Detecting blade anomaly of wind turbine

Abstract

A method of monitoring one or more blades of a rotor of a wind turbine, the method comprising: obtaining bending information indicating bending of one or more of the blades; obtaining pitch information indicating a pitch angle of one or more of the blades; when wind speed is below a rated wind speed: operating the wind turbine in a partial load mode in which pitch angles of the blades are controlled to maximise power captured by the rotor; operating a first detection mode in which the bending information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the first detection mode coincides with the partial load mode over a first range of wind speeds; when wind speed is above the rated wind speed: operating the wind turbine in a full load mode in which the pitch angles of the blades are varied to capture a maximum allowed power with the rotor; and operating a second detection mode in which the pitch information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the second detection mode coincides with the full load mode over a second range of wind speeds.

Description

[0001] DETECTING BLADE ANOMALY OF WIND TURBINE

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to a method of monitoring a blade of a rotor of a wind turbine; a method of controlling a wind turbine; and apparatus for monitoring a blade of a rotor of a wind turbine.

[0004] BACKGROUND OF THE INVENTION

[0005] US20220186714 discloses a method for detecting an accretion of ice on a rotor blade of a rotor of a wind turbine. The wind turbine 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.

[0006] SUMMARY OF THE INVENTION

[0007] A first aspect of the invention provides a method of monitoring one or more blades of a rotor of a wind turbine, the method comprising: obtaining bending information indicating bending of one or more of the blades; obtaining pitch information indicating a pitch angle of one or more of the blades; when wind speed is below a rated wind speed: operating the wind turbine in a partial load mode in which pitch angles of the blades are controlled to maximise power captured by the rotor; operating a first detection mode in which the bending information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the first detection mode coincides with the partial load mode over a first range of wind speeds; when wind speed is above the rated wind speed: operating the wind turbine in a full load mode in which the pitch angles of the blades are varied to capture a maximum allowed power with the rotor; and operating a second detection mode in which the pitch information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the second detection mode coincides with the full load mode over a second range of wind speeds.

[0008] Optionally in the first detection mode a blade anomaly is determined to exist if the analysis of the bending information indicates bending of the blade(s) which is less than expected.

[0009] Optionally in the second detection mode a blade anomaly is determined to exist if the analysis of the pitch information indicates a pitch angle of the blade(s) which is less than expected.

[0010] Optionally in the first detection mode and / or the second detection mode the bending information and the pitch information are both analysed to determine whether a blade anomaly exists and generate a corresponding output.

[0011] Optionally in the second detection mode a blade anomaly is determined to exist if the analysis of the pitch information indicates a pitch angle of the blade(s) which is less than expected and an analysis of the bending information indicates a bending of the blade(s) which is more than expected.

[0012] Optionally the pitch information is obtained by determining an average or cumulative sum of pitch angles of one of the blades over a time period.

[0013] Optionally the pitch information is obtained by determining an average or sum of pitch angles of all of the blades of the rotor.

[0014] Optionally the bending information is obtained by determining a mean or cumulative sum of bending measurements of one of the blades over a time period.

[0015] Optionally the bending information is obtained by determining an average or sum of bending measurements of all of the blades of the rotor. Optionally the bending information is obtained by assigning a plurality of bending measurements to a bin based on wind speed, and combining the bending measurements to obtain the bending information; and the pitch information is obtained by assigning a plurality of pitch angle measurements to a bin based on wind speed, and combining the pitch angle measurements to obtain the pitch information.

[0016] Optionally in the first detection mode the bending information is analysed to determine whether a blade anomaly exists by comparing the bending information to a stored reference associated with the bin.

[0017] Optionally in the second detection mode the pitch information is analysed to determine whether a blade anomaly exists by comparing the pitch information to a stored reference associated with the bin.

[0018] Optionally the blade anomaly is a blade icing anomaly.

[0019] Optionally the first detection mode operates below a wind speed threshold which is the same as the rated wind speed, or offset from the rated wind speed by a predetermined amount.

[0020] Optionally the second detection mode operates above a wind speed threshold which is the same as the rated wind speed, or offset from the rated wind speed by a predetermined amount.

[0021] Optionally the method further comprises measuring wind speed to obtain wind speed measurements, and comparing the wind speed measurements with the rated wind speed to determine whether to operate in partial load mode or full load mode.

[0022] Optionally the method further comprises measuring wind speed to obtain wind speed measurements, and using the wind speed measurements to determine whether to operate in the first or second detection mode.

