JP6609462B2 - Wind power generation system - Google Patents

Description

  The present invention relates to a wind power generation system, and particularly compares a control value with measured wind condition information.

  As information used to control the wind turbine generator, as a means of grasping the wind speed value and wind direction value on the rotor surface of the windmill, actually measured by installing an anemometer or anemometer directly on the rotor blade (Patent Document 1) There is a technique. Alternatively, there are a method (Patent Document 2) for measuring the wind speed distribution on the windmill using light scattering, and a method using a Doppler radar (Patent Document 3).

JP 2014-47742 A JP 2013-177885 JP JP 2002-152975 A

  In Patent Document 1, since the wind speed / wind direction sensor is installed in the vicinity of the tip of the rotating blade, it is a problem to raise the cost of the apparatus and to consider durability. In addition, it is necessary to appropriately correct the speed vector of the rotating blade, which is much larger than the wind speed of the surrounding environment. In Patent Documents 2 and 3, it is necessary to consider the cost of Lidar (LIght Detection and Ranging) and Radar (RAdio Detection and Ranging) devices, which calculate the spatial distribution of wind velocity vectors, and light and electromagnetic waves that leak to the surrounding environment. It becomes. Any of the measuring means is not necessarily sufficient for measuring with high accuracy while simplifying the equipment added for measurement. It is an object of the present invention to provide a wind power generation system that can realize highly accurate measurement with a simple configuration.

In order to solve the above-described problems, a wind power generation system according to the present invention includes a rotor having blades that rotate by receiving wind, and depends on a wind turbine that generates electric power using the rotational energy of the rotor and the wind direction or wind speed. An operation for comparing the wind turbine control command value determined from the measured value other than the wind direction or the wind speed with a comparison value determined from the wind direction or the wind speed obtained by the wind information measuring device for detecting the wind direction or the wind speed in the wind turbine. equipped with a device, the control command value is characterized pitch angle der Rukoto of the blade.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the wind power generation system which can implement | achieve highly accurate measurement with a simple structure.

The figure which shows the outline of the same control value calculation by two or more methods The figure which shows the outline of the method of calculating the pitch angle by multiple methods and taking the deviation Correlation between wind speed value by anemometer and pitch angle (pitch angle A) by windmill control Example of pitch angle estimation from wind speed value by anemometer Correlation between deviation between pitch angle A and pitch angle B and Yaw error Application example for windmill protection Comparison with Yaw error over-protection operation by wind vane Example of using deviation between pitch angle A and pitch angle B for wind turbine control Improved Yaw error overjudgment example Changing the threshold according to the degree of turbulence Correction of translation and rotation of floating body Windmill overview

  Hereinafter, preferred embodiments for carrying out the present invention will be described with reference to the drawings.

[Relationship between various measured values and control values]
Various measurement values or control values measured in the operation of the windmill are not independent of each other but operate while maintaining correlation with each other. This is necessary to realize appropriate control, and these measured values and control values change in a relevant manner due to fluctuations in wind speed, which is the main input that determines the operation of the wind turbine. is there.

  Therefore, the wind turbine control value determined from the measured value other than the anemometer or anemometer, which changes depending on the wind direction or wind speed, and measures wind information, and the measured value by sensors such as anemometer and anemometer Fluctuate during normal times while maintaining relevance. On the other hand, depending on the input conditions, the relevance may be lost. By using this degree of disruption of the relationship, a new index relating to the state of the windmill can be obtained. In this embodiment, attention is paid to such a relationship.

