exists, such as a pitch offset of a rotor blade or a problem associated with the physical structure of the rotor blade. Distinction between such types of faults or fault conditions may be made based on information obtained from the wind turbine control or from other sensors. Generally, each fault condition of the wind turbine operation may affect the wind turbine operation differently, which may be detected. For example, a structural fault or pitch offset of a rotor blade may, besides a moment imbalance, cause a nacelle acceleration or vibration, which may be detected and may distinguish such fault condition from a fault of a blade moment measurement. An optional mitigation action may be chosen in dependence on the type of fault.

[0085] Fig. 1 schematically illustrates a wind turbine 100 comprising a wind turbine tower 101 and a wind turbine rotor 120. The rotor 120 has three rotor blades 121, 122 and 123. Wind turbine 100 may be an onshore, an offshore, or a 2024PF00540 Subsequent Filing

[0086] 16

[0087] floating wind turbine. Such wind turbines are generally known in the art, and wind turbine 100 may have any of the known configurations.

[0088] To measure blade bending moments, a monitoring system 10 according to an embodiment is provided. Monitoring system 10 comprises load sensors 21, 22 and 23, each being provided at a respective blade 121, 122 and 123. The load sensors 21 to 23 may each comprise one or more optical fibers that measure strain in the blade by changes to their optical characteristics caused by the strain. Any other types of respective load sensors may additionally or alternatively be employed. Measurement of a blade moment may include measuring a respective sensor signal by the corresponding load sensor and processing the sensor signal to derive the blade moment. Such measurement of blade moment may be based on a respective calibration. Monitoring system 10 may store such calibration data.

[0089] Monitoring system 10 may include a processing unit 11 coupled to a memory 12. Processing unit 11 may be a micro-processor, a digital signal processor, an application-specific integrated circuit, or the like. Memory 12 may comprise RAM, ROM, Flash-Memory, a hard disk drive and other types of memory. Memory 12 may store control instructions which, when executed by processing unit 11, cause the monitoring system 10 to perform any of the methods disclosed herein. Monitoring system 10 may be implemented in one or more controllers and may accordingly comprise plural processing units and plural memories. Monitoring system 10 may comprise further components common to such monitoring system, for example input / output interfaces, a system bus, a user interface and the like.

[0090] Monitoring system 10 may provide as an output measurements of different blade moments for each blade, such as flapwise moment, edgewise moment, in-plane moment, out-of-plane moment, and the like. Monitoring system 10 may form part of a 2024PF00540 Subsequent Filing

[0091] 17

[0092] control system 15 of the wind turbine, or may be coupled to such control system. The control system 15 may employ the output of the monitoring system 10 to control the wind turbine, for example to reduce the loads acting on rotor 120 if excessive moment imbalances are detected.

[0093] The monitoring system 10 is configured to detect a moment imbalance condition, e. g. a faulty measurement of a moment, which may for example be due to a calibration error, a sensor error, or the like. An embodiment of a method for detecting such faulty blade moment measurement that may be implemented on the monitoring system 10 is illustrated in the flow diagram of Fig. 2. In step S10, measurements of the blade moment are obtained for each rotor blade, in particular by means of the sensors 21 to 23. Further, measurements of an operating parameter, such as wind speed, are obtained simultaneously, for example via a dedicated wind speed sensor of wind turbine 100 or from a nearby wind speed sensor.