[0023] Optionally the first and second detection modes are each mixed detection modes in which the bending information and the pitch information are both analysed to determine whether a blade anomaly exists and generate the corresponding output; in the first detection mode the bending information is dominant; and in the second detection mode the pitch information is dominant.

[0024] A further aspect of the invention provides a method of controlling a wind turbine, the method comprising monitoring one or more blades of a rotor of the wind turbine by a method according to the preceding aspect; and controlling the wind turbine based on the output.

[0025] A further aspect of the invention provides apparatus for monitoring one or more blades of a rotor of a wind turbine, wherein the apparatus is configured to perform a method according to any preceding aspect.

[0026] BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Embodiments of the invention will now be described with reference to the accompanying drawings, in which:

[0028] Figure 1 is a front view of a wind turbine;

[0029] Figure 2 is a side view of the wind turbine;

[0030] Figure 3 shows a blade of the wind turbine

[0031] Figure 4 shows strain gauges in the root of the blade;

[0032] Figure 5 shows apparatus for monitoring the blades;

[0033] Figure 6 shows a control system of the wind turbine;

[0034] Figure 7 shows a control scheme for operating the wind turbine in partial load mode and full load mode;

[0035] Figure 8a is a graph showing variation of blade flap moment with wind speed;

[0036] Figure 8b is a graph showing variation of blade pitch angle with wind speed;

[0037] Figure 9 shows a method of monitoring the blades to flag ice events;

[0038] Figure 10 shows the different modes associated with Figure 9;

[0039] Figure 11 shows an alternative method using wind speed thresholds offset from the rated wind speed;

[0040] Figure 12 shows an alternative method of monitoring the blades to flag ice events; Figure 13 shows a pair of steep sigmoid functions; and

[0041] Figure 14 shows a pair of shallow sigmoid functions. DETAILED DESCRIPTION OF EMBODIMENT(S)

[0042] Figures 1 and 2 show a wind turbine 1 including a tower 3 mounted on a foundation and a nacelle 4 disposed at the apex of the tower. The wind turbine 1 depicted here is an onshore wind turbine such that the foundation is embedded in the ground, but the wind turbine 1 could be an offshore installation in which case the foundation would be provided by a suitable marine platform.

[0043] A rotor is operatively coupled via a gearbox to a generator housed inside the nacelle 4. The rotor includes a central hub and a plurality of rotor blades 2a-c, which project outwardly from the central hub. It will be noted that the wind turbine 1 is the common type of horizontal axis wind turbine (HAWT) such that the rotor is mounted at the nacelle to rotate about a substantially horizontal axis defined at the centre at the hub. The example shown has three blades, but it will be realised by the skilled person that other numbers of blades are possible.

[0044] When wind blows against the wind turbine 1, the blades generate a lift force which causes the rotor to rotate, which in turn causes the generator within the nacelle 4 to generate electrical energy.

[0045] All of the blades 2a-c are substantially identical, and an exemplary one of the blades 2a is shown in Figures 3 and 4. Each of the blades has a root end 6 proximal to the hub and a tip end 7 distal from the hub. A leading edge 5a and a trailing edge 5b extend between the root end 6 and the tip end 7, and each of the blades has a respective aerodynamic high pressure surface (i.e. the pressure surface) and an aerodynamic low pressure surface (i.e. the suction surface) extending between the leading and trailing edges of the blade.

[0046] Each blade has an edgewise direction 14 (between the leading edge 5a and the trailing edge 5b) and a flapwise direction 15 (at right angles to the edgewise direction).

[0047] Ice accretion on the blades can impact both performance and safety of the wind turbine. A clear indication of the presence of ice is important to determine reasons for performance losses, changes to control strategy, or legal requirements. From a safety standpoint, ice presence can lead to ice throw which can pose risk if a turbine is placed near people, roads, or animals. Improved ice detection can also lead to bill of materials (BOM) savings, reducing the need for further material, making the wind turbine 1 lighter, cheaper, and easier to transport.

[0048] As shown in Figure 4, the blade includes a blade shell 10 defining a hollow interior space 11. Strain gauges 12a, 13a are adhered to the interior blade shell, or embedded within the blade shell, at the root end 6 of the blade.

[0049] Each strain gauge 12a, 13a measures linear extension of the blade root. An example of a suitable strain gauge is an optical strain gauge such as a fibre Bragg grating (FBG) as described in EP3317531 B1. Such an optical strain gauge comprises equally spaced reflection points in the core of the optical fibre that reflect difference wavelengths of light under different levels of strain. Such a sensor is well known to the skilled person.

