WO2010061255A2

WO2010061255A2 - Active blade pitch control for reduction of wind turbine noise or loads

Info

Publication number: WO2010061255A2
Application number: PCT/IB2009/007262
Authority: WO
Inventors: Khanh Quoc Nguyen
Original assignee: Clipper Windpower, Inc.
Priority date: 2008-11-01
Filing date: 2009-10-29
Publication date: 2010-06-03
Also published as: WO2010061255A3

Abstract

A method of blade pitch control actuation, which reduces rotor noises during operation of a wind turbine. A microphone or a blade mounted pressure sensor provides a noise level feedback signal to a blade pitch controller. The blade pitch controller uses the feedback signal to adjust the blade pitch to reduce the measured noise decibel level to an acceptable decibel level.

Description

ACTIVE BLADE PITCH CONTROL FOR REDUCTION OF WIND TURBINE

NOISE OR LOADS

Field of the Invention

The invention relates to fluid-flow turbines, such as wind turbines and more particularly to an apparatus and method to reduce wind turbine operating noise.

DESCRIPTION OF THE PRIOR ART

The development of practical, wind-powered generating systems creates problems, which are unique and not encountered in the development of conventional power generating systems. One such problem is the amount of noise emanating from the turning rotor blades of an operating wind turbine.

In the past, wind turbines have been operated at constant speed. The torque produced by the blades and main shaft determines the power delivered by such a wind turbine. The turbine is typically controlled by a power command signal, which is fed to a turbine blade pitch angle servo motor control. This servo controls the pitch of the rotor blades and therefore the power output of the wind turbine .

To alleviate the problems of power surges and mechanical loads and to increase performance compared to constant speed wind turbines, the wind power industry has been moving towards the use of variable speed wind turbines . A variable speed wind turbine is described in US Patent 7,042,110.

Large modern wind turbines have rotor diameters of up to 126 meters or larger with towers at a height to accommodate them. Operational noise is likely to be more pronounced in the case of larger rotors . Loading across turbine rotor blades may vary because of wind shear, or differences in wind speed at different heights of the rotor plane. With positive wind shear, wind at the highest altitude reached by the rotor blades will have a higher velocity and the least wind speed will be at the lowest point of the rotor. Negative wind shear is also possible. Loading also varies horizontally across the rotor. Thus, at any point in time, each blade may have a different load due to wind depending upon its real-time rotational position. These loads and spatially varying wind environment contribute to noise generated by the operation of a wind turbine . Turbine control systems can reduce loads, reduce motor torque, and provide better control. Control systems range from the relatively simple proportional, integral derivative (PID) collective blade controllers to independent blade state space controllers . These controls do not address the issue of noise emanating from an operating wind turbine, which, if it exceeds an acceptable level, will severely limit the environment in which the turbines will be allowed to operate . The blade pitch is used to control aerodynamic angle of attack. If all else, is equal, then a greater value of angle of attack can typically cause greater blade noise due to trailing edge separation, thicker boundary layers, and increased small-scale turbulence.

Current rotor designs use low tip speed to limit rotor noise, which incurs a performance penalty. Passive methods using blade tip shapes achieve limited noise reduction (4 dB) . The following patents illustrate some of the work in measuring turbine noise that has been undertaken.

Wobben Patent 6,966,754 relates to a system for monitoring the operation of wind turbines, wherein image and acoustic monitoring is performed in order to improve the maintenance, safety and efficiency of the wind turbine. Acoustic and optical sensors are employed to monitor the operation of the wind turbine . The optical sensor images a particular location in the nacelle housing when the acoustic sensor detects a sound emanating from that location in the nacelle housing. When a deviation between an operating acoustic spectrum and a reference acoustic spectrum exceeds a threshold, the component that generated the operating acoustic spectrum is imaged.