[0094] In step S11, a moment measure is determined for each rotor blade from the blade moment measurements obtained for the respective blade. The moment measure may be determined from the blade moment measurements obtained over at least one revolution of the wind turbine rotor. The moment measure can be determined by a statistical processing of the blade moment measurements, such as by calculating an average or standard deviation of the measurement values. Two or more moment measures may be derived for each blade. It is beneficial if the determined moment measures for each blade include at least a mean flapwise moment, a mean edgewise moment and an edgewise moment standard deviation. The mean edgewise moment and the edgewise moment standard deviation may be monitored to provide a more reliable detection of edge sensor failures. Since the mean value of the edgewise moment may be close to zero, and a sensor failure to zero may not be detectable by only a comparison of the mean value, the standard deviation is employed as an additional moment measure. 2024PF00540 Subsequent Filing

[0095] 18

[0096] Optionally, in step S12, the moment measure of each blade can be averaged over a number of revolutions of the wind turbine rotor. As such averaging may reduce outliers, it may avoid false detection of a moment imbalance condition. The greater the number of rotor revolutions used in the averaging, the lower is the expected imbalance during normal operation.

[0097] However, for a fast detection, only few revolutions should be averaged, or no averaging may be used. This may allow a faster detection of larger errors. Both may be employed simultaneously, i. e. an average of a larger number of revolutions may be monitored simultaneously with an average over a low number of revolutions, wherein different thresholds may be set, as described further below.

[0098] In step S13, a blade moment imbalance measure is determined for each rotor blade by comparing the moment measure for the respective rotor blade to an average moment measure of the remaining two rotor blades. This may be performed for each type of moment measure, in particular for the mean flapwise moment and the mean edgewise moment, as well as for the edgewise moment standard deviation. For a moment measure M of the blades A, B, C, the blade moment imbalance measure Mimbmay be determined as follows for each blade:

[0099] WjmbA — (MB+ Me) / 2

[0100] ^imbB (Eq. 1 )

[0101]

[0102] ^imbC ( M J T- M MS3J / / 7

[0103] Accordingly, the moment measure or each blade is compared to the moment measures of the other blades, which serve as a reference. By providing a comparison between blades, no expected moment values are needed, as in the prior art, and the comparison can be made regardless of the operating conditions or environmental conditions. The comparison is still valid even when the wind turbine operates with curtailment, noise-reduced operation, wake adaptation modes, different air densities, or the like. 2024PF00540 Subsequent Filing

[0104] 19

[0105] To facilitate the detection of a moment imbalance condition, e. g. a fault in the measurement, an overall magnitude of the imbalance may be determined from the three blade imbalance measures. The following equation may be used to calculate the overall magnitude of the moment imbalance as an overall imbalance measure Mlmb0:

[0106] I

[0107] ;, 1

[0108] Mjmbo = 'WimbA- +

[0109]

[0110] \ (Eq. 2 )

[0111] Such overall imbalance measure, in particular overall moment imbalance magnitude, may be used because it allows the capturing of an overall rotor imbalance, and not just of an individual blade moment imbalance. Such overall imbalance measure may be determined for each of the mean flapwise moment, the mean edgewise moment, and the edgewise moment standard deviation. It should be clear that this is only one example of calculating such overall imbalance measure, and other ways of determining this overall imbalance measure may likewise be used, such as averaging the absolute values of the blade imbalance measures. It should also be clear that averages of the individual blade moments described for use in Eq. 1 and Eq. 2 to determine the imbalances could be replaced by averages of Eq. 1 and Eq 2 using non-averaged blade moments to determine the imbalances.

[0112] Turning back to Fig. 2, the overall imbalance measure is determined in step S14. In other implementations, only the blade imbalance measures of step S13 may be employed, or the overall imbalance measure of step S14 may be determined without the intermediate step S13.

[0113] In the diagram of Fig. 3, the overall imbalance measure based on the mean flapwise moment MimbF is shown. The data illustrated in Fig. 3 is based on a simulation for different wind speeds. In this simulation, the overall imbalance 2024PF00540 Subsequent Filing