[0050] A first one of the strain gauges is an edge strain gauge 12a which is positioned to measure bending of the blade in the edgewise direction 14. When the blade bends one way in the edgewise direction 14, the edge strain gauge 12a extends and generates a positive edge bending signal. When the blade bends the other way in the edgewise direction 14, the edge strain gauge 12a contracts and generates a negative edge bending signal. In this case there is only a single edge strain gauge 12a, but in other embodiments there may be two edge strain gauges positioned on opposite sides of the blade root 6.

[0051] A second one of the strain gauges is a flap strain gauge 13a which is positioned to measure bending of the blade in the flapwise direction 15. When the blade bends one way in the flapwise direction 15, the flap strain gauge 13a extends and generates a positive flap bending signal. When the blade bends the other way in the flapwise direction 15, the flap strain gauge 13a contracts and generates a negative flap bending signal. In this case there is only a single flap strain gauge 13a, but in other embodiments there may be two flap strain gauges positioned on opposite sides of the blade root 6.

[0052] Figure 5 is a schematic diagram of apparatus 20 for monitoring the blades 2a-c of the rotor of the wind turbine 1. All elements of the apparatus 20 may be part of the wind turbine 1, or some elements may be located remotely from the wind turbine 1 (for instance at a central wind park controller).

[0053] Each blade has a pair of strain gauges 12a, 13a; 12b, 13b; 12c, 13c which generate bending signals which are input into a computer-implemented processing system 21.

[0054] The processing system 21 is coupled to a wind turbine control system 25 which is shown schematically in Figure 6, along with other elements of the wind turbine 1.

[0055] A wind speed sensor 26, which may be mounted on the nacelle or at any other location, generates wind speed measurements which are used by the processing system 21 and the wind turbine control system 25.

[0056] The rotor is mechanically connected to an electrical generator 50 via a gearbox 51 (in direct drive systems the gear box is not present). The electrical power generated by the generator 50 is injected into a power grid 54 via an electrical converter 55. The electrical generator 50 and the converter 55 may be based on a full scale converter (FSC) architecture or a doubly fed induction generator (DFIG) architecture, but other types may be used.

[0057] The control system 25 comprises a number of elements, including at least one main controller 200 with a processor and a memory, so that the processor is capable of executing computing tasks based on instructions stored in the memory. In general, the main controller 200 ensures that in operation the wind turbine generates a requested power output level. This is obtained by adjusting the pitch angle of the blades and / or the power extraction of the converter 55. To this end, the control system 25 comprises a pitch system including a pitch controller 57 using a pitch reference signal 58, and a power system including a power controller 59 using a power reference signal 56. The rotor blades 2a-c can be pitched by a pitch mechanism. The rotor comprises an individual pitch system which is capable of individual pitching of the rotor blades, and may comprise a common pitch system which adjusts all pitch angles on all rotor blades at the same time. The control system 25, or elements of the control system 25, may be connected to a power plant controller (not shown) or other control system to receive externally provided instructions. When wind speed measurements from the wind speed sensor 26 are below a rated wind speed, the control system 25 operates the wind turbine in a partial load mode in which pitch angles of the blades are controlled to maximise power captured by the rotor.

[0058] When wind speed measurements from the wind speed sensor 26 are above the rated wind speed, the control system 25 operates the wind turbine in a full load mode in which the pitch angles of the blades are varied to capture a maximum allowed power with the rotor, which is referred to below as a rated power.

[0059] Figure 7 gives an example of how the blade pitch, generator speed and produced power may vary as a function of wind speed in partial load mode and in full load mode. In the example shown in Figure 7, the partial load mode operates from a cut-in wind speed (here 5 m / s but it could be different) up to a rated wind speed (here 13 m / s). When operating in partial load mode, the turbine produces as much power as is possible with the available wind. Full load mode occurs from the rated wind speed up to a cut-out wind speed. In this full load mode the turbine produces rated power (e.g. 3 MW). The rated power captured with the rotor is typically substantially constant when operating the wind turbine in the full load mode.

[0060] The generator speed may be ramped up from cut-in wind speed until a maximum is reached at some wind speed (here at 9 m / s).

[0061] In partial load mode the optimum pitch is almost constant for wind speeds above 5 m / s and up to the rated wind speed. The pitch reference signal 58 shown in Figure 6 may be set to follow the optimum pitch during partial load mode operation.

[0062] Note that the wind speeds and rated power given in Figure 7 are examples only, and will vary from turbine to turbine.

