WO2017035325A1

WO2017035325A1 - Techniques for determining yaw misalignment of a wind turbine and system and method using the same

Info

Publication number: WO2017035325A1
Application number: PCT/US2016/048629
Authority: WO
Inventors: Thomas J. Nostrand; Enrique D. ANGOLA; Junda ZHU
Original assignee: Nrg Systems Inc.
Priority date: 2015-08-25
Filing date: 2016-08-25
Publication date: 2017-03-02

Abstract

Techniques are disclosed for actively determining a yaw misalignment value for a wind turbine based in part on identifying the position of a tower shadow effect that occurs the moment a blade passes in front of the wind turbine tower. The present disclosure has identified that when a wind turbine is properly aligned with the wind, tower shadowing occurs the moment when a blade is substantially in front of the tower (e.g., 6 o'clock position). A wind turbine control system consistent with the present disclosure correlates incidences of tower shadowing effect with blade position at the moment the tower shadowing occurs, and therefore may identify whether tower shadowing occurs substantially directly in front of the tower, and thus, the wind turbine is properly aligned with oncoming wind, or whether tower shadowing occurs prior to or after passing through the front of the tower, thus indicating a yaw misalignment.

Description

TECHNIQUES FOR DETERMINING YAW MISALIGNMENT OF A WIND TURBINE AND

SYSTEM AND METHOD USING THE SAME

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 62/209,830 filed on August 25, 2015, the entire teachings of which are hereby incorporated herein by reference.

TECHNICAL FIELD

[0002] This disclosure relates to wind turbines, and in particular, to determining a yaw misalignment value based on identifying the position of tower shadow effect relative to blade position.

BACKGROUND

[0003] Wind turbines operate optimally when properly aligned with incoming wind. However, changes in wind speed and direction occur on a regular basis. To compensate, control mechanisms associated with a wind turbine seek to ensure the main rotor axis of the turbine is parallel to the wind direction. When such alignment occurs, the wind turbine is generally said to have a yaw angle error of 0 degrees.

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Reference should be made to the following detailed description which should be read in conjunction with the following figures, wherein like numerals represent like parts:

[0005] FIG. 1 shows an example top plan view of a wind turbine consistent with the present disclosure;

[0006] FIG. 2 shows an example wind turbine and illustrates a position of the blades where a momentary drop in rotations per minute (RPM) occurs due to tower shadow effect;

[0007] FIG. 3 illustrates a profile view of an example wind turbine and illustrates positions of the blades where effects of tower shadowing are relatively small or non-existent;

[0008] FIG. 4A is a schematic illustration of a wind turbine control system consistent with an embodiment of the present disclosure;

[0009] FIG. 4B is a perspective view of an example wind turbine implementing the wind turbine control system of FIG. 4A in accordance with an embodiment of the present disclosure; [0010] FIG. 5 shows an example process for detecting a yaw misalignment during operation of a wind turbine in accordance with an embodiment of the present disclosure;

[0011] FIG. 6 shows an example process for calculating a yaw misalignment value based on identifying the position of tower shadow effect relative to the position of wind turbine blades, in accordance with an embodiment of the present disclosure;

[0012] FIG. 7 shows a plurality of time- synchronized shaft speed data samples and blade position data samples received during performance of the example process of FIG. 6, in accordance with an embodiment of the present disclosure;

[0013] FIG. 8 shows the plurality of shaft speed data samples and blade position data samples of FIG. 7 separated into chunks during performance of the example process of FIG. 6, in accordance with an embodiment of the present disclosure;

[0014] FIG. 9 shows a plurality of example composite waveforms in a 1 -revolution domain for the chunks of FIG. 8 after time synchronous averaging (TSA) is performed on the chunks during the example process of FIG. 6, in accordance with an embodiment of the present disclosure;

[0015] FIG. 10 shows an example shaft speed composite waveform in a 1-revolution domain prior to band-pass filtering, in accordance with an embodiment of the present disclosure;

[0016] FIG. 11 shows an example graph that plots a shaft speed composite waveform and blade position composite waveform in a 1-revolution domain and illustrates a zero or near- zero yaw misalignment value based on an associated phase shift of about 60 degrees, in accordance with an embodiment of the present disclosure;

[0017] FIG. 12 shows another example graph that plots a shaft speed composite waveform and blade position composite waveform in a 1-revolution domain and illustrates a yaw misalignment value greater than zero value based on an associated phase shift being less than 60 degrees, in accordance with an embodiment of the present disclosure;

[0018] FIG. 13 is an example polar plot that shows a zero or near- zero yaw misalignment, e.g., based on a phase shift of 60 degrees between a shaft speed composite waveform and blade position composite waveform in a 1-revolution domain, in accordance with an embodiment of the present disclosure; and

[0019] FIG. 14 is another example polar plot that shows a yaw misalignment greater than zero, e.g., based on a phase shift of less than 60 degrees between a shaft speed composite waveform and blade position composite waveform in a 1 -revolution domain, in accordance with an embodiment of the present disclosure.

DETAILED DESCRIPTION

[0020] Wind turbines often include active yaw control strategies to mitigate yaw misalignment, which is also referred to as yaw error. FIG. 1 shows a top plan view of an example wind turbine 102 relative to the direction of wind 104. As shown, the wind 104 includes a vector with a direction component 108 that extends substantially in parallel with the main rotor axis 106 of the wind turbine 102. The component of the vector that comes in perpendicular to the rotor of the wind turbine 102 produces energy. A wind turbine in this orientation has a theoretical yaw misalignment of zero. In a more general sense, a yaw misalignment of zero means a wind turbine "faces" oncoming wind such that a substantial portion of incoming wind encounters the rotor in a perpendicular fashion. In reality, wind direction changes frequently and randomly relative to the main rotor axis 106, and wind turbine control systems must continuously adjust to prevent yaw misalignment from negatively impacting energy production. Wind encountering the wind turbine 102 at an angle Θ offset from main rotor axis 106 represents yaw misalignment.

[0021] Some active yaw correction strategies include determining wind direction and wind speed using an anemometer and wind vane. Unfortunately, such devices generally require precise calibration in production to account for deployment characteristics, e.g., the environmental and geographical peculiarities associated with the location of a deployed wind turbine, and a control system that utilizes sensor data effectively and accurately. As a result, many wind turbines operate with yaw misalignment of about two percent to four percent, or more, despite active yaw alignment strategies. Even a one percent further reduction in yaw misalignment may yield a substantial increase in power generation for a given wind turbine on an annual basis.

[0022] Thus, in accordance with an embodiment, techniques are disclosed for actively determining a yaw misalignment value for a wind turbine based in part on identifying the position of a tower shadow effect that occurs the moment a blade passes in front of the wind turbine tower. As discussed in greater detail below, the present disclosure has identified that when a wind turbine is properly aligned with the wind, e.g., with a yaw misalignment under one degree or about zero, tower shadowing occurs the moment when a blade is substantially in front of the wind turbine tower. A measurable drop in power, or rotations per minute (RPM), characterizes tower shadowing as the tower structure adds resistance to the wind flow which causes velocity and pressure fluctuations. A wind turbine control system consistent with the present disclosure may then utilize a measured tower shadowing effect, and more particularly a drop in rotations per minute (RPM) of a rotor, e.g., using a tachometer, due to tower shadowing, in combination with a sensor, e.g., a key phasor, that detects blade position relative to the wind turbine tower.