[Relationship between control value and estimated value of control value]
FIG. 1 is a diagram showing an outline of a procedure for calculating the same wind turbine control value by a plurality of methods. FIG. 12 shows an overall view of the windmill 101. The windmill 101 has a blade 1 that rotates by receiving wind input and a rotor that rotates together with the blade 1 and has a hub 2 disposed at the center of the blade 1 and a generator that generates electric power using the rotational energy of the rotor. doing. The generator is not shown, but is provided so as to be rotatable in a substantially horizontal plane with respect to the tower 4 and is disposed, for example, in a nacelle 3 that rotatably supports a blade. A yaw actuator that rotates the nacelle 3 relative to the tower 4 is provided between the nacelle 3 and the tower 4. The operation of the wind turbine and the connected generator at this time fluctuates in accordance with the increase and decrease of the wind speed in a state in which the pitch angle, the rotor rotational speed generator torque, and the like are approximately related by the pitch control 102 of the wind turbine body. The wind turbine control value x 103 is a value controlled by the pitch control 102 of the wind turbine main body or a measured value observed as a result of the left control. Control values and measurement values related to the wind power generator are mutually determined according to the result of the pitch control 102 of the wind turbine body or the physical characteristics of the wind turbine itself (for example, the Cp value or the specifications such as the airfoil and rotor diameter). There is a correlation. By selecting the appropriate correlation depending on the current control mode of the windmill, the estimated value 113 of the windmill control value x corresponding to the windmill control value x103 is obtained from the physical value detected by the sensor 111, for example. The sensor value → control value conversion 112 may be used for calculation. If the deviation between the wind turbine control value x 103 and the estimated value 113 of the wind turbine control value x is calculated by the arithmetic unit 104 having a comparison function, the deviation value 105 which is the left output is obtained. The arithmetic device itself may be arranged inside or outside the windmill. If the wind turbine operating state is a typical state, the wind turbine control value x can be accurately estimated by the sensor value → control value conversion 112, and the deviation value 105 becomes small. On the other hand, when deviating from the above typical state, the deviation value 105 increases. Specific examples of non-typical cases include a case where a cross wind flows in and a case where the wind distribution is not uniform. By using this property and using the deviation value 105, an index related to the state of the windmill can be obtained.

[Specific example of pitch angle]
Next, the above-described method will be described more specifically with reference to FIG. In the example shown in the figure, the pitch angle is used as the control value x calculated by a plurality of methods. First, the wind turbine 101 rotor is rotated by receiving wind. At this time, a variable-speed variable-pitch horizontal-axis wind turbine, which is a typical large-scale wind power generator, enters a mode in which the pitch angle is controlled so that the rotor speed is constant and the generator torque is constant at a rated wind speed or higher. Therefore, an output selected as the control value x is a pitch angle A203. Next, an anemometer 211 is used as a sensor (wind information measuring device). The anemometer 211 may be, for example, installed on the nacelle or installed at a location away from the windmill main body, but preferably has a location that well reflects the state of the wind flowing into the windmill. Next, using the wind speed value 214 measured by the anemometer 211, the wind speed → pitch angle conversion unit 212 performs an operation for converting the wind speed value into the pitch angle. This calculation result is defined as a pitch angle B213. This calculation utilizes the property that various measured values or control values measured in the operation of the windmill operate while maintaining correlation with each other as described above.

  FIG. 3 shows an example of the correlation between the wind speed value by the anemometer and the pitch angle (pitch angle A) by the wind turbine control. By using this correlation and generating an approximate curve, the pitch angle is estimated from the wind speed value of the anemometer. Focusing on the distribution shape of FIG. 3, while there are plots that are densely distributed, there are also plots that are thinly distributed. Although the thin plot is expected to be out of the typical operating state, in this embodiment, focusing on the sample points that are out of the typical state, the wind turbine is detected from the out of the sample points. A new index related to the state is extracted.

  Next, an example in which the pitch angle (pitch angle B) is estimated from the wind speed value obtained by the anemometer will be described with reference to FIG. In this figure, the same signals (measured values and control values) as in FIG. 2 are assigned the same reference numerals. As shown in FIG. 4, the pitch angle (pitch angle A) 203 determined as a result of the windmill control and the pitch angle B (213) estimated from the wind speed value 214 by the anemometer show good agreement in many cases. There are also places where divergence is observed.

  The above is the flow and basis of the operation of the wind speed → pitch angle conversion unit 212 in FIG. Subsequently, a deviation 205 between the pitch angle A203 and the pitch angle B213 is taken. The absolute value of the deviation 205 between the pitch angle A203 and the pitch angle B213 takes a small value if the windmill is in a typical state (shows a good correlation as shown in the dark plot in FIG. 3). On the other hand, when it deviates from a typical state, it takes a large value. A typical state is, for example, a case where a wind that is nearly uniform flows into the rotor surface of a windmill from the front.