[0063] US10364797 discloses a known example of operating in partial and full load as shown in Figure 7. Other suitable examples of operating in partial and full load mode are given in WO 2022 / 105975 and WO 2024 / 094263. Figure 8a is a graph showing how the blade flap moment may vary with wind speed for an iced turbine compared with a non-iced turbine. The blade flap moment for the noniced turbine increases to a maximum 60 in partial load mode and then decreases in full load mode. The two curves cross at a cross-over point 61.

[0064] Figure 8b is a graph showing how the blade pitch angle may vary with wind speed for an iced turbine compared with a non-iced turbine. The pitch angle in both cases is relatively constant in partial load mode, then increases when operating in full load mode.

[0065] When operating in partial load mode, the presence of ice reduces the airfoil efficiency of the blades. This results in lower blade moment at similar wind speeds (as shown in Figure 8a) with little effect on pitch angle (as shown in Figure 8b)

[0066] When operating in full load mode, a turbine with ice on the blades will have to be set at a lower-than-expected pitch angle (as shown in Figure 8b) to maintain the rated power at a given wind speed. The lower airfoil efficiency of an iced blade also results in a higher blade moment in full load mode (as shown in Figure 8a).

[0067] A method of monitoring one or more blades 2a-c of the rotor of the wind turbine is shown in Figure 9. In an input stage, blade moment data 70 is obtained indicating bending of one or more of the blades. In this example the blade moment data 70 comprises blade moment measurements from one or more blade sensors, such as strain gauges. For example the blade moment data 70 may indicate bending of one or more of the blades in a flapwise direction, for instance based on measurements by the flap strain gauge 13a.

[0068] Pitch angle data 71 is also obtained, indicating a pitch angle of one or more of the blades. This pitch angle data 71 may be obtained by one or more appropriate sensors, for example, or it may be based on the pitch reference signal 58.

[0069] The ambient temperature 72 of the environment is also obtained, as an indication of whether icing conditions are likely. Wind speed is also measured to obtain wind speed data 73 associated with the blade moment data 70 and the pitch angle data 71. The wind speed data 73 may be obtained by the wind speed sensor 26 shown in Figure 5.

[0070] The wind speed data 73 is binned to produce a wind speed measurement 74, and for each bin an average blade moment 75 is obtained along with an average pitch angle 76. The averages 75, 76 may each be obtained by determining a cumulative sum of measurements for a respective wind speed bin, for just one blade or for all blades of the rotor.

[0071] The temperature is checked at 77, and if the temperature is above a temperature threshold, for instance 3degc, then a training process 78 is performed in which the cumulative sum continues to be obtained for blade moment and pitch angle, and a model (such as a Kalman filter) is trained. During training of the model, the average bending moment is stored as a calibrated reference and updated when a higher value is measured. Thus each wind speed bin has a stored calibrated reference.

[0072] If the temperature is below the temperature threshold, then a detection process is performed. The first step of the detection process is to check a wind speed measurement at 80, for instance by comparing the wind speed measurement 74 with a wind speed threshold. The wind speed measurement 74 may be the average wind speed of a bin for example

[0073] When the check 80 indicates that the wind speed measurement 74 is below the wind speed threshold, then a first “blade load only” blade anomaly detection mode is operated. In this first detection mode, the average blade moment 75 is analysed at 81 to determine whether a blade anomaly exists and generate a corresponding output. A blade anomaly is determined to exist if the analysis of the average blade moment 75 indicates bending of the blade(s) which is less than expected. For example the analysis at 81 may check whether the average blade moment 75 is less than the calibrated reference for the current bin which was stored during the training process 78. The use of wind speed bins ensures that the effects of ice on the blade are compared under the same wind conditions. Referring to Figure 8a, a relatively low wind speed (below the rated wind speed) is indicated by a dashed vertical line 91. For this low wind speed 91 the blade moment 92 for an iced blade is lower than the expected blade moment 93 for a non-iced blade. For this wind speed 91 when the wind turbine is operating in partial load mode, the wind turbine is operating in the first (“blade load only”) detection mode.

[0074] When the wind speed measurement 74 is above the wind speed threshold, then the wind turbine will typically be operating in a full load mode as described above, in which the pitch angles of the blades are varied to capture a rated power with the rotor. When the check 80 indicates that the wind speed measurement 74 is above the wind speed threshold, then a second “double-check” blade anomaly detection mode is operated. In this second detection mode, the average blade moment 75 and the average pitch angle 76 are both analysed at 82. A blade anomaly is determined to exist at 82 if the analysis of the average pitch angle 76 indicates a pitch angle of the blade(s) which is less than expected and an analysis of the averaging blade moment 75 indicates a bending of the blade(s) which is more than expected.