[0023] Correlating incidences of tower shadowing effect with blade position at the moment the tower shadow effect occurs allows the wind turbine controller to identify whether tower shadowing occurs substantially directly in front of the wind turbine tower, and thus, has a zero or near-zero yaw misalignment. Conversely, the wind turbine controller may determine a yaw misalignment value based on incidences of tower shadowing that occur away from substantially directly in front of the wind turbine tower.

[0024] The wind turbine control system may utilize the determined yaw misalignment to perform yaw correction, and/or report an indication of yaw misalignment to a wind farm controller, for example. In some cases, the wind turbine control system corrects yaw misalignment by sending one or more signals to a yaw control arrangement to cause a yaw motor to rotate a nacelle of the wind turbine for instance. The wind turbine control system may also send sensor data to, for example, a remote computing device such as a wind farm controller, with the wind farm controller performing yaw misalignment detection in accordance with aspects and embodiments of the present disclosure. In this example, the wind farm controller may be accurately referred to as a yaw misalignment detector.

[0025] Yaw correction may include incremental changes, e.g., less than 1 degree of corrective rotation, in order for additional yaw misalignment determinations to occur prior to further corrective rotation. In this manner, relatively fine-grain yaw correction may occur over long periods of time, e.g., over hours, days, weeks, months, and so on. The turbine control system may also use sensor data for other sensors, such as data from an anemometer and wind vane, when detecting yaw misalignment. For example, if mean wind speed reported by an anemometer is zero, then yaw misalignment detection may be delayed as tower shadowing is unlikely to occur in this scenario. In still other cases, the detected yaw misalignment value may be used to validate wind sensor data. Accordingly, the turbine control system may utilize the determined yaw misalignment value to reduce ineffective yaw correction that would otherwise occur without the ability to validate data from wind sensing devices. Moreover, a technician may use a determined yaw misalignment value to adjust an orientation of a wind sensor mounting bracket or other portion of a wind sensor, e.g., firmware, to correct for a misconfiguration.

[0026] Various examples and scenarios disclosed herein include a horizontal axis, three-blade (3-blade) turbine that includes a rotor system forward of the nacelle (upward turbine), although this disclosure is not necessarily limited in this regard. Various aspects and embodiments disclosed herein also are also applicable to other wind turbine configurations, e.g., wind turbines that have 2-blades, with minor modification and are also within the scope of this disclosure.

[0027] With reference to FIG. 2, some aspects of tower shadowing are best understood by way of illustration. Tower shadowing phenomenon is common to virtually all horizontal axis, 3- blade turbines that include a rotor system forward of the nacelle. Tower shadowing phenomenon, also known as tower shadowing effect, generally refers to the loss of angular momentum suffered by the rotor of a wind turbine as a consequence of the alteration of wind flow caused by the presence of the tower. When a wind turbine is aligned with the wind such that wind encounters the rotor in a substantially perpendicular fashion, as previously discussed, tower shadowing occurs generally at the moment when a blade travels substantially directly in front of the tower. FIG. 2 illustrates a wind turbine 102 having a blade 110 in front of the tower 112, at about a 6 o'clock position. Supposing the wind direction is as shown, the effects of tower shadowing are greatest at this moment. Tower shadowing manifests as a drop in power generation due to the decrease of the rotor's RPM. Within the context of FIG. 2, such tower shadowing occurs three (3) times for every one (1) rotation of the rotor, e.g., 3 cycles/ one full revolution.

[0028] On the other hand, and as shown in FIG. 3, when blades travel 60 degrees away from the tower 112, the effects from tower shadowing on power generation/RPM are relatively low or otherwise non-existent. As further shown, the blades are separated by 120 degrees such that only one blade passes by the tower 112 at a time. As such, the maximum power generation (e.g., RPM) for the wind turbine occurs when each blade is 60 degrees before/after the tower 112. Therefore, the waveform representing power generation (or RPM) is sinusoidal and for a given rotation of a rotor includes 3 cycles/per revolution (e.g., each of the three blades pass by the tower 112 once per one revolution). The RPM waveform may be filtered, e.g., to remove noise and other conditions not related to tower shadowing, and processed using, for example, time synchronous averaging to transform the RPM waveform from the time domain to a 1 -revolution domain.

[0029] The RPM waveform in a 1 -revolution domain, also known as a composite RPM waveform, allows the same to be compared to a composite waveform representing blade position also in the same 1 -revolution domain. Correlation strategies may then be utilized to determine whether the RPM drops due to tower shadowing occur at the moment each blade passes substantially in front of the tower 112, thus indicating a yaw misalignment of zero or near- zero, or if tower shadowing is occurring at a point before/after a position away from the 6 o'clock blade position shown in FIG. 2, thus indicating a yaw misalignment value greater than zero. FIGs. 11-12 illustrate some such example composite waveforms in the 1-revolution domain.

[0030] The present disclosure has identified that a phase shift of about 60 degrees between a shaft speed waveform (e.g., an RPM waveform) and a blade position waveform in a 1-revolution domain indicates, in a general sense, proper alignment of a wind turbine. FIG. 11 illustrates one such example of a shaft speed composite waveform and blade position composite waveform in a 1-revolution domain having an associated phase shift of 60 degrees. On the other hand, the present disclosure has identified a phase shift of less than or greater than 60 degrees indicates a yaw misalignment condition. FIG. 12 illustrates one such example of a shaft speed composite waveform and blade position composite waveform in a 1-revolution domain having an associated phase shift less than 60 degrees.

[0031] Although the present disclosure specifically references transforming signals from a time domain to a 1-revolution domain for comparison, this disclosure is not necessarily limited in this regard. For example, an N-revolution domain such as a 2-revolution, 3-revolution, 4- revolution domain, or 10-revolution domain may be used depending on a desired configuration. Moreover, while the present disclosure specifically references using time domain averaging (TSA) to transform signals from the time domain into an N-revolution domain, other approaches to correlating blade position with a drop in RPM, e.g., tower shadow effect, may be utilized. For example, raw signals corresponding to shaft speed and blade position may be processed and analyzed in the time domain using various signal processing techniques, e.g.,., filtering and averaging, to determine a yaw misalignment value. However, TSA advantageously allows for raw signals to be framed/represented in an N-revolution domain, e.g., made synchronous with the wind turbine shaft as discussed in greater detail below, which allows for a relatively simple and efficient way of extracting information from signals, such as a RPM signal, that are characterized by a relatively small dynamic range mixed with noise.

[0032] The yaw misalignment value may therefore be simply a delta D (or difference D) between a measured phase shift and the expected phase shift of 60 degrees. The actual degree of yaw misalignment for a wind turbine may not necessarily equal the delta between a 60-degree phase shift and the measured result. For example, a measured phase shift of 64 degrees does not necessarily equal a 4-degree yaw misalignment of the wind turbine. Instead, the correlation between a measured phase shift and actual degrees of misalignment is dependent on multiple factors including, but not limited to, geometry of the wind turbine, length of each blade, horizontal distance between tower and blades, location of the nacelle relative to the base of the tower, and so on. Therefore, known geometries and wind turbine properties may be utilized by a wind turbine control system to directly convert a yaw misalignment value to an actual degree of misalignment.