  FIG. 5 shows the correlation between the deviation 205 between the pitch angle A and the pitch angle B and the yaw error. In the figure, the horizontal axis is the deviation 205 between the pitch angle A and the pitch angle B, and the vertical axis is the Yaw error. The Yaw error of a horizontal axis windmill is the direction in which the nacelle's direction of orientation (roughly speaking, the direction between the rotor and the tower) on the rotation axis between the nacelle and the tower and the direction in which the wind flowing into the windmill is projected onto the horizontal plane. Is the difference.

  The reason for the correlation as shown in FIG. 5 is considered as follows. First, regarding the calculation of the pitch angle A, in a horizontal axis wind turbine with a variable speed and variable pitch, control is performed to change the pitch angle so that the rotor speed is constant and the generator torque is constant at a rated wind speed or higher. This control is performed independently of input from a wind speed sensor such as an anemometer. In other words, the windmill is a huge anemometer, and the pitch angle (pitch angle A) changes so that the rotor rotational speed and the generator torque satisfy certain conditions as the wind speed fluctuates. Here, if there is a Yaw error, the amount of energy that flows in is less than when the wind flows from the front of the rotor surface at the same wind speed. Accordingly, correspondingly, the pitch angle becomes shallower (in other words, on the fine side), and becomes a pitch angle corresponding to the case where the wind speed is reduced. In other words, this is based on the fact that the factor that the inflowing energy is reduced is mistakenly caused by the decrease in the wind speed. Actually, the wind speed does not decrease, but due to the yaw error, only the component directly facing the rotor of the wind vector has decreased, and the cross wind component remains. In windmills, due to the size of the rotor surface and mechanical / structural constraints of the drive mechanism of the Yaw shaft, such as the yaw actuator, the response time is long, and it is difficult to adjust to the wind direction with frequent and frequent fluctuations. On the other hand, an anemometer can be adjusted to a wind direction that fluctuates frequently and instantaneously. This is because it is small (such as a propeller type) or, in principle, has no directivity with respect to the wind direction in a horizontal plane (such as a cup type). In addition, some of them are less susceptible to mechanical restrictions (ultrasonic type, etc.). Therefore, it can be assumed that the wind speed value obtained by the anemometer relatively reflects the value when the wind is received from the front of the anemometer. Therefore, the pitch angle B converted from the wind speed value from the anemometer is not as shallow as the pitch angle A (not on the fine side). Therefore, if there is a Yaw error, the difference (absolute value) between the pitch angle A and the pitch angle B increases.

  FIG. 6 shows an example in which the deviation 205 between the pitch angle A and the pitch angle B is applied to wind turbine protection using the above properties. This figure is the most simplified configuration. In this configuration, the threshold 205 determines the magnitude relationship with the threshold 303 with respect to the deviation 205 between the pitch angle A and the pitch angle B, and the result 305 triggers the start of wind turbine protection operation due to excessive Yaw error. One of them. In the example of the figure, threshold judgment is performed without inserting an integrator (or low-pass filter processing) for the deviation 205 between the pitch angle A and the pitch angle B. This is because the deviation 205 is smaller in time variation than the measured value of the wind direction by the anemometer. Of course, before input to the threshold determination 302, integration processing or low-pass filter processing may be performed on the 205 to perform more stable determination. In the above case, it is preferable to consider the trade-off between stability of determination and response time.