[0114] 20

[0115] measure of the mean flapwise moment has been determined by averaging over five rotor revolutions, wherein the data points correspond to a maximum value of this overall imbalance measure determined within a ten-minute simulation interval for the particular wind speed. The data points 31 correspond to the overall imbalance measure for a first design load case (DLC 1.2 ) and the data points 32 correspond to the overall imbalance measure for a second design load case (DLC 1.3). A threshold 33 can be set in dependence on wind speed for the overall imbalance measure. The threshold 33 can be determined such that it lies above the maximum values of the overall imbalance measure that are experienced during this simulation, e. g. by a predetermined margin. False detection of a moment imbalance condition may thus be reduced. During operation, the determined overall imbalance measure for mean flapwise moment MimbF may then be compared to the threshold 33 to detect the presence of a moment imbalance condition, such as a measurement fault or fault condition of the wind turbine operation. In the example of Fig. 3, threshold 33 is composed of two linear segments, which may be described by linear function parameters. In other implementations, different mathematical functions may be used as threshold, or the threshold 33 may be stored in the form of discrete values in a lookup table for a certain set of wind speed bins.

[0116] Fig. 4 shows respective simulated data for the overall imbalance measure based on mean edgewise moment MimbE. Again, data points 41 correspond to the design load case DLC 1.2 and data points 42 to the design load case DLC 1.3. The mean edgewise moment overall imbalance values MimbEare significantly lower than for Fig. 3. A second threshold 43 can be set for the overall imbalance measure based on mean edgewise moment. Again, the simulation data has been averaged over five revolutions.

[0117] Fig. 5 shows a third example of respective simulated data points, wherein the overall imbalance measure based on 2024PF00540 Subsequent Filing

[0118] 21

[0119] edgewise moment standard deviation MimbSTDis illustrated. Data points 51 correspond to the design load case DLC 1.2 and data points 52 correspond to the design load case DLC 1.3. A third threshold 53 for the edgewise moment standard deviation overall imbalance measure can be set based on the simulated data.

[0120] Again, in Figs. 4 and 5, the thresholds 43, 53 are composed of two linear segments, but the thresholds may have any other shape that provides a respective dependency on wind speed. By making the threshold dependent on wind speed, the detection of the moment imbalance condition, e. g. of a measurement fault, may be made more reliable, as the threshold can be set closer to the overall imbalance measure expected during normal operation. At the same time, as the threshold increases for larger wind speeds, the risk for a false detection of a measurement fault can be reduced. Operating parameters other than wind speed may be employed for this purpose, such as wind turbine rotor speed, generator speed, wind turbine output power, or any other parameter indicative of the current operating condition of the wind turbine.

[0121] Besides using simulation data for determining the respective thresholds for the overall imbalance measure, other methods may be used for setting the threshold. For example, respective data may be collected during operation of the wind turbine and may be used to set the threshold. Further, it would likewise be possible to employ respective thresholds for each blade imbalance measure additionally or alternatively to using such threshold for the overall imbalance measure.

[0122] Turning back to Fig. 2, the overall imbalance measure is compared to the respective threshold in step S15. This may be performed for each type of overall imbalance measure that is employed in the method, e. g. for the mean flapwise moment, the mean edgewise moment, and / or the edgewise moment standard deviation. If it is found in step S15 that the overall imbalance measure is below the threshold, no faulty 2024PF00540 Subsequent Filing

[0123] 22

[0124] measurement is detected and the method continues in step S10. In case that the threshold is exceeded, it is determined that the moment measurement is faulty for at least one blade.

[0125] Optionally, the method may check in step S16 if a replication mode (see below) is active. If this is not the case, the method continues in step S17, wherein the blade having the faulty blade moment measurement is determined from the blade imbalance measures. These may be compared with each other, and the blade imbalance measure that has the largest magnitude may be determined as corresponding to the blade for which the moment measurement is faulty. Other ways of determining for which blade the measurement is faulty based on the blade imbalance measures may also be employed.