[0033] Alternatively, a wind turbine control system consistent with the present disclosure may use the determined yaw misalignment as an indicator of misalignment and apply a relatively small, e.g., about 0.1 to 4 degrees, incremental correction over time until the turbine control system determines that the rotor is aligned, e.g., the rotor has a zero or near-zero yaw misalignment value. In other cases, production and/or deployment of a wind turbine may include tests to develop a calibration curve to empirically or heuristically establish a lookup table or other mechanism to correlate a determined yaw misalignment value with an actual degree of misalignment. For example, a procedure may include purposely misaligning a wind turbine by a predetermined degree and acquiring a corresponding yaw misalignment value. This procedure may occur a number of times to derive a plurality of data points. These data points may then be used in combination with curve fitting approaches (e.g., non-linear regression, statistical regression, and so on) to estimate a degree of misalignment. Turbine control systems may perform such interpolation during operation and/or may utilize lookup tables based in part on a calibration curve, for example, to translate a calculated yaw misalignment values into a value that represents an actual degree of misalignment. [0034] As generally referred to herein, the term signal refers to any electrical quantity or effect (e.g., current, voltage, or electromagnetic waves, and so on), that can be varied in such a way as to convey information. Signals may comprise a portion of information, e.g., data, and a portion of noise. As used herein, the term "noise" when used to describe a signal or waveform refers to non-informational portions of a signal that may be introduced during, for example, yaw movement, wind shear, and other anomalies. Sensory of a wind turbine, e.g., a shaft speed sensor and a blade position sensor, may provide signals in a so-called "raw" form with that form being set by the manufacturer of a given device. A wind turbine control system, and more particularly a controller or circuit thereof, may convert such raw signals into, for example, digital representations, e.g., bits, or other forms that are suitable for capturing and processing signals. However, the term signal should not be construed as limited to "raw" signals; rather, the term generally refers to any signal, either raw or processed (e.g., digitized, filtered, resampled, and so on), that may be used by a turbine control system to preform various yaw misalignment detection processes disclosed herein.

Example Wind Turbine Control System and Operation

[0035] Turning to FIG. 4A, an example wind turbine control system 400 is shown in accordance with an embodiment of the present disclosure. The example wind turbine control system 400 is illustrated in a highly simplified form and other configurations are also within the scope of this disclosure. The example wind turbine control system 400 may also be referred to as a yaw misalignment detector arrangement, or yaw misalignment detector. As shown, the example wind turbine control system 400 includes a housing 402. The housing 402 may comprise, for example, a nacelle of a wind turbine, such as the nacelle 403 of the horizontal axis, 3-blade wind turbine 450 shown in FIG. 4B.

[0036] As further shown, the housing 402 includes a controller 404, yaw control 406, pitch control 408, power conversion and control circuitry 416, and main shaft 422. The wind turbine control system 400 includes an optional wind sensor 410, blade position sensor 412, and shaft speed sensor 414.

[0037] The yaw control 406 includes associated circuitry and one or more controllers to control the yaw motor 407 for rotating the nacelle about a yaw axis, for example. The pitch control 408 includes one or more controllers and associated circuitry configured to adjust the pitch of turbine blades, such as the blades collectively shown as 452 and individually shown as blades 452-1, 452-2 and 452-3 in FIG. 4B. The main shaft 422 couples to the rotor 424 for power generation purposes. The main shaft 422 couples to power generation equipment, e.g., the power conversion and control circuitry 416, via one or more shafts (not shown) such as an intermediate shaft and a high-speed shaft, for example. The main shaft 422 may be referred to as a low- speed shaft.

[0038] The shaft speed sensor 414 comprises, for example, a tachometer device or other suitable device for measuring RPM of a shaft, such as the main shaft 422. Some shaft speed sensors, such as tachometers, may be configured to output a signal with N pulses per revolution of a shaft. For instance, a digital tachometer may output a bit stream that represents shaft speed while an analog tachometer may output a sinusoidal waveform. In any such cases, an output signal from a tachometer may be easily converted to an equivalent RPM signal by the controller 404, or other suitable circuitry, and for the purpose of clarity and ease of description this disclosure may generally refer to the output signal as an RPM signal without necessarily discussing intermediate conversion. Also note that this disclosure is not necessarily limited to using an RPM signal and may utilize the N pulse per revolution signal in various yaw misalignment detection and correction processes disclosed herein without necessarily performing conversion. The shaft speed sensor 414 couples to the main shaft 422 and comprises circuitry configured to measure RPM of the main shaft 422 and output a proportional electrical signal to the controller 404. The shaft speed sensor 414 may further include conversion circuitry such as AC/DC converters depending on a desired configuration. Alternatively, or in addition to coupling to the main shaft 422, the shaft speed sensor 414 may couple to any shaft that is part of the power generating arrangement connected to the rotor 424. In some cases, the wind turbine control system 400 includes a dedicated shaft speed sensor (e.g., at least a first and second tachometer device) for each shaft. However, the RPM measurements from the main shaft 422 may be converted to equivalent high-speed RPM measurements, and vice-versa, by simple calculation that divides or multiplies by the known gear ratios, and therefore, multiple shaft speed sensors are not necessarily required.

[0039] Power conversion and control circuitry 416 includes one or more controllers (not shown) and associated circuitry for converting rotations of the rotor 424 into power, and for supplying power to components of the wind turbine control system 400. Power conversion and control circuitry 416 may further include AC to DC conversion circuitry for converting power generated by the rotors into a stable, usable DC voltage for internal loads of the wind turbine control system.

[0040] The optional wind sensor 410 may comprise any suitable device for measuring of wind components, e.g., wind speed and/or direction of wind 420, such as a digital/analog anemometer, a digital/analog wind vane, and a LIDAR-based sensor, just to name a few. The optional wind sensor 410 may include circuitry configured to provide an electrical signal to the controller 404 that includes measured wind speed and direction.

[0041] The blade position sensor 412 comprises, for example, a key phasor or other suitable device for detecting the presence of a blade, such as the blades 452. The blade position sensor 412 may be disposed along the tower of a wind turbine adjacent the blades 452, such as on the tower 454 of the wind turbine 450 as shown in FIG. 4B. The blade position sensor 412 may output a pulse each time each of the blades 452 pass by.

[0042] The controller 404 comprises at least one processing device/circuit such as, for example, a digital signal processor (DSP), a field-programmable gate array (FPGA), Reduced Instruction Set Computer (RISC) processor, x86 instruction set processor, microcontroller, and an application- specific integrated circuit (ASIC). The controller 404 may be implemented, for example, using software (e.g., C or C++ executing on the controller/processor 404), hardware (e.g., hardcoded gate level logic or purpose-built silicon) or firmware (e.g., embedded routines executing on a microcontroller), or any combination thereof. The controller 404 may further include a hardware and/or software clock for timestamping data samples received from, for example, the shaft speed sensor 414 and the blade position sensor 412. In an embodiment, the controller 404 may be configured to carry out the processes 500 and 600 of FIGs. 5 and 6, respectively.

[0043] In operation, the controller 404 may perform active yaw alignment processes on a periodic basis, e.g., daily, hourly, weekly, monthly, and so on. During active yaw alignment the controller 404 receives sensor data from the shaft speed sensor 414 and blade position sensor 412 to correlate incidences of tower shadow effect with blade position. Analog data from sensors may be digitized and/or converted as needed. The controller 404 may use signal processing techniques such as filtering and time synchronous averaging (TSA) to transform time- synchronized samples representing blade position and tower shadow effect into composite waveforms in an N-revolution domain for comparison purposes. For example, signals in the time domain may be transformed to a 1-revolution domain using time synchronous averaging (TSA), and filtered accordingly, e.g., using a band-pass filter, in order to generate composite waveforms that represent 3 cycles (e.g., 3 blades pass by the tower) per revolution of a rotor. Cross- correlation may then be used in order to calculate the phase shift between composite waveforms representing shaft speed and blade position in the 1-revolution domain.