  Fig. 7 shows a comparison with the Yaw error excessive protection operation based on the wind direction by the anemometer. The figure shows the time change of various measured values when a Yaw error value (difference in wind direction with respect to the nacelle front) by a conventional anemometer is measured and a protection operation is performed due to excessive Yaw error. Further, in addition to the left plot, a plot of deviation 205 between pitch angle A and pitch angle B shown in the present embodiment is added. In the figure, a light solid line 414 is a Yaw error value (difference in wind direction with respect to the nacelle front) by an anemometer, and this time the absolute value is plotted. A dotted line 214 is an anemometer wind speed value, and a solid line 205 is a deviation between the pitch angle A and the pitch angle B. In the horizontal axis of the same figure, time 412 is the timing of stop determination due to excessive Yaw error when using only anemometer data, and at time 413, too much Yaw error is excessive when using only anemometer data. It is the timing of stop determination in the case. These are whether or not the Yaw error value 414 (smooth value) by the anemometer reaches a predetermined value (not shown) and the wind speed value 214 (smooth value) by the anemometer exceeds a predetermined value (not shown). Judge with. The determination threshold value at 413 is larger than at least one of the threshold value for the Yaw error value 414 by the anemometer and the threshold value of the wind speed value 214 by the anemometer as compared to the determination threshold value by 412 (none is shown). Although the Yaw error value 414 and the wind speed value 214 by the anemometer seem to be small only at 412 and 413, these are due to a time delay due to the smoothing process. Actually, as a result of the high value continuing in a certain period preceding in time, the smooth value is set slightly and the threshold value is reached. Actually, a delay is also added in the sensor information collection cycle, control / communication, and the like. Here, paying attention to the plot 205 of the deviation between the pitch angle A and the pitch angle B shown in the present embodiment, not only at the time 413, but at a sufficiently early time (t1) 415 before reaching the timing at the time 412, The judgment threshold 303 is exceeded. The excess timing on the left is (t1) 415 in the figure. The plot of 205 after t1 is continuously high, suggesting that a quick and stable determination can be made. This represents that the deviation 205 between the pitch angle A and the pitch angle B shown in the present embodiment represents the state of the wind received by the entire rotor surface, particularly the state of the Yaw error received by the entire rotor surface. Because. Therefore, a case where the wind direction changes suddenly as a large tendency including the entire rotor surface can be detected with higher probability. Here, considering the Yaw error value by the conventional anemometer, since the Yaw error value by the anemometer has a larger fluctuation than normal time, noise must be removed by time integration processing or low-pass filter processing. Therefore, it is difficult to perform yaw control at an early timing. In addition, the wind direction that can be observed by the anemometer is, for example, the wind direction value at the installation position of the anemometer on the nacelle, and the wind direction on the entire rotor surface of the windmill is not necessarily measured. In fact, according to the example shown in FIG. 7, at t1 slightly before the timing of the time 412, the Yaw error value 414 by the anemometer becomes smaller and the Yaw error value may decrease in the future. It shows a change in value that could lead to misjudgment. On the other hand, at the same t1, the plot 205 of the deviation between the pitch angle A and the pitch angle B shown in the present embodiment stably maintains a high value, and the wind turbine as a whole shows that the Yaw error increases. It can be detected accurately. Therefore, if the wind turbine device is controlled using the comparison result (specifically, deviation 205) between the pitch angle A and the pitch angle B shown in the present embodiment, the wind turbine protection operation due to the excessive Yaw error can be more accurately performed. In addition, since it can be started at an early timing, the accumulation of windmill fatigue can be reduced. Since the degree of fatigue damage of a wind turbine is proportional to the higher order power of the fluctuating stress, it is useful to avoid a state where a large stress may be applied even if it is infrequent. Examples of wind turbine device control include shutdown due to excessive yaw error, correction of yaw error offset, replacement of wind direction measurement values by an anemometer, or compensation for accuracy of wind direction measurement values.

  Also, the change in wind direction starts on average from a high altitude location. This is because the average wind speed is higher in the sky. Therefore, since the vicinity of the upper end of the rotor surface starts to be affected by changes in the wind direction before the anemometer installed in the nacelle, using the method of this embodiment, Yaw is faster than using the anemometer from that point of view. An error may be detected. In the sense of detection on the entire rotor surface, for example, it is possible to detect changes in wind conditions in a wider range than an anemometer that measures the wind conditions in a centralized place such as on a nacelle. Can grasp the characteristic.