[0126] This faulty measurement can no longer be used. In step S18, the faulty blade moment measurement is therefore substituted by a blade moment measurement of a preceding rotor blade. Since the preceding rotor blade is offset by 120° for a three blade rotor, the moment measurement of the preceding blade will be ahead by one third of a revolution. The substitute moment measurement may thus be delayed by one third of the rotation period of the wind turbine rotor. By the substitution of step S18, it may be ensured that the monitoring system 10 can still deliver rather reliable measurements of the bending moments of all three blades although the measurement for one blade is faulty. Wind turbine control and the blade moment monitoring system can thus continue to operate. Optionally in step S18, the control behavior of the wind turbine may be changed and / or the wind turbine performance may be reduced to ensure safe operation. Operation with such substitute blade moment measurement may be termed "replication mode". Operation may then continue in step S10.

[0127] If in step S16 it is determined that the replication mode is already active, and the overall imbalance measure has been found to exceed the threshold in step S15, it may not be possible to provide further replication or to determine the 2024PF00540 Subsequent Filing

[0128] 23

[0129] blade that suffers from the measurement fault. In this case, the method continues in step S19 and operates the wind turbine in a safe mode. Operation in such safe mode may include a power reduction, or may even include operating the wind turbine in an idle mode or shutdown of the wind turbine.

[0130] Steps S16 to S19 are optional. Further, it should be clear that the order of the steps may be changed.

[0131] The method may allow a fast and efficient detection of an error in the sensor calibration or a sensor fault. In particular, by using several imbalance measures, the reliability of the detection may be improved. Since for detecting the imbalance, the moment measure does not need to be compared to any fixed, theoretical, or otherwise determined reference, but is compared to respective values obtained from the other two rotor blades, the method can be employed in different operating modes, different operating conditions and thus benefits from an improved versatility. Nontypical conditions, such as a wake adaptation mode, derated operation mode and the like, do not impede operation of the method. The method may correspondingly be employed to detect other moment imbalance conditions, e. g. caused by a fault condition of wind turbine operation, such as pitch offset or the like.

[0132] While specific embodiments are disclosed herein, various changes and modifications can be made without departing from the scope of the invention. The present embodiments are to be considered in all respects as illustrative and non-restrictive, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.

Claims

2024PF00540 Subsequent Filing24Patent claims1. A method of monitoring a blade moment of a rotor blade of a wind turbine rotor for detecting a faulty blade moment measurement, wherein the wind turbine rotor ( 120) comprises at least three rotor blades ( 121, 122, 123), wherein the method comprises:obtaining measurements of a blade moment of each of the rotor blades ( 121, 122, 123);determining a moment measure (MA, MB, Mc) for each of the rotor blades ( 121, 122, 123) from the obtained blade moment measurements;determining a moment imbalance measure (MimbA, MimbB, MimbC; MimbO) for at least one of the rotor blades; wherein determining the moment imbalance measure comprises comparing the moment measure (MA) determined for one rotor blade ( 121 ) to the moment measure (MB, Mc) determined for the other rotor blades ( 122, 123); employing the moment imbalance measure (MimbA, MimbB, MimbC; MimbO) to detect if a blade moment imbalance condition exists for at least one of the rotor blades ( 121, 122, 123), which comprises detecting if the measurement of the blade moment is faulty for at least one of the rotor blades, wherein whether the blade moment imbalance condition is caused by the presence of a faulty blade moment measurement is determined based on the blade moment measurements and / or by obtaining measurements of one or more sensors configured to detect a rotational imbalance of the wind turbine rotor.

2. The method according to claim 1, wherein the moment measure (MA, MB, Mc) of a blade is determined by statistically processing the measurements of blade moment of the respective rotor blade.

3. The method according to claim 1 or 2, wherein the moment measure (MA, MB, Mc) comprises at least one, preferably two or three of the following types of moment measures:2024PF00540 Subsequent Filing25a mean flapwise moment;a mean edgewise moment; andan edgewise moment standard deviation.

4. The method according to any of the preceding claims, wherein determining whether the blade moment imbalance condition is caused by the presence of a faulty blade moment measurement based on the blade moment measurements comprises statistically processing the measurements of blade moment of the respective rotor blade.