[0044] Composite waveforms, as generally referred to herein, refer to waveforms that may be formed from a combination of a number of sinusoidal waves or other basis functions summed together. FIGs. 10-12 illustrate some such example composite waveforms in a 1-revolution domain, and are discussed in more detail further below.

Example Yaw Alignment Processes and Architecture

[0045] A yaw misalignment detection process may be implemented in a variety of ways. FIG. 5 is a flow chart illustrating one yaw misalignment detection process 500 useful in connection with a system and method consistent with the present disclosure. While flowcharts presented herein illustrate various operations according to example embodiments, it is to be understood that not all of the depicted operations are necessary for other embodiments. Indeed, it is fully contemplated herein that in other embodiments of the present disclosure, the depicted operations, and/or other operations described herein, may be combined in a manner not specifically shown in any of the drawings, but still fully consistent with the present disclosure. Thus, claims directed to features and/or operations that are not exactly shown in one drawing are deemed within the scope and content of the present disclosure.

[0046] The yaw misalignment detection process 500 may be performed by the controller 404, or any other suitable component or combination of suitable components of the wind turbine control system 400. In some cases, the process 500 may be performed by a computing device remote from a wind turbine. For example, a wind farm controller, or server computer, may receive a plurality of samples representing shaft speed and blade position from the wind turbine control system 400. The wind farm controller may then perform the yaw misalignment detection process 500 and/or the process 600, in whole or in part, depending on a desired configuration.

[0047] In any such cases, the yaw misalignment detection process 500 may be performed periodically, e.g., hourly, daily, weekly, monthly, and so on. In some cases, the yaw misalignment detection process 500 is performed continuously to ensure proper alignment of the nacelle relative to the wind.

[0048] In act 502, the controller 404 calculates a yaw misalignment value X based in part on a phase shift between a time- synchronized shaft speed composite waveform and blade position composite waveform in a 1 -revolution domain. One such example calculation to determine the yaw misalignment value X is shown and described in greater detail with regard to the example process 600 of FIG. 6, which is discussed in greater detail below.

[0049] In act 504, the controller 404 determines if the calculated yaw misalignment value X exceeds a predetermined threshold value Y. In some cases, the predetermined threshold value Y is about 1.0 degrees, although other larger and or smaller threshold values are within the scope of this disclosure. For example, a threshold value of between 0.1 to 0.5 degrees may be utilized. In another non-limiting example, the threshold value may be greater than 1.0 degrees, such as 2.0 degrees, 3.0 degrees, and so on. If the calculated yaw misalignment value X is greater than the predetermined threshold value Y, the process 500 continues to act 506. Otherwise, the process 500 continues to act 510.

[0050] In act 510, the controller 404 may initiate a timer or otherwise delay further yaw misalignment detection. For example, the controller 404 may implement a programmable schedule that waits for a period of time X to elapse prior to returning to act 502. In other cases, the controller 404 returns to act 502 without delay to continuously perform yaw misalignment detection.

[0051] In act 506, the controller 404 calculates a correction value Z and a direction D for the correction value. As previously discussed, the yaw misalignment value X does not necessarily represent the actual degree of yaw misalignment because of various wind turbine geometries and characteristics. Thus, the controller 404 may apply a relatively small static correction value of, for example, 0.1 to 4 degrees. In other cases, the controller 404 may use empirical and/or heuristics to algorithmically translate the yaw misalignment value X into an actual degree of misalignment. For example, the controller 404 may utilize a lookup table that maps yaw misalignment values to actual degrees of error to derive a correction value Z. In other examples, the controller 404 performs interpolation using data from a calibration curve, or other similar data points, to derive a correction value Z. A calculated yaw misalignment value X that is greater than 60 degrees may indicate a clockwise correction of the nacelle may bring an associated rotor into alignment with oncoming wind. On the other hand, a calculated yaw misalignment value X that is less than 60 degrees may indicate a counter clockwise correction of the nacelle may bring an associated rotor into alignment with oncoming wind.

[0052] In act 508, the controller 404 performs a corrective action. In some cases, the corrective action includes the controller 404 providing a signal to the yaw control 406 to cause rotation of the nacelle based on the correction value Z and the calculated direction D. Alternatively, or in addition to the controller 404 providing a signal to the yaw control 406, the controller 404 may send a signal to a wind farm controller, e.g., via a TCP/IP network or other suitable network, for purposes of reporting the yaw misalignment value.

[0053] Turning to FIG. 6, one example process 600 for detecting a yaw misalignment value is shown in accordance with an aspect of the present disclosure. As discussed above, the process 600 may be performed during act 502 of the process 500. The process 600 may be performed by the controller 404, or other suitable component of the wind turbine control system 400.

[0054] In act 602, the controller 404 receives a plurality of shaft speed data samples representing a first period of time T from, for example, the shaft speed sensor 414. In act 604, the controller 404 receives a plurality of blade position samples also representing the first period of time T from the blade position sensor 412. For example, as shown in FIG. 7, the controller 404 receives shaft speed data samples 702 and blade position data samples 704 for the first period of time To ... To₊₁. The controller 404 may synchronize the shaft speed data samples 702 and blade position data samples 704 using, for example, a high-resolution hardware clock within the controller 404. The duration of the first period of time To ... To₊i may include, for example, 1 minute or more of data. In some cases, 1 to 10 minutes, or more, may be captured for processing purposes.

[0055] Returning to FIG. 6, and in act 606, the controller 404 splits (or separates) the plurality of shaft speed data samples 702 and the blade position data samples 704 into a first sequence of chunks 702-1...702-N (or portions) and a second sequence of chunks 704-1...704-N (or portions). For example, as shown in FIG. 8, the controller 404 may split each plurality of data samples into substantially equal one minute chunks, although other chunk sizes are also within the scope of this disclosure. In any event, each of the chunks 702-1...702-N and 704-1...704-N represent consecutive intervals of time over the period of time To ... To₊i. Thus, chunks 702- 1...702-N and 704-1...704-N may be time synchronized such that, for example, chunk 702-1 corresponds to chunk 704-2, chunk 702-2 corresponds to chunk 704-2, and so on.

[0056] Continuing with FIG. 6, and in act 608, the controller 404 applies time synchronous averaging (TSA) to each of the chunks 702-1...702-N and 704-1...704-N to transform each associated signal from the time domain into a 1 -revolution domain (or a N-revolution domain depending on a desired configuration). In particular, the controller 404 may use TSA to remove portions of each of the shaft speed data sample chunks, e.g., chunks 702-1 ... 702-N, not related to tower shadowing, e.g., noise, and also to reduce the number of data points down to a single revolution. In a raw form, the chunks 702-1...702-N may include components (noise) from other phenomena (e.g., drops in RPM due to wind gusts, yawing, and so on) that occur during operation of a wind turbine, and that are not necessarily related to tower shadow effect. Each chunk may also include a large number of data points representing potentially N number of revolutions of the shaft 422 relative to how fast each revolution is occurring (e.g., the RPM of the shaft) and the overall duration of time each chunk represents. The controller 404 may therefore use the shaft speed data samples 702 as a reference to determine how many revolutions occur over a given chunk. The controller 404 may then split/separate each chunk into a number of sub-chunks (or sub-portions), with each sub-chunk delineated along 1-revolution boundaries, or N-revolution boundaries depending on a desired configuration. For example, a 1 minute chunk having 10 revolutions of the rotor represented therein may be split into 10 sub-chunks. As each sub-chunk may include varying numbers of samples, the controller 404 may then resample the sub-chunks, e.g., by performing decimation and/or interpolation, such that each sub-chunk includes an equal number of samples. The controller 404 may then apply TSA to each of the sub-chunks to derive an associated composite waveform that represents the average shaft speed and blade position over 1 full revolution of a rotor, or over N full revolutions of the rotor depending on a desired configuration.