In Wu Patent 7,330,396, an algorithm diagnoses a noise source strength distribution on a wind turbine installation based on acoustic pressures measured in the far field. This enables one to acquire a quick estimate of the acoustic pressure at locations that are off limit to traditional measurement microphones or blade mounted pressure sensors . The noise source is modeled in terms of a plurality of virtual spherical wave sources distributed on an auxiliary surface conformal to a source boundary from the inside, but not on the source boundary itself . Sound is then measured at a plurality of measurement points external to the source boundary. The sound field is reconstructed on the source boundary surface itself. What is needed is a method to measure the rotor noise of a turbine, either directly or indirectly, and then reduce it. In Wobben's patent, the method employs image and acoustic monitoring inside a turbine, nacelle, tower, or hub and processes the information to detect faults or anomalies during turbine operations. While Wobben's method employs noise measurements, Wobben's objective is to detect faults in an open-loop sense, not to reduce noise. The faults are detected in Wobben's method by comparing the measured noise to a stored reference signal. It is desirable to provide noise reduction as a component of a turbine control system. It is also desirable to process measured noise into a metric (decibel scale) and use it as a feedback signal in a rotor blade control system to reduce noise to an acceptable decibel level .

SUMMARY OF THE INVENTION Briefly, the present invention relates to an apparatus and method of controlling a wind turbine having a number of rotor blades to affect noise reduction. The noise may be caused by rotation of the rotor blades during turbine operation. The wind turbine uses a pitch command to control pitch of the rotor blades of the wind turbine. The method first determines and stores a nominal metric (decibel level) that represents an acceptable noise level set point.

The control uses one more acoustic devices, such as microphones or blade mounted pressure sensors, to sense instantaneous noise of the wind turbine resulting in a noise signal. The control uses the noise signal to calculate a blade pitch modulation needed to compensate for the instantaneous noise, normally reducing blade angle of attack to bring down the nominal noise level . The calculated blade pitch modulation is combined with the nominal pitch command determined to control, for example, the rotor rpm. Finally, the combination is used to control pitch of the rotor blades in order to reduce the instantaneous noise of the wind turbine . The invention therefore uses an output of conventional control systems and adds compensation for instantaneous conditions deviating from nominal or mean conditions by modulation of the control signals. However, the basic control mechanism providing the basic pitch command is not affected, since only the output signal is modulated. Therefore, the system can smoothly and stably return to the unmodulated control values if deviations from the nominal values are not registered. Because this method uses active blade pitch control to reduce noise, it has a potential for large noise reduction. In the present invention, the method measures the rotor noise of a turbine, either directly or indirectly, and then reduces it. In the above-described Wobben patent the method employs image and acoustic monitoring inside a turbine, nacelle, tower, or hub, and processes the information to detect faults or anomalies during turbine operations. While the present invention employs noise measurements, Wobben ¹S method only detect faults, it does not reduce noise.

Furthermore, the methods differ in how noise measurements are processed. Wobben 's method compares the measured noise to a reference signal, while the present invention processes the measured noise into a metric (on a decibel scale) and uses it as a feedback signal in a control system.

In the above-described Wu patent, the method uses acoustic arrays to measure the far field noise and employs signal processing to identify the noise source. It is a measurement technique involving no controls . The method of the present invention also measures far field noise, but it processes the measurement and uses it as a feedback signal for noise control. Thus, the present invention focuses on noise reduction and not on identifying the noise source. In addition, the noise processing technique used in the method of this invention differs from and is simpler than the one described in Wu 's patent.

The low noise technology of the invention enables quiet turbine operations for public acceptance and enables high tip speed rotor designs for enhanced power production. The invention has the advantage that existing turbines can be retrofitted with a noise reduction package; including a noise reduction apparatus embedded in the turbine controller and microphones or blade-mounted pressure sensors. The way in which the pitch modulation is chosen or calculated on basis of the noise information may vary. In a preferred embodiment of the invention the pitch modulation generally moves towards "feather", or reduced angle of attack, on an individual blade that is sensed to be at a higher noise level (or to have unsteady pressure measurements) at a certain azimuth position. At other azimuth positions, the blade is modulated to its nominal position.