5. The method according to claim 4, wherein the statistical processing of the measurements of blade moment comprises determining a moment measure in form of an edgewise moment standard deviation and employing the edgewise moment standard deviation to determine if the blade moment imbalance condition is caused by the presence of a faulty blade moment measurement.

6. The method according to any of the preceding claims, wherein determining a moment imbalance measure comprises determining a blade moment imbalance measure (MimbA, MimbB, Mimbc) for each blade ( 121, 122, 123), wherein the blade moment imbalance measure is determined for a specific rotor blade ( 121 ) by averaging the moment measure of the other two rotor blades ( 122, 123) and comparing the moment measure of the specific rotor blade ( 121 ) to the averaged moment measure, preferably by subtraction.

7. The method according to any of the preceding claims, wherein determining a moment imbalance measure comprises determining an overall imbalance measure (Mimbo) from the moment measure of each rotor blade ( 121, 122, 123).

8. The method according to claim 7, wherein determining a moment imbalance measure comprises determining a blade moment imbalance measure (MlmbA, MlmbB, Mlmbc) for each blade ( 121, 122, 123), wherein the overall imbalance measure (Mimbo) is2024PF00540 Subsequent Filing26determined from each of the blade moment imbalance measures (MimbA, MimbB, Mimbc) of the blades ( 121, 122, 123).

9. The method according to any of the preceding claims, wherein employing the moment imbalance measure to detect a blade moment imbalance condition comprises comparing the moment imbalance measure to a threshold (33, 43, 53) and detecting that the blade moment imbalance condition exits for at least one of the rotor blades ( 121, 122, 123) if the moment imbalance measure exceeds the threshold (33, 43, 53).

10. The method according to claim 9, wherein the method comprises obtaining measurements of an operating parameter indicative of a wind turbine operating condition when measuring the blade moment of the rotor blades ( 121, 122, 123), wherein the threshold (33, 43, 53) is dependent on the operating parameter.

11. The method according to any of the preceding claims, wherein the method comprises determining a first moment imbalance measure from an averaged moment measure of each blade that is averaged over a first number of revolutions of the wind turbine rotor ( 120), and determining a second moment imbalance measure from an averaged moment measure of each blade that is averaged over a second number of revolutions of the wind turbine rotor ( 120), the second number being larger than the first number, wherein preferably, detecting if an imbalance condition exists for at least one of the rotor blades comprises comparing the first moment imbalance measure to a first threshold and comparing the second moment imbalance measure to a second threshold.

12. The method according to any of the preceding claims, wherein the method further comprises, if it is detected that the measurement of the blade moment is faulty for one of the rotor blades ( 121, 122, 123), substituting the measurements of the blade moment for said rotor blade by the measurements of the blade moment for a rotationally preceding rotor blade,2024PF00540 Subsequent Filing27wherein preferably, the substitute measurements for said rotor blade are delayed by 1 / 3 of the rotation period of the wind turbine rotor ( 120).

13. The method according to claim 12, wherein, if it is detected that the measurement of the blade moment is faulty for one of the rotor blades ( 121, 122, 123) when operating with substitute measurements for a rotor blade, the method further comprises changing an operation mode of the wind turbine ( 100) to a safe mode.

14. A monitoring system configured to monitor a blade moment of a rotor blade of a wind turbine rotor, wherein the wind turbine rotor ( 120) comprises at least three rotor blades ( 121, 122, 123), wherein the monitoring system ( 10) is configured to perform the method according to any of the preceding claims.

15. A computer program for monitoring a blade moment of a rotor blade of a wind turbine rotor by a monitoring system, wherein the wind turbine rotor ( 120) comprises at least three rotor blades ( 121, 122, 123), wherein the computer program comprises control instructions which, when executed by a processing unit ( 11 ) of the monitoring system ( 10), cause the processing unit ( 11 ) to perform the method according to any of claims 1-13.

Images