[0057] So, because the tower shadow effect is synchronous to the rotation of the shaft 422, and more particularly to the moment blades 452-1 to 452-3 pass by the front of tower 454, the controller 404 may use TSA to remove/filter out portions of signals within the shaft speed data sample chunks 702-1 ... 702-N not synchronous to the shaft. Thus, the controller 404 derives a relatively 'clean' signal for each chunk 702-1 ... 702-N by performing TSA on each chunk and its associated blade position data chunk. For example, the controller 404 may perform TSA on chunks 702-1 and 704-1 using the shaft speed signal 702 as a reference for the TSA in order to identify portions of the shaft speed data signal that are synchronous with the position of the blades. In similar fashion, the controller 404 may use TSA on each blade position chunk, e.g., chunks 704- 1 ... 704-N, to transform each associated signal from the time domain to the 1- revolution domain.

[0058] TSA algorithmically determines a mathematical average of chunks of time-domain data/signal where each chunk represents N-revolutions. In an embodiment, each chunk includes an equal number of data points which may be achieved through resampling and interpolating, as discussed above. The controller 404 may use the following equation to perform TSA:

(Equation 1)

where (TV) is the total number of revolutions, (t) is time, f(t) i is the chunk of time domain signal, e.g., after resampling and interpolation, Θ is the degree of revolution, e.g., ranges from 0 to 2*pi, and z(6>)is the TSA signal over the revolution domain signal. The result is a composite waveform that allows for cross-correlation to be performed during act 616, as discussed in greater detail below.

[0059] FIG. 9 illustrates how each of chunks 702- 1 ... 702-N and 704- 1 ... 704-N may be processed in accordance with act 608 to transform the chunks of the shaft speed data samples 702 and blade position data samples 704 from the time domain to a 1-revolution domain (e.g., having 3 cycles per revolution, or 3Hz). The composite waveforms depicted in FIG. 9 are not intended to depict actual resulting waveforms; rather, the waveforms are representational to more clearly show how the controller 404 may process chunks associated with each interval of time to derive a waveform in the 1-revolution domain.

[0060] Turning to FIG. 10, an example chunk signal 750 is shown for a chunk of the shaft speed data samples 702 after performing TSA in accordance with act 608 prior to band-pass filtering. As shown, the waveform 750 is a composite waveform in the 1-revolution domain representing the RPM of the shaft 422 for a given time interval. For example, the waveform 750 may correspond to the time interval associated with chunk 702-1. The waveform 750 plots tower shadow effect at positions 752-1, 752-2 and 752-3, which is consistent with each of the blades 452-1 to 452-3 passing by the tower 454. As will be discussed further below with regard to FIGs. 11-12, if the drop in RPM due to tower shadow effect occur at substantially each moment a blade passes in front of the tower 454, then the yaw misalignment is zero, or near- zero. On the other hand, tower shadow effect occurring before/after each moment the blade passes in front of the tower 454 indicates a yaw misalignment value greater than zero.

[0061] Returning to FIG. 6, and also in act 608, the controller 404 may also optionally apply band-pass filtering to each chunk of the shaft speed data samples 702-1...702-N and each chunk of the of the blade position data samples 704-1...704-N to further filter each signal. The controller 404 may use a band-pass filter with a cut-off frequency of 2-4Hz. This may cause some additional components, such as the frequency fluctuations shown in the waveform of FIG. 10, to be filtered out to produce a relatively "clean" waveform. FIGs. 11 and 12 show some such example composite waveforms after applying band-pass filtering.

[0062] In act 610, the controller 404 may scale each shaft speed composite waveform and blade position composite waveform for comparison purposes. For example, the amplitude of the signals may be scaled accordingly such that the controller 404 may compare each shaft speed composite waveform to a corresponding blade position composite waveform for a given time interval.

[0063] In act 612, the controller 404 calculates a phase shift value for each shaft speed chunk 702-1 ... 702-N relative to the composite waveforms corresponding to the blade position chunks 704-1 ... 704-N. For example, FIG. 11 plots a shaft speed composite waveform, e.g., associated with a chunk 702-N, relative to a blade position composite waveform, e.g., associated with chunk 704-N, in a 1-revolution domain. As shown, the RPM drop 812 of the shaft speed signal 702-N, e.g., due to tower shadowing, correspond to pulses (peaks) 810 of the blade position signal, thus indicating tower shadowing is occurring substantially directly in front of a wind turbine tower. As shown, the phase shift between the drops 812 of the shaft speed composite waveform and peaks 810 of the blade position composite waveform is substantially 60 degrees. Note that because the composite waveforms shown in FIG. 11 are represented in a 1-revolution domain, 0.0 to 1.0 of the revolution represents 360 total degrees of rotation by a rotor, such as rotor 424. Therefore, each oscillation of the composite waveforms represent 0-120 degrees, 120- 240 and 240-360 degrees, respectively, of the 1 full rotor revolution.

[0064] By way of contrast, FIG. 10 shows a shaft speed composite waveform with a drop in RPM, e.g., due to tower shadowing, that is occurring away from the tower as evidenced by the blade position composite waveform having pulses (peaks) 814 just after the RPM drop 816, e.g., due to tower shadowing. The phase shift between the shaft speed composite waveform and the blade position composite waveform is less than 60 degrees in this particular example, and thus indicates a yaw misalignment of a wind turbine is present.

[0065] Returning to FIG. 6, and continuing in act 612, the controller 404 may calculate a phase shift value for each chunk of shaft speed data samples and its corresponding chunk of blade position data samples, e.g., chunk 702-1 and its corresponding chunk 704-1. In some cases, the controller 404 may calculate each phase shift in the time domain using, for example, cross-correlation using the following equation:

Equation (2)

m=-∞

where (g) is a shaft speed composite waveform for a given chunk and (f) is a blade position composite waveform for an associated chunk. The result is a composite waveform which is a cross-correlation of (f) to (g). As discussed below, the peaks of the resulting composite waveform allow the controller 404 to identify where the signals potentially correlate the most, e.g., a phase shift. Other approaches to determining a phase shift value may be utilized such as in the frequency domain using fast Fourier transform (FFT), for example. The controller 404 may use any other suitable approaches to calculate a phase shift value, and the particular examples provided are not intended to be limiting.

[0066] In act 614, the controller 404 filters out, e.g., removes, any chunks from the shaft speed data samples and the blade position data samples that have an associated phase shift value that is outside of a predetermined window, e.g., outliers. For example, the controller 404 may filter out any chunk with an associated phase shift value less than a first value and greater than a second value. For example, the controller 404 may filter out chunks with an associated phase shift value of less than 30 degrees and greater than 70 degrees as phase shift measurements outside of the 30- to 70-degree window may be considered erroneous and caused by external events unrelated to tower shadowing effect such as, for example, wind shear, yaw movement, and so on.