However, in other embodiments a solution could also be devised where acoustic sensing devices could be used to determine an individual blade pitch strategy that could be used for average or common wind conditions or typical wind shear values. In this case, the pitch strategy could be open loop as a table lookup and could reduce noise of the turbine in common wind regimes .

In the case of blade-mounted pressure sensors, individual blades could even be modulated to higher angles of attack than nominal if the pressure sensors indicate less pressure than optimum for performance . It is important that the invention may be used not only to reduce noise of a wind turbine but also for loads reduction. When implemented as described for reducing noise, a loads reduction is an almost certain side effect. However, even when the turbine noise is not an issue of major importance (e.g. devices installed offshore) the invention may be used to reduce loads. In this connection, noise is considered as an indication for loads on the blades. The analysis of the kind or intensity of noise is detected by noise sensors. The relation between the detected noise and the loads may be based on empirical values which may be used in order to derive a load condition in relation to detected noise. Therefore, according to special embodiments of the invention the noise reduction may even be a side effect of the load reduction for which the noise is detected as an indicator.

In this connection all characteristics of the detected noise signals may be analyzed in order to deduce a load condition. Depending on the noise analysis there may even be different modulations, depending on the results of the analysis .

BRIEF DESCRIPTION OF THE DRAWINGS The invention and its mode of operation will be more fully understood from the following detailed description when taken with the appended drawings in which:

FIGURE 1 is a block diagram of a variable speed wind turbine in an example of a positive wind shear in which the present invention is embodied;

FIGURE 2 is a block diagram of a general noise compensator in parallel with a conventional collective controller;

FIGURE 3 is a top view of wind turbine illustrating noise reduction system using field microphones; FIGURE 4 is a side view of a wind turbine illustrating noise reduction using proximity microphones located on a wind turbine tower;

FIGURE 5 is a block diagram of a noise processor (PID controller) , which processes measured noise signals into a metric, which is used to compute a blade pitch schedule for noise reduction; and

FIGURE 6 is a flow chart of the method of noise reduction in a wind turbine in accordance with the invention.

DETAILED DESCRIPTION OF THE INVENTION

Refer to FIGURE 1, which is a block diagram of a variable- speed wind turbine apparatus in accordance with the present invention. The wind power-generating device includes a turbine which turns one or more electric generators and has a nacelle 100, which is mounted atop a tower structure 102 anchored to the ground 104 or another type of foundation. The nacelle 100 rests on a yaw platform 101 and is free to rotate in the horizontal plane about a yaw pivot 106 and is maintained in the path of prevailing wind current 108, 110.

The turbine has a rotor with variable pitch blades, 112, 114, attached to a rotor hub 118. The blades rotate around axis 122 in response to wind current, 108, 110. Each of the blades may have a blade base section and a blade extension section such that the rotor is variable in length to provide a variable diameter rotor. As described in US patent 6,726,439, the rotor diameter may be controlled to fully extend the rotor at low flow velocity and to retract the rotor, as flow velocity increases such that the loads delivered by or exerted upon the rotor do not exceed set limits. The nacelle 100 is held on the tower structure in the path of the wind current such that the nacelle is held in place horizontally in approximate alignment with the wind current. The electric generator is driven by the turbine to produce electricity and is connected to power carrying cables inter-connecting to other units and/or to a power grid.

Vertical wind shear is the change in wind speed with height above ground. Positive wind shear is illustrated in FIGURE 1 by the greater wind speed arrow 108 and the lower wind speed arrow 110 closer to ground. Among other influences, vertical wind shear is caused by height-dependent friction with the ground surface 104. The higher the height is above ground, 108, the less the effect of surface friction 104 and the higher the wind speed. Generally, the closer the height to ground, 110, the more the effect surface friction 104 has and the lower the wind speed. Because of wind shear, the wind turbine should control the pitch of the rotor blades independently of each other .