  According to the present embodiment, while changing depending on the wind direction or the wind speed, the wind turbine control value (specifically, the pitch angle control value) determined from the measurement value other than the wind direction or the wind speed, the wind direction meter in the wind turbine, By using the difference between the wind direction obtained by the anemometer or the comparison value determined from the wind speed, an amount related to the wind distribution on the rotor surface of the wind turbine can be obtained. By using the entire rotor surface as if it were a sensor, high-precision measurement can be performed with a simple configuration.

  FIG. 8 shows an example of the Yaw error excessive determination. In this embodiment, a Yaw error excessive determination result 460 using an anemometer and a determination result 305 obtained by threshold determination of the deviation 205 between the pitch angle A and the pitch angle B described in the first embodiment are logically calculated. In the example shown in FIG. 8, the logical sum operation 442 is specifically performed. With this configuration, it is possible to quickly determine an excessive Yaw error. Various operations including the logical sum operation 442 may be performed by one arithmetic device, or may be performed separately for a plurality of arithmetic devices.

  A method of calculating the Yaw error excessive determination result 460 using an anemometer will be described. Time average processing (D2) 421 is applied to the wind direction value from the anemometer 411. The time average may be any method as long as it can remove a fluctuation component and acquire a trend of a low frequency, such as integration calculation, low-pass filter processing, and moving average processing. Next, a threshold determination 423 is performed on the time average result with a threshold (d2) 422, and one of the inputs to the logical product 441 is configured.

  Next, a time average process 431 is applied to the wind speed value from the anemometer 211 in the process related to the wind speed. The time average processing method may be integration or low-pass filter, as in the case of the wind direction. There is no need to match the type with the time average processing in the wind direction. Next, a threshold determination 433 is performed on the time average result with a threshold 432, which is used as another input to the logical product 441. Further, after the time average processing 451 is similarly added to the output from the logical product 441, the threshold determination 453 is performed with the threshold (a2) 452, and the Yaw error excessive determination result 460 is output. The reason why the wind speed value is added to the determination of the excessive Yaw error is that it is difficult to affect the structure of the wind turbine in a weak wind, and it is not necessary to stop for preventing fatigue. In the present embodiment, the Yaw error excessive determination result 460 obtained as described above is logically ORed with the determination result 305 described in the first embodiment.

  The time average processing 451 and threshold determination processing 453 from the logical product 441 can be omitted depending on the time average processing and the threshold setting related to the wind direction and wind speed from the logical product 441, and the output of the direct logical product 441 is determined as a result of the determination. It is good. In the figure, V2 is a time constant of time average processing at the wind speed, for example, a time constant of temporary delay calculation. Similarly, v2 is a threshold value for determining the threshold value after the averaging process using the time constant in the wind speed. D2 in the figure is a time constant of time average processing in the wind direction, for example, a time constant of temporary delay calculation. Similarly, d2 is a threshold value for determining the threshold value after the averaging process using the time constant in the wind direction. Further, A2 in the figure is a time constant of time average processing in the logical product 441, for example, a time constant of temporary delay calculation. Similarly, a2 is a threshold value when the value after average processing by the time constant in the logical product 441 is determined as a threshold value. In the wind direction time averaging process, in order to detect a Yaw error, the absolute values may be averaged after averaging, or the positive and negative values may be averaged. When the positive and negative values are averaged, the absolute value of the threshold value may be set to a different value between positive and negative. This is because the wind power generator has asymmetry due to the influence of the rotation direction and tilt of the rotor.

  The determination criteria may be added to other determination factors depending on the system configuration. However, in the above example, since the essence is not impaired, the description is omitted, and only the main factors of the wind direction and the wind speed are described.

  The anemometer 211 is commonly used for the input to the wind speed → pitch angle conversion unit 212 and the input to the time averaging process 431 in the figure, but independent anemometers may be used. It is advantageous to use an independent anemometer when the accuracy of wind speed measurement and the type of anemometer required for the wind speed → pitch angle conversion unit 212 and the conventional time averaging process 431 are different. For example, as an input to the time averaging process 431, when it is determined whether or not the wind speed is equal to or higher than a certain wind speed at the time of excessive Yaw error determination, it may be sufficient if the accuracy is high enough to make the determination. Of course, if the threshold value of the Yaw error excessive determination is made variable according to the wind speed value, it is necessary to cope with it. An anemometer has a time response speed and a characteristic difference with respect to a wind in an oblique direction with respect to a rotating surface according to a type such as a cup type, a propeller type, and an ultrasonic type. Therefore, when the anemometer 211 is provided separately, the appropriate type should be selected in consideration of the relationship with the surrounding obstacles at the installation location of the anemometer, the difference in wind conditions, the influence of the wake from the surrounding windmill, etc. Is possible.