[0067] In act 616, the controller 404 determines an overall average phase shift value using, for example, mean circular quantities. The controller 404 may determine an overall average phase shift using other approaches such as summing each phase shift value associated with the chunks and diving by the total number of phase shift values represented. For instance, if ten phase shift values remain after act 614, then the controller 404 may simply sum each of the ten phase shift values and then divide by 10. Accordingly, other approaches to determining in overall average phase shift in act 616 may be utilized and are also within the scope this disclosure.

[0068] In act 618, the controller 404 outputs a yaw misalignment value (or yaw error value) in degrees based on the overall average phase shift value. This yaw misalignment value thus indicates where tower shadow effect is occurring relative to the position of the blades. The yaw misalignment value may also include an indication of a particular direction in which the nacelle may be rotated to correct alignment. The particular direction identified for correction may be based on, in a general sense, whether there is a "delay" of tower shadow effect (e.g., tower shadow happens after the 6 o clock position) or as "rushed" (tower shadow occurs before the 6Ό clock position). In the example case of FIG. 12 tower shadow effect is "rushed" and therefore a counter-clockwise rotation may be used to align the wind turbine with oncoming wind, assuming a clockwise rotation of the blades. On the other hand, a composite waveform that shows a "delay" of tower shadowing effect may indicate a clockwise rotation for correction. For example, an average overall phase shift less than 60 degrees may indicate that a counterclockwise rotation may correct alignment as tower shadow effect may be "rushed" and occurring to the left of the tower 454 shown in FIG 4B, assuming clockwise rotation of the blades. Other reference points for measuring phase shift between the shaft speed composite waveform and blade position composite waveform may be utilized to determine rotation direction for correction of the nacelle/rotor, and the provided examples are not intended to be limiting

[0069] FIG. 13 further illustrates the yaw misalignment value in a polar plot in accordance with an aspect of the present disclosure. As shown, the difference between the peak of lobe 801 associated with the RPM of the shaft 422 and the peak of lobe 802 associated with the blade position are separated by a distance of 60 degrees, and thus indicates a yaw misalignment value of zero or near-zero. In contrast, the polar plot of FIG. 14 shows a yaw misalignment value greater than zero. This is because, as shown, the difference between the peak of the lobe 803 associated with the RPM of shaft 422 and the peak of the lobe 804 associated with the blade position is separated by a distance of less than 60 degrees. As such, the difference from 60 degrees may be considered the yaw misalignment value. For example, a 40-degree difference may indicate a yaw misalignment value of 20 degrees. The yaw misalignment value thus indicates whether the blades are passing through the tower shadow, or whether the tower shadow is occurring at a position away from where the blades pass in front of the tower.

[0070] The wind turbine control system 400 may thus advantageously provide a number of options when determining whether yaw misalignment has occurred, and moreover, multiple options for validating data from a wind sensor and/or other devices that detect yaw misalignment.

[0071] Moreover, a wind turbine control system consistent with the present disclosure may provide an indication of potential loss of power production to be sent to a wind farm operator (e.g., via a network coupled to a wind turbine), or owner, and also an indication that larger stresses are being applied to the wind turbine. The wind turbine control system 400 may send this indication to a system that monitors the status of the yaw alignment of the wind turbine relative to the incoming wind direction, with this information being aggregated with information from other wind turbines in the same farm for monitoring purposes.

[0072] In any event, the wind turbine control system 400 provides numerous advantageous. For example, stand-alone or part of Condition Monitoring System (CMS) may alert wind farm operators if a turbine is out of alignment (yaw misalignment). This alert may be based on the measured yaw misalignment exceeding a predetermined threshold. Such an alert allows adjustment of the wind turbine, e.g., adjustment to a wind sensor and/or rotation of the nacelle, to ensure the wind turbine is properly aligned with the wind in order to generate power optimally. By extension, corrective adjustments also create less wear and tear on the machine thus extending time between failures/machine lifetime.

[0073] Embodiments of the methods and processes described herein for yaw misalignment detection and correction, e.g., the processes 500 and 600 of FIGs. 5 and 6, respectively, may be implemented using a processor and/or other programmable device. To that end, the methods described herein may be implemented on a tangible, computer readable storage medium having instructions stored thereon that when executed by one or more processors perform the methods. Thus, for example, the transmitter and/or receiver may include a storage medium (not shown) to store instructions (in, for example, firmware or software) to perform the operations described herein. The storage medium may include any type of non-transitory tangible medium, for example, any type of disk including floppy disks, optical disks, compact disk read-only memories (CD-ROMs), compact disk re-writables (CD-RWs), and magneto-optical disks, semiconductor devices such as read-only memories (ROMs), random access memories (RAMs) such as dynamic and static RAMs, erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), flash memories, magnetic or optical cards, or any type of media suitable for storing electronic instructions.

[0074] Any block diagrams herein represent conceptual views of illustrative circuitry embodying the principles of the disclosure. Similarly, any flow charts, flow diagrams, state transition diagrams, pseudocode, and the like represent various processes which may be substantially represented in computer readable medium and so executed by a computer or processor, whether or not such computer or processor is explicitly shown.

[0075] The functions of the various elements shown in the figures, including any functional blocks, may be provided through the use of dedicated hardware as well as hardware capable of executing software in association with appropriate software. When provided by a processor, the functions may be provided by a single dedicated processor, by a single shared processor, or by a plurality of individual processors, some of which may be shared. Moreover, explicit use of the term "processor" or "controller" should not be construed to refer exclusively to hardware capable of executing software, and may implicitly include, without limitation, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for storing software, random access memory (RAM), and non-volatile storage. Other hardware, conventional and/or custom, may also be included.

[0076] As used in any embodiment herein, "circuit" or "circuitry" may comprise, for example, singly or in any combination, hardwired circuitry, programmable circuitry, state machine circuitry, and/or firmware that stores instructions executed by programmable circuitry. In at least one embodiment, the transmitter and receiver may comprise one or more integrated circuits. An "integrated circuit" may be a digital, analog or mixed-signal semiconductor device and/or microelectronic device, such as, for example, but not limited to, a semiconductor integrated circuit chip. The term "coupled" as used herein refers to any connection, coupling, link or the like by which signals carried by one system element are imparted to the "coupled" element. Such "coupled" devices, or signals and devices, are not necessarily directly connected to one another and may be separated by intermediate components or devices that may manipulate or modify such signals. As used herein, use of the term "nominal" or "nominally" when referring to an amount means a designated or theoretical amount that may vary from the actual amount. Further Example Embodiments

[0077] In accordance with an aspect of the present disclosure a wind turbine control system is disclosed. The wind turbine control system comprising a controller configured to receive a first signal representing shaft speed of a shaft of a wind turbine over a first period of time T and a second signal representing blade position over the first period of time T, and to provide a yaw misalignment output value based on the received first and second signals, the yaw misalignment output value indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

[0078] In the wind turbine control system the controller may further be configured to determine the yaw misalignment output value based in part on identifying a position of tower shadow effect relative to blade position over the first period of time T using at least a portion of the received first and second signals.

[0079] In the wind turbine control system the controller may further be configured to provide the yaw misalignment output value based at least in part on applying time synchronous averaging (TSA) to at least a first portion of the first signal to transform the first portion from a time domain to a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

[0080] In the wind turbine control system the controller may be further configured to apply TSA to a least a first portion of the second signal to transform the first portion of the second signal from a time domain to the N-revolution domain.

[0081] In the wind turbine control system the controller may further be configured to determine the yaw misalignment output value based in part on calculating a phase shift between the first portion of the first signal and the first portion of the second signal in the N-revolution domain.