[0075] A relatively high wind speed (above the rated wind speed) is indicated by a dashed vertical line 95 in Figures 8a and 8b. As shown in Figure 8a, for this high wind speed 95 the blade moment 96 for an iced blade is higher than the expected blade moment 97 for a non-iced blade. As shown in Figure 8b, for this high wind speed 95 the pitch angle 98 for an iced blade is lower than the expected pitch angle 99 for a non-iced blade. For this wind speed 95 when the wind turbine is operating in full load mode, the wind turbine is operating in the second (“double check”) detection mode.

[0076] The double check at 82 of both the blade moment and the pitch angle is advantageous, because it makes any determination of a blade anomaly in full load mode more robust. However, in alternative embodiments only the pitch angle may be checked at 82 in the second detection mode to determine a blade anomaly.

[0077] The outputs of the comparisons 81 , 82 enters a fusion block 83 which flags an ice event at 84.

[0078] The processing system 21 of Figure 5 may perform the process of Figure 9 to determine whether a blade anomaly exists and generate a corresponding output 24 which is indicated in Figure 5. This output 24 may simply be an indication on a display device, or a flag in a computer memory, that a blade anomaly does or does not exist. Alternatively, the output 24 may be fed to the wind turbine control system 25 which controls the wind turbine 1 based on the output 24. For example, if the output 24 indicates a blade anomaly then the wind turbine control system 25 may initiate ice adaption measures such as activating heaters on the blades 2a-c, pitching out the blades 2a-c to enter idle operation, suspending calibration procedures, or changing a control strategy of the wind turbine to extract more power from the rotor. Similarly, if the output 24 indicates that there is no blade anomaly, then such ice adaption measures may be stopped so the wind turbine returns to normal operation.

[0079] The information on blade condition provide by the above method could enable advanced control strategies, that can extract more power from the rotor. Instead of traditional mixed profile assumptions, the control strategy could use the estimated blade condition to define the control strategy.

[0080] Typically the blade anomaly detected in the process of Figure 9 is an ice-related blade anomaly, i.e. an indication that ice has accreted on the blade. The method may also be used to indicate a blade anomaly caused by accretion of another material on the blade.

[0081] Figure 10 replicates the curves of Figures 8a and 8b, and is annotated to indicate the rated wind speed and the two detection modes for the method of Figure 9. In this case the wind speed threshold which is used at step 80 to switch between the detection modes coincides with the cross-over point 61 between the two blade flap moment curves. The first detection mode (step 81 of Figure 9) is performed for wind speeds below the cross-over point 61 , and the second detection mode (step 82 of Figure 9) is performed for wind speeds above the cross-over point 61.

[0082] In the example of Figure 10, the wind speed threshold is offset from the rated wind speed, but in other embodiments the wind speed threshold may be the rated wind speed. Figure 11 shows an alternative method which employs two wind speed thresholds and three detection modes. The two wind speed thresholds are each offset from the rated wind speed by a predetermined amount.

[0083] A first wind speed threshold is offset below the rated wind speed by a predetermined amount (for instance by 1m / s so the first wind speed threshold is the rated wind speed - 1m / s).

[0084] The first “blade load only” detection mode is performed for wind speeds below the first wind speed threshold and above a cut-in wind speed. The first detection mode coincides with the partial load mode over a first range of wind speeds.

[0085] A second wind speed threshold is offset above the rated wind speed by a predetermined amount (for instance by 2m / s so the second wind speed threshold is the rated wind speed + 2m / s). The second (“double-check”) detection mode is performed for wind speeds above the second wind speed threshold. The second detection mode coincides with the full load mode over a second range of wind speeds.

[0086] For wind speeds between the first and second wind speed thresholds, a third “pitch-only” detection mode is used, in which only the pitch information is analysed to determine whether a blade anomaly exists and generate a corresponding output. The pitch-only detection mode coincides with both the partial load mode and the full load mode.

[0087] Figure 12 shows an alternative method of changing detection mode based on wind speed. Certain steps of the method of Figure 12 are the same as in Figure 9, and are given the same reference number.

[0088] For temperatures below 3degC a Moment ratio may be calculated at 90 by the algorithm:

[0089] Moment ratio = |(Moment reference - Current moment)| / Moment reference The Moment reference is the calibrated blade moment reference for the current wind speed bin (as determined at 78), and the Current moment is the average blade moment for the current wind speed bin (as determined at 75).