The apparatus shown in FIGURE 1 compensates for noise in a wind turbine 100. The pitch of the blades is controlled in a conventional manner by a command component, conventional pitch command logic 148, which uses generator RPM 138, which is an output of shaft rotation sensor 132, to develop a nominal rotor blade pitch command signal 154. A storage 144 contains a stored value a nominal noise decibel level and the blade pitch limits for noise compensation. The pitch limits are used to prevent blade loads exceeding their fatigue limits. These limits depend on the properties of the blades and are determined analytically and experimentally. Noise sensor microphones or blade mounted pressure sensors have an output, which represents noise signals 147. The microphones 160, 161, 162, 163 are located on the ground 104 and/or microphones 164,165, 166, 167 are attached to the tower 102. The microphones should surround the tower to triangulate blade noise source and to also account for the changing yaw position of the turbine in response to changes in wind direction. A yaw angle sensor 124 is provided to generate a yaw angle 143.

The blade-mounted pressure sensors are located on the rotor blades .

Conversion logic 146 is connected to the noise signals 147, to the blade rotational positions 140, and to the blade pitch sensors 141, which results in a calculated pitch modulation command 152. Combining logic connected to the calculated blade pitch modulation command 152 and to the pitch command 154, provides a combined blade pitch command 156 capable of commanding pitch of the rotor blades, which includes compensation for instantaneous noise levels of the wind turbine . Conversion logic 146 is also connected with the blade rotational position signal 140, which is an output of blade position sensor 134, and yaw angle 143, which is an output of yaw angle sensor 124. The yaw angle 143 is used to activate the appropriate microphones around the circumference of the tower, depending upon the circumferential position of the rotor blades .

Refer to FIGURE 2, which is a block diagram of a general noise compensator in parallel with a conventional collective controller. The apparatus shown in FIGURE 2 compensates for noise imbalance in a wind turbine 200. The pitch of the blades is controlled in a conventional manner by a command component, conventional collective controller 248, which uses actual generator RPM 238 fed back to and combined with a desired RPM 239 to develop a collective pitch command signal 254. Conversion logic (not shown) connected to a noise signal provides an output for each of the blades #1, #2 and #3, which is a calculated pitch modulation command 252. Combining logic 250 connected to the calculated noise blade pitch modulation command 252 and to the collective pitch command 254, provides a combined blade pitch command 256 capable of commanding pitch of the rotor blades, which includes compensation for high instantaneous noise levels of the wind turbine 200. The collective controller 248 therefore provides a control signal used as basis for controlling each of the blades #1, #2 and #3. However, the combining logic 250 outputs individual blade commands by modulating the collective command signal 254 by individual blade pitch modulation command 252. The conventional collective controller is a PID, PI, state space or any other type of control system. A three-bladed turbine is illustrated, however any number of blades may be used. A collective controller with pitch as its only output is illustrated, however generator torque and any other output is possible. A collective controller with generator rpm as its only input is illustrated, however, actual blade pitch and any other inputs are within the scope of this invention. The preferred apparatus for measurement of noise is illustrated in FIGURE 3. The noise is measured by microphones located about a circle surrounding the wind turbine tower at equal distances from each other and from the turbine tower. The microphones may be located on the ground as shown in FIGURE 3 or on the tower as shown in FIGURE 4.

Refer to FIGURE 3 , which is a top view of wind turbine illustrating noise reduction system using field microphones. Four microphones 160,161,162,163 are located at ground level, circumferential around and equidistant from each other and from the wind turbine tower 102. Microphone wires or wireless connection 301,303,305,307 relay microphone signals to a noise processor 312 in the nacelle 100. The blades 112, 114 attached to a rotor hub 118 of the turbine blades can be pitched by pitch bearings 308, 310. In this arrangement, the four microphones mounted on the ground measure rotor blade noise. The noise processor 312 uses the microphone signals, along with the rotor azimuth (blade rotational position) and yaw angle 143 (FIGURE 1) , to determine the noise levels. The processed noise signal is used by the noise processor 312 to compute the blade pitch signal for noise reduction, as described with respect to FIGURE 1 OR FIGURE 2.