[0082] In the wind turbine control system the controller may further be configured to determine the yaw misalignment output value based on a difference D between the calculated phase shift and a predetermined phase shift of 60 degrees.

[0083] The wind turbine control system may further comprise the wind turbine, and wherein the wind turbine is a horizontal axis, three-blade wind turbine having the rotor forward of a nacelle, and wherein the yaw misalignment output value indicates whether a horizontal axis of the rotor of the wind turbine is substantially in parallel with a directional component of oncoming wind.

[0084] The wind turbine control system may further comprise a blade position sensor disposed along a tower of the wind turbine, the blade position sensor configured to detect a blade passing in front of the tower and provide a pulse, and wherein the second signal representing blade position is based in part on the pulse from the blade position sensor.

[0085] The wind turbine control system may further comprise a shaft speed sensor coupled to a shaft of the wind turbine and configured to provide an output signal proportional to a speed at which the shaft rotates, and wherein the first signal representing shaft speed for the shaft is based in part on the output signal of the shaft speed sensor.

[0086] In the wind turbine control system the controller may be further configured to determine if the yaw misalignment output value exceeds a maximum threshold value, and performing a corrective action in response to the yaw misalignment output value exceeding a maximum threshold value. Further, the corrective action may comprise at least one of sending an indication of the yaw misalignment output value to a remote computing device and/or sending a signal to a yaw control arrangement to cause the yaw control arrangement to rotate the rotor of the wind turbine in a direction that reduces yaw misalignment with oncoming wind.

[0087] In accordance with another aspect of the present disclosure a method is disclosed. The method comprising receiving a plurality of shaft speed data samples for a shaft of a wind turbine over a first period of time T, receiving a plurality of blade position samples representing blade position for the wind turbine over the first period of time T, generating a first signal based on at least a first portion of the plurality of shaft speed data samples, generating a second signal based on at least a first portion of the plurality of blade position samples, and calculating an average phase shift value based at least in part on a phase shift between the first signal and the second signal, the average phase shift indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

[0088] The method may further comprise outputting a yaw misalignment value equal to a delta between the calculated average phase shift value and a predetermined phase shift value of about 60 degrees.

[0089] The method may further comprise filtering the first signal to remove noise related to events other than tower shadow effect. [0090] The method may further comprise transforming the first signal from a time domain into a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

[0091] The method may further comprise transforming the second signal from a time domain into the N-revolution domain.

[0092] The method may further comprise calculating the average phase shift value between the first signal and the second signal further comprises calculating a phase shift value between peaks of the first signal and the second signal in the N-revolution domain using a cross- correlation approach in a time domain or a fast Fourier transform (FFT) in a frequency domain.

[0093] In accordance with another aspect of the present disclosure a method is disclosed. The method comprising receiving a first signal comprising shaft speed samples for a shaft of a wind turbine over a first period of time T, receiving a second signal comprising blade position samples representing blade position of the wind turbine over the first period of time T, splitting the first signal and second signal into a first and a second sequence of chunks, respectively, wherein the first signal and the second signal are time- synchronized such that each chunk of the first and second signals represent consecutive intervals of time over the first period of time T, performing time synchronous averaging (TSA) to transform each chunk of the first and second sequences from a time domain to a N-revolution domain, the N-revolution domain representing at least one full revolution of a rotor of the wind turbine, and calculating a yaw misalignment output value based in part on an average phase shift between each chunk of the first sequence of chunks and a corresponding chunk of the second sequence of chunks.

[0094] In the method, performing time synchronous averaging (TSA) may further comprise filtering each chunk of the first sequence of chunks to remove samples associated with events unrelated to tower shadowing effect.

[0095] In the method, calculating the yaw misalignment output value may further comprise comparing the average phase shift to a predetermined phase shift value of 60 degrees to determine a delta D, and wherein when the delta D is greater than zero indicates a yaw error of the rotor of the wind turbine relative to oncoming wind.

[0096] In yet another aspect of the present disclosure a wind turbine is disclosed. The wind turbine comprising a tower, a nacelle including a rotor with a plurality of blades coupled thereto, the nacelle coupled to the tower and configured to rotate along a yaw axis to align the rotor relative to oncoming wind, a shaft speed sensor coupled to a shaft of the wind turbine and to output a first signal representing shaft speed, a blade position sensor disposed adjacent the plurality of blades and configured to output a second signal representing blade position, and a controller configured to receive the first signal representing shaft speed over a first period of time T and a second signal representing blade position over the first period of time T, and to provide a yaw misalignment output value based in part on the received first and second signals, the yaw misalignment output value indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

[0097] In the wind turbine, the controller may be further configured to determine the yaw misalignment output value based in part on identifying a position of tower shadow effect relative to blade position over the first period of time T using at least a portion of the received first and second signals.

[0098] In the wind turbine, the controller may be further configured to provide the yaw misalignment output value based at least in part on applying time synchronous averaging (TSA) to at least a first portion of the first signal to transform the first portion from a time domain to a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

[0099] In the wind turbine, the controller may be further configured to apply TSA to a least a first portion of the second signal to transform the first portion of the second signal from a time domain to the N-revolution domain.

[00100] In the wind turbine, the controller may be further configured to determine the yaw misalignment output value based in part on calculating a phase shift between the first portion of the first signal and the first portion of the second signal in the N-revolution domain.

[00101] In the wind turbine, the controller may be further configured to determine the yaw misalignment output value based on a difference D between the calculated phase shift and a predetermined phase shift of 60 degrees.

[00102] The wind turbine may be implemented as a horizontal axis, three-blade wind turbine having the rotor forward of the nacelle, and wherein the yaw misalignment output value indicates whether a horizontal axis of the rotor of the wind turbine is substantially in parallel with a directional component of oncoming wind.

[00103] In the wind turbine, the blade position sensor may be disposed along the tower adjacent the plurality of blades, and wherein the second signal representing blade position is based in part on a pulse from the blade position sensor as each blade of the plurality of blades pass in front of the tower.

[00104] In the wind turbine, the first signal representing shaft speed may comprise N pulses for each revolution of the rotor, and wherein the controller is further configured to convert the first signal representing shaft speed to a rotations-per-minute (RPM) signal.

[00105] In the wind turbine, the controller may be further configured to determine if the yaw misalignment output value exceeds a maximum threshold value, and performing a corrective action in response to the yaw misalignment output value exceeding a maximum threshold value.

[00106] In the wind turbine, the corrective action may comprise at least one of sending an indication of the yaw misalignment output value to a remote computing device and/or sending a signal to a yaw control arrangement to cause the yaw control arrangement to rotate the nacelle of the wind turbine in a direction that reduces yaw misalignment relative to oncoming wind.

[00107] The terms and expressions which have been employed herein are used as terms of description and not of limitation, and there is no intention, in the use of such terms and expressions, of excluding any equivalents of the features shown and described (or portions thereof), and it is recognized that various modifications are possible within the scope of the claims. Also features of any embodiment described herein may be combined or substituted for features of any other embodiment described herein.

Claims

What is claimed:

1. A wind turbine control system comprising:

a controller configured to receive a first signal representing shaft speed of a shaft of a wind turbine over a first period of time T and a second signal representing blade position over the first period of time T, and to provide a yaw misalignment output value based on the received first and second signals, the yaw misalignment output value indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

2. The system of claim 1, wherein the controller is further configured to determine the yaw misalignment output value based in part on identifying a position of tower shadow effect relative to blade position over the first period of time T using at least a portion of the received first and second signals.