[0090] The Moment ratio calculated at 90 is a positive number which is the absolute or modulus of the difference between the Current moment and the Moment reference. The absolute or modulus is used to ensure that the Moment ratio is a positive value above and below the cross-over point 61. The Moment ratio is calculated at 90 for all wind speeds above the cut-in wind speed.

[0091] Alternatively, the method of calculating the Moment ratio at 90 may depend on the wind speed relative to the wind speed threshold coinciding with the cross-over point 61 in Figure 10.

[0092] For example, at low wind speeds below the wind speed threshold, the Moment ratio may be calculated at 90 by the algorithm:

[0093] Moment ratio = (Moment reference - Current moment) / Moment reference

[0094] This algorithm only looks at positive values for this Moment ratio. In other words, the algorithm ignores negative values. As shown in Figure 10, below the wind speed threshold the Moment ratio will tend to be a positive number.

[0095] At high wind speeds above the wind speed threshold, the Moment ratio may be calculated at 90 by the algorithm:

[0096] Moment ratio = (Moment reference - Current moment) / Moment reference

[0097] This algorithm only looks at negative values for this Moment ratio. In other words, the algorithm ignores positive values. As shown in Figure 10, above the wind speed threshold the Moment ratio will tend to be a negative number.

[0098] For temperatures below 3degC a Pitch ratio is also calculated at 91 by the algorithm:

[0099] Pitch ratio = (Pitch reference - Current pitch) / Pitch reference The Pitch reference is the calibrated pitch reference for the current wind speed bin (as determined at 78), and the Current pitch is the average pitch angle for the current wind speed bin (as determined at 76).

[0100] The pitch ratio calculated at 91 is a positive number (or very small) for all wind speeds so the absolute or modulus is not required. The Pitch ratio is calculated at 91 for all wind speeds above the cut-in wind speed.

[0101] At 92 the Moment ratio is multiplied by a bending moment weight shown in Figure 13, to output a weighted moment ratio. The bending moment weight varies based on wind speed as a first sigmoid function 101.

[0102] At 93 the Pitch ratio is multiplied by a pitch angle weight shown in Figure 13, to output a weighted pitch ratio. The pitch angle weight varies based on wind speed as a second sigmoid function 102. The functions 101, 102 meet at a crossing-point of about 10 m / s which may be the rated wind speed (as in this case) or may be offset by a predetermined amount from the rated wind speed.

[0103] At 94 the weighted moment ratio is added to the weighted pitch ratio (or combined in some other way) to generate a combined output value.

[0104] The combined output value is then analysed at 95 to determine whether a blade anomaly exists and generate a corresponding output. For example the combined output value may be compared with a threshold (for instance 0.2) and an icing event is flagged if the combined output value is above 0.2.

[0105] Optionally the analysis at 95 may use a CUSIIM algorithm or other sequential analysis technique to detect a change in the combined output value which is indicative of an ice event, rather than a simple comparison with a threshold.

[0106] CUSIIM (Cumulative Sum) is a statistical method used to detect small, persistent changes in a process over time. It works by calculating the cumulative sum of deviations from a target value, making it effective for identifying subtle shifts or trends in the combined output value. The effect of the sigmoid functions 101 , 102 is to transition gradually between detection modes based on wind speed, rather than switching suddenly based on a wind speed threshold.

[0107] For very low wind speeds (below about 6 m / s) the second sigmoid function 102 is zero, so the wind turbine is operating in a blade load only detection mode in which only the bending information is used to determine whether a blade anomaly exists and generate a corresponding output.

[0108] For very high wind speeds (above about 14 m / s) the first sigmoid function 101 is zero, so the wind turbine is operating in a pitch only detection mode in which only the pitch information is used to determine whether a blade anomaly exists and generate a corresponding output.

[0109] Between about 6 m / s and 14 m / s the wind turbine operates in mixed detection modes in which the bending information and the pitch information are both analysed to determine whether a blade anomaly exists and generate a corresponding output. In the mixed detection modes the determination of the blade anomaly is made on the basis of bending information, pitch information, bending moment weight 101 and pitch angle weight 102. For example the bending information and the pitch information may be combined at 94 based on their respective weights to generate a combined output value which is used to detect the blade anomaly at 95. Below the crossing point of 10 m / s in Figure 12 the bending moment weight 101 is higher so the wind turbine is operating in a first detection mode (a “blade load dominant” mixed detection mode) in which the bending information is dominant; and above the crossing point the pitch angle weight 102 is higher so the wind turbine is operating in a second detection mode (a “pitch dominant” mixed detection mode) in which the pitch information is dominant.