Refer to FIGURE 4, which is a side view of the wind turbine of FIGURE 3 illustrating noise reduction using proximity microphones 164, 165, 166, 167, located circumferentially around the wind turbine tower 102. For the proximity microphones 164, 165, 166, 167, mounted on the tower 102, the noise signals are transferred via wires 401, 402, 403, 404, to the noise processor 312 to determine the noise levels. The processed noise signal is used by the noise processor 312 to compute the blade pitch signal for noise reduction, as described with respect to FIGURE 1 OR FIGURE 2. Refer to FIGURE 5, which is a block diagram of a noise controller , which processes measured noise signals into a noise metric and uses it to compute a blade pitch schedule to reduce noise. While different types of controller can be implemented based on this invention, the PID controller is described here. In this invention, the PID controller of

FIGURE 5 attempts to correct the error between a desired set point 500 (acceptable noise level) and a measured noise variable 512, by calculating and then outputting a corrective action 506 that can adjust the rotor blade pitch 508 accordingly.

The set point 500 is a nominal noise metric (decibel level) that represents an acceptable noise level. Microphones pick up noise from the turning rotor blades of a wind turbine, resulting in noise signals 512. The noise signals are input 514 to logic 516 that processes the noise signals into a measure noise metric 502.

The measured noise metric 502 and the nominal noise metric 500 (set point) are summed 501 and output to a controller 504. The output 506 of the controller 504 is a blade pitch command 506 that changes the pitch of each of the rotor blades of the wind turbine 508 in a direction that will attempt to reduce the rotor blade noise 510.

In addition to the PID controller, this invention can be implemented using the harmonic controller. Referring back to Figure 5, the harmonic controller 504 outputs blade pitch 506 in terms of sine or cosine of the rotor azimuth (blade rotational position) . The other elements of the harmonic controller are identical to those of the PID controller described previously

Method of Noise Reduction

Refer to FIGURE 6, which is a flow chart of the method of noise reduction in a wind turbine .

A value of a predetermined or learned noise level limit metric is determined and stored. At flow start 600, the controls maintain the blade pitch to desired operating position 602. The noise processor 608 uses the microphone signals 607, along with the yaw angle 606, to determine the noise level 609. The yaw angle defines the rotor orientation with respect to the microphones locations. The noise processor 608 uses the yaw angle to identify the appropriate weighting factor for each microphone signal before processing.

Sensing and processing of an instantaneous noise of the wind turbine results in an instantaneous noise signal 609. The noise signal 609 is presented to a decision block 610. If the noise signal exceeds the stored noise level limit metric, the flow continues to logic block 612. The instantaneous noise signal 609 is converted 612 to a blade pitch modulation 614 and the flow proceeds to logic 615.

If the blade pitch exceeds a design limit 615 the blades are not allowed to pitch farther. If the blade pitch does not exceed the design limit 615 the blades are allowed to pitch farther and the flow proceeds to logic 624. If the blade pitch does exceed the design limit 615 the blades are not allowed to pitch farther and the flow returns to block 610. Concurrently with the above steps, actual rotor RPM is detected 616 and presented to a decision block 618. Sensing of actual RPM of the wind turbine results in an instantaneous RPM signal. To maintain a target RPM, the instantaneous RPM is converted 620 to a nominal blade pitch command 622. During operation at less than rated RPM, the pitch and rotor speed may be varied to optimize performance.

The blade pitch modulation 614 is added at logic 624 to the nominal blade pitch command 622 resulting in a combined pitch command 625 to the blade pitch control logic 628. The blade is incremented accordingly and the flow returns 629 to logic 610.

The above-described steps are replicated for each rotor blade, typically three blades, each preferably controlled independently. Thusly, the combined pitch command 625 is used to control pitch of the rotor blades in order to compensate for the instantaneous noise of the wind turbine rotor blades. While the invention has been particularly shown and de- scribed with reference to preferred embodiments thereof, it will be understood by those skilled in the art that the foregoing and other changes in form and detail may be made therein without departing from the scope of the invention. Therefore, the invention is applicable to any wind powered device with adjustable pitch. The inventive concept is not restricted to to devices with turbines and/or generators housed in a nacelle but may be applied to any setup with a rotor generating rotor torque which is somehow coupled to generators or other force transforming means.