3. The system of claim 1, wherein the controller is further configured to provide the yaw misalignment output value based at least in part on applying time synchronous averaging (TSA) to at least a first portion of the first signal to transform the first portion from a time domain to a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

4. The system of claim 3, wherein the controller is further configured to apply TSA to a least a first portion of the second signal to transform the first portion of the second signal from a time domain to the N-revolution domain.

5. The system of claim 4, wherein the controller is further configured to determine the yaw misalignment output value based in part on calculating a phase shift between the first portion of the first signal and the first portion of the second signal in the N-revolution domain.

6. The system of claim 5, wherein the controller is configured to determine the yaw misalignment output value based on a difference D between the calculated phase shift and a predetermined phase shift of 60 degrees.

7. The system of claim 1, further comprising the wind turbine, and wherein the wind turbine is a horizontal axis, three-blade wind turbine having the rotor forward of a nacelle, and wherein the yaw misalignment output value indicates whether a horizontal axis of the rotor of the wind turbine is substantially in parallel with a directional component of oncoming wind.

8. The system of claim 1, further comprising a blade position sensor disposed along a tower of the wind turbine, the blade position sensor configured to detect a blade passing in front of the tower and provide a pulse, and wherein the second signal representing blade position is based in part on the pulse from the blade position sensor.

9. The system of claim 1, further comprising a shaft speed sensor coupled to a shaft of the wind turbine and configured to provide an output signal proportional to a speed at which the shaft rotates, and wherein the first signal representing shaft speed for the shaft is based in part on the output signal of the shaft speed sensor.

10. The system of claim 1, wherein the controller is further configured to:

determine if the yaw misalignment output value exceeds a maximum threshold value; and performing a corrective action in response to the yaw misalignment output value exceeding a maximum threshold value.

11. The system of claim 10, wherein the corrective action comprises at least one of sending an indication of the yaw misalignment output value to a remote computing device and/or sending a signal to a yaw control arrangement to cause the yaw control arrangement to rotate the rotor of the wind turbine in a direction that reduces yaw misalignment with oncoming wind.

12. A method, the method comprising:

receiving a plurality of shaft speed data samples for a shaft of a wind turbine over a first period of time T;

receiving a plurality of blade position samples representing blade position for the wind turbine over the first period of time T; generating a first signal based on at least a first portion of the plurality of shaft speed data samples;

generating a second signal based on at least a first portion of the plurality of blade position samples; and

calculating an average phase shift value based at least in part on a phase shift between the first signal and the second signal, the average phase shift indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

13. The method of claim 12, further comprising outputting a yaw misalignment value equal to a delta between the calculated average phase shift value and a predetermined phase shift value of about 60 degrees.

14. The method of claim 12, further comprising filtering the first signal to remove noise related to events other than tower shadow effect.

15. The method of claim 12, further comprising transforming the first signal from a time domain into a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

16. The method of claim 15, further comprising transforming the second signal from a time domain into the N-revolution domain.

17. The method of claim 16, wherein calculating the average phase shift value between the first signal and the second signal further comprises calculating a phase shift value between peaks of the first signal and the second signal in the N-revolution domain using a cross- correlation approach in a time domain or a fast Fourier transform (FFT) in a frequency domain.

18. A method, the method comprising:

receiving a first signal comprising shaft speed samples for a shaft of a wind turbine over a first period of time T;

receiving a second signal comprising blade position samples representing blade position of the wind turbine over the first period of time T;

splitting the first signal and second signal into a first and a second sequence of chunks, respectively, wherein the first signal and the second signal are time- synchronized such that each chunk of the first and second signals represent consecutive intervals of time over the first period of time T;

performing time synchronous averaging (TSA) to transform each chunk of the first and second sequences from a time domain to a N-revolution domain, the N-revolution domain representing at least one full revolution of a rotor of the wind turbine; and calculating a yaw misalignment output value based in part on an average phase shift between each chunk of the first sequence of chunks and a corresponding chunk of the second sequence of chunks.

19. The method of claim 18, wherein performing time synchronous averaging (TSA) further comprises filtering each chunk of the first sequence of chunks to remove samples associated with events unrelated to tower shadowing effect.

20. The method of claim 18, wherein calculating the yaw misalignment output value further comprises comparing the average phase shift to a predetermined phase shift value of 60 degrees to determine a delta D, and wherein when the delta D is greater than zero indicates a yaw error of the rotor of the wind turbine relative to oncoming wind.

21. A wind turbine comprising:

a tower;

a nacelle including a rotor with a plurality of blades coupled thereto, the nacelle coupled to the tower and configured to rotate along a yaw axis to align the rotor relative to oncoming wind;

a shaft speed sensor coupled to a shaft of the wind turbine and to output a first signal representing shaft speed;

a blade position sensor disposed adjacent the plurality of blades and configured to output a second signal representing blade position; and

a controller configured to receive the first signal representing shaft speed over a first period of time T and a second signal representing blade position over the first period of time T, and to provide a yaw misalignment output value based in part on the received first and second signals, the yaw misalignment output value indicating whether a rotor of the wind turbine has a yaw error relative to oncoming wind.

22. The wind turbine of claim 21, wherein the controller is further configured to determine the yaw misalignment output value based in part on identifying a position of tower shadow effect relative to blade position over the first period of time T using at least a portion of the received first and second signals.

23. The wind turbine of claim 21, wherein the controller is further configured to provide the yaw misalignment output value based at least in part on applying time synchronous averaging (TSA) to at least a first portion of the first signal to transform the first portion from a time domain to a N-revolution domain, the N-revolution domain representing at least one complete revolution of the rotor.

24. The wind turbine of claim 23, wherein the controller is further configured to apply TSA to a least a first portion of the second signal to transform the first portion of the second signal from a time domain to the N-revolution domain.

25. The wind turbine of claim 24, wherein the controller is further configured to determine the yaw misalignment output value based in part on calculating a phase shift between the first portion of the first signal and the first portion of the second signal in the N-revolution domain.

26. The wind turbine of claim 25, wherein the controller is configured to determine the yaw misalignment output value based on a difference D between the calculated phase shift and a predetermined phase shift of 60 degrees.

27. The wind turbine of claim 21, wherein wind turbine is implemented as a horizontal axis, three-blade wind turbine having the rotor forward of the nacelle, and wherein the yaw misalignment output value indicates whether a horizontal axis of the rotor of the wind turbine is substantially in parallel with a directional component of oncoming wind.

28. The wind turbine of claim 21, wherein the blade position sensor is disposed along the tower adjacent the plurality of blades, and wherein the second signal representing blade position is based in part on a pulse from the blade position sensor as each blade of the plurality of blades pass in front of the tower.

29. The wind turbine of claim 21, wherein first signal representing shaft speed comprises N pulses for each revolution of the rotor, and wherein the controller is further configured to convert the first signal representing shaft speed to a rotations-per-minute (RPM) signal.

30. The wind turbine of claim 21, wherein the controller is further configured to: determine if the yaw misalignment output value exceeds a maximum threshold value; and performing a corrective action in response to the yaw misalignment output value exceeding a maximum threshold value.

31. The wind turbine of claim 30, wherein the corrective action comprises at least one of sending an indication of the yaw misalignment output value to a remote computing device and/or sending a signal to a yaw control arrangement to cause the yaw control arrangement to rotate the nacelle of the wind turbine in a direction that reduces yaw misalignment relative to oncoming wind.