[0110] Figure 14 shows more shallow sigmoid functions 103, 104 which may be used instead of the relatively steep sigmoid functions 101, 102 of Figure 13. The shallower sigmoid functions 103, 104 result in the wind turbine operating in only two modes: a first detection mode (a “blade load dominant” mixed detection mode) below the rated wind speed; and a second detection mode (a “pitch dominant” mixed detection mode) above the rated wind speed. To sum up: the methods of Figure 9 and Figure 12 obtain bending information (for example an average blade moment 75) indicating bending of one or more of the blades; obtain pitch information (for example an average pitch angle 76) indicating a pitch angle of one or more of the blades; and measure wind speed to obtain wind speed measurements (for example a wind speed 74 associated with a wind speed bin). The wind speed measurements may be associated with the bending information and the pitch information.

[0111] The wind speed measurements could be average values of different kinds - for instance each wind speed measurement may be an average wind speed 74 of a bin.

[0112] The bending information may be obtained by assigning a plurality of bending measurements to a bin based on wind speed, and combining (e.g. averaging) the bending measurements in the bin to obtain the bending information. In this case the bending information may be analysed at 81, 82 or 90 to determine whether a blade anomaly exists by comparing the bending information to a stored reference associated with the bin.

[0113] The pitch information may be obtained by assigning a plurality of pitch measurements to a bin based on wind speed, and combining (e.g. averaging) the pitch measurements in the bin to obtain the pitch information. In this case the pitch information may be analysed at 82 or 91 to determine whether a blade anomaly exists by comparing the pitch information to a stored reference associated with the bin.

[0114] When wind speed measurements are below a wind speed threshold, a first detection mode may be operated in which the bending information is analysed to determine whether a blade anomaly exists and generate a corresponding output. The first detection mode either completely coincides with the partial load mode so the two modes operate over the same range of wind speeds as in Figure 14; or the first detection mode and the partial load mode may operate over partially overlapping wind speeds as in Figures 10-12.

[0115] More generally the first detection mode coincides with the partial load mode over a first range of wind speeds. For example in Figures 10 and 14 the first range of wind speeds is between the cut-in wind speed and the rated wind speed. In Figure 11 the first range of wind speeds is between the cut-in wind speed and the first wind speed threshold. In Figure 13 the first range of wind speeds is between 4 m / s and 6 m / s or between 6 m / s and 10 m / s.

[0116] When wind speed measurements are above a wind speed threshold, a second detection mode is operated in which the pitch information is analysed to determine whether a blade anomaly exists and generate a corresponding output. The second detection mode either completely coincides with the full load mode so the two modes operate over the same range of wind speeds as in Figure 14; or the second detection mode and the full load mode may operate over partially overlapping wind speeds as in Figures 10, 11 and 13.

[0117] More generally the second detection mode coincides with the full load mode over a second range of wind speeds, which is higher than the first range of wind speeds. In Figure 10 the second range of wind speeds is between the wind speed threshold and the cut-out wind speed. In Figure 11 the second range of wind speeds is between the second wind speed threshold and the cut-out wind speed. In Figure 13 the second range of wind speeds is between 10 m / s and 14 m / s or between 14 m / s and 16m / s. In Figure 14 the second range of wind speeds is between 10 m / s (rated wind speed) and 16 m / s (cut-out wind speed).

[0118] In the first detection mode, a blade anomaly may be determined to exist if the analysis 81 of the bending information indicates bending of the blade(s) which is less than expected. In the case of Figure 12, in the first detection mode a blade anomaly may be determined to exist if the analysis 90 of the bending information indicates bending of the blade(s) which is different to an expected reference (the moment reference).

[0119] In the second detection mode, a blade anomaly may be determined to exist if the analysis 82 of the pitch information indicates a pitch angle of the blade(s) which is less than expected and the analysis 82 of the bending information indicates a bending of the blade(s) which is more than expected. In the examples above, wind speed measurements may be used to determine the detection mode, for example by comparison with one or more wind speed thresholds. That is, wind speed measurements may be used to determine whether to analyse the bending information or the pitch information to determine whether a blade anomaly exists (or in the case of Figures 12-14 to decide which information is more dominant). The wind speed measurements may be obtained from a wind speed sensor, or estimated based on other inputs.

[0120] In other embodiments of the invention, wind speed measurements may not be used to switch between the detection modes. For example the power generation mode may be used to infer the wind speed, and to act as a trigger to switch between detection modes. For example a switch between detection modes may be triggered by a switch between partial load mode and full load mode.