Claims

1. An apparatus that reduces noise in a wind turbine during operation of rotor blades comprising: a nominal rotor blade pitch command signal 154; a noise sensor, an output of which is a noise signal 147; conversion logic 146 connected to said noise signal 147, an output of said conversion logic being a pitch modulation command 162; and, combining logic 150 connected to receive said blade pitch modulation command 162 and to receive said nominal pitch command 154, an output of said combining logic 150 being a combined blade pitch command 156 capable of commanding pitch of the rotor blades, which includes compensation for instantaneous noise of said wind turbine.

2. The apparatus of claim 1 further including at least one blade rotational position sensor 134 an output of which is a blade position signal 140, the conversion logic 146 being connected to receive said blade position signal of said at least one blade position sensor.

3. The apparatus of claim 1, or 2 further including a storage 144 wherein the storage stores (i) nominal noise value, which is output 145 to the conversion logic 146 and (ii) pre-determined blade pitch limits to prevent blade fatigues or excessive blade loads .

4. The apparatus of one of claims 1, 2, or 3, wherein said conversion logic 146 calculates multiple individual blade pitch modulation commands 162, one command assigned to each of the rotor blades .

5. A wind turbine noise reduction apparatus comprising: a blade pitch controller 146, 148 providing pitch commands 154; and, means 160,161,162,162 for ground base acoustic sensors or 164,165,166,167 for acoustic sensors mounted on tower for sensing noise level from an operating wind turbine 100 resulting in a feedback metric 147 whose magnitude is proportional to said noise level; said blade pitch controller including means 146 responsive to said feedback metric 147 for adjusting blade pitch to reduce the magnitude of said feedback signal to a metric 145.

6. The apparatus of claim 5 further including at least one blade rotational position sensor 134 or rotor azimuth position sensor 124, an output of which is a blade position signal 140, the conversion logic 146 being connected to receive said blade position signal of said at least one blade position sensor.

7. The apparatus of claim 5, or 6 further including a storage 144 wherein the storage stores (i) nominal noise value, which is output 145 to the conversion logic 146 and (ii) pre-determined blade pitch limits to prevent blade fatigues or excessive blade loads.

8. The apparatus of one of claims 5, 6, or 7, wherein said conversion logic 146 calculates multiple individual blade pitch modulation commands 162, one command assigned to each of the rotor blades .

9. A method of noise or loads reduction in an operating wind turbine comprising steps of: A. sensing an external noise level from an operating wind turbine resulting in a feedback signal 147 sent to a blade pitch controller 144, 146, 148; and

B. using the feedback signal in said blade pitch controller to adjust turbine rotor blade pitch 156 to reduce the external noise level to a nominal sensed or predetermined level .

10. A method of noise reduction in a wind turbine, which uses a nominal pitch command 154 to control pitch of rotor blades of said wind turbine, comprising steps of:

A. storing 144 a value of a nominal noise level 145;

B. sensing an instantaneous noise of said wind turbine resulting in an instantaneous noise signal 147 or sensing a sustained noise above a threshold level resulting in a sustained noise signal 136;

C. converting 146 said instantaneous noise signal 147 or said sustained noise signal 136 and said nominal noise level 145 to a blade pitch modulation 162; D. combining 150 said blade pitch modulation 162 with said nominal pitch command 154 resulting in a combined pitch command 156; and

F. using said combined pitch command 156 to control pitch of the rotor blades in order to compensate for said instantaneous noise of the wind turbine rotor blades.

11. The method of claim 9 or 10 wherein the method uses acoustic signals, along with the rotor azimuth position and yaw angle 143, to determine noise levels.

12. The method of claim 11 wherein acoustic signals are sampled by one or more acoustic sensors .

13. The method of claim 11 wherein acoustic signals or air pressure variations are sampled by one or more blade mounted pressure sensors.

14. The method of claim 11 wherein acoustic signals or air pressure variations are sampled by one or more ground based or tower mounted pressure sensors.