[0121] In this case, the power generation mode may be detected by monitoring the power or any other parameter. For instance: when power is below rated power, then the first detection mode is operated; and when power is at or above rated power, then the second detection mode is operated.

[0122] The embodiments above provide a method of monitoring one or more blades of the rotor of a wind turbine based on bending information indicating bending of one or more of the blades as the rotor rotates. The bending information could be strain data obtained from one or more edge strain gauges, from one or more flap strain gauges, or from a combination of both flap and edge strain gauges. Other embodiments of the invention may obtain the bending information from other sensors, for example a displacement sensor, camera, lidar sensor or accelerometer. The bending information may be obtained directly from a specific sensor or from a fusion of different sensors. The bending information may be obtained by directly measuring bending of the blade, or by indirectly measuring bending of the blade using an engineering model. The bending information for a particular blade may also originate from a model based on the other two blades of the rotor.

[0123] Although the invention has been described above with reference to one or more preferred embodiments, it will be appreciated that various changes or modifications may be made without departing from the scope of the invention as defined in the appended claims

Claims

CLAIMS1. A method of monitoring one or more blades of a rotor of a wind turbine, the method comprising:obtaining bending information indicating bending of one or more of the blades; obtaining pitch information indicating a pitch angle of one or more of the blades; when wind speed is below a rated wind speed: operating the wind turbine in a partial load mode in which pitch angles of the blades are controlled to maximise power captured by the rotor;operating a first detection mode in which the bending information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the first detection mode coincides with the partial load mode over a first range of wind speeds;when wind speed is above the rated wind speed: operating the wind turbine in a full load mode in which the pitch angles of the blades are varied to capture a maximum allowed power with the rotor; andoperating a second detection mode in which the pitch information is analysed to determine whether a blade anomaly exists and generate a corresponding output, wherein the second detection mode coincides with the full load mode over a second range of wind speeds.

2. A method according to claim 1 , wherein in the first detection mode a blade anomaly is determined to exist if the analysis of the bending information indicates bending of the blade(s) which is less than expected.

3. A method according to claim 1 or 2, wherein in the second detection mode a blade anomaly is determined to exist if the analysis of the pitch information indicates a pitch angle of the blade(s) which is less than expected.

4. A method according to any preceding claim, wherein in the first detection mode and / or the second detection mode the bending information and the pitch information are both analysed to determine whether a blade anomaly exists and generate a corresponding output.

5. A method according to claim 4, wherein in the second detection mode a blade anomaly is determined to exist if the analysis of the pitch information indicates a pitch angle of the blade(s) which is less than expected and an analysis of the bending information indicates a bending of the blade(s) which is more than expected.

6. A method according to any preceding claim, wherein the bending information is obtained by assigning a plurality of bending measurements to a bin based on wind speed, and combining the bending measurements to obtain the bending information; and the pitch information is obtained by assigning a plurality of pitch angle measurements to a bin based on wind speed, and combining the pitch angle measurements to obtain the pitch information.

7. A method according to claim 6, wherein in the first detection mode the bending information is analysed to determine whether a blade anomaly exists by comparing the bending information to a stored reference associated with the bin.

8. A method according to claim 6 or 7, wherein in the second detection mode the pitch information is analysed to determine whether a blade anomaly exists by comparing the pitch information to a stored reference associated with the bin.

9. A method according to any preceding claim, wherein the first detection mode operates below a wind speed threshold which is the same as the rated wind speed, or offset from the rated wind speed by a predetermined amount.

10. A method according to any preceding claim, wherein the second detection mode operates above a wind speed threshold which is the same as the rated wind speed, or offset from the rated wind speed by a predetermined amount.

11. A method according to any preceding claim, further comprising measuring wind speed to obtain wind speed measurements, and comparing the wind speed measurements with the rated wind speed to determine whether to operate in partial load mode or full load mode.

12. A method according to any preceding claim, further comprising measuring wind speed to obtain wind speed measurements, and using the wind speed measurements to determine whether to operate in the first or second detection mode.

13. A method according to any preceding claim, wherein the first and second detection modes are each mixed detection modes in which the bending information and the pitch information are both analysed to determine whether a blade anomaly exists and generate the corresponding output; in the first detection mode the bending information is dominant; and in the second detection mode the pitch information is dominant.

14. A method of controlling a wind turbine, the method comprising monitoring one or more blades of a rotor of the wind turbine by a method according to any preceding claim; and controlling the wind turbine based on the output.

15. Apparatus for monitoring one or more blades of a rotor of a wind turbine, wherein the apparatus is configured to perform a method according to any preceding claim.

Images