  In the above description, the case where the logical sum operation 442 is used as an example has been described. However, when the logical product is used, the erroneous determination rate of the Yaw error excessive determination can be easily reduced. When using a logical product, at least one of the determination threshold values 303, 432, 422, and 452 may be set lower. The lower threshold setting on the left can reduce the risk of undetected Yaw errors even though they are excessive. As a place where the logical product operation is performed, in addition to the method of replacing the logical sum operation 442 in FIG. There is a way. In this case, the time constant of the time averaging process 451 and the threshold value 452 may be appropriately changed. As an example, a time averaging process with a shorter time constant or a lower threshold value is changed. This is because the determination result 305 is newly added as a determination factor, and the reliability is improved, so that the necessity of reducing false detections by a long time averaging process and a higher threshold is reduced, and the determination result This is to make use of the quickness of the determination of 305.

FIG. 9 is an example of correcting the wind direction value of an anemometer used for various wind turbine equipment control using the deviation 205 between the pitch angle A and the pitch angle B. Regarding the wind direction value from the anemometer 411, the deviation 205 is used to correct the deviation of the wind direction value of the anemometer. First, the deviation 205 determines the direction of the Yaw error using the Yaw error direction determination 206. The direction of the Yaw error is a polarity indicating whether the wind direction flowing into the rotor surface is shifted to the right side or to the left side on the average on the entire rotor surface. As an example of the polarity determination on the left, the pitch angle control for each blade is used, and the polarity of the minute fluctuation of the rotor torque that occurs when the azimuth angle adds a minute fluctuation to the pitch angle near the zenith and the pitch angle near the bottom. And so on. The polarity of the minute fluctuations on the left varies depending on the rotor rotation direction, the polarity of fluctuation applied to the pitch, and the direction of the Yaw error. Since all the two are known, the direction of the Yaw error is known. Further, the deviation of the deviation 205 may be observed by actually changing the direction of the nacelle. Since the rotation of the nacelle on the left is limited in consideration of depletion, if it is necessary to reversely rotate, reverse rotation at a timing that considers the reaction time constant of the structure that causes movement by rotation after Yaw rotation stops The starting torque can be slightly reduced. Various measures such as those on the left are repeated to reduce mechanical stress. After determining the direction of the Yaw error, the pitch angle deviation is converted into a Yaw error at (215). For the conversion on the left, the correlation shown in FIG. 5 is used. Using the converted value on the left, 207 corrects the wind direction value from the anemometer. In addition to the above method, this correction method may be a method in which the wind direction selection unit 208 selectively uses the wind direction value from the anemometer 411 and the deviation of the wind direction from the pitch angle deviation 205. In accordance with the selection, the wind turbine general control 100 side switches the input value between the wind direction value itself and the deviation amount of the wind direction using a route (not shown). The wind turbine general control 100 is a function for performing overall control of a single wind turbine or a plurality of wind turbines, and includes the blade blade pitch angle control 102 of the wind turbine main body and other processing units in the drawing in a broad sense. In the present specification, some of the functions that can be included as the wind turbine general control 100 are described outside the 100 frame, and the other parts are shown as the 100 frame for convenience. The arithmetic device 104 having the comparison function described in the first embodiment and the device that performs the wind turbine general control may be configured as the same device, or may be provided as separate independent devices.
Selection instruction in the wind direction selection unit 208 is given by the selection adjustment mechanism 210. The left mechanism uses an input (not shown) from the anemometer, and outputs an instruction to the wind direction selection unit 208 to change the selection to control using the pitch angle deviation 205 when the variation in the wind direction is large. In addition to the alternative selection, the wind direction weight adjustment unit 209 that performs weighted averaging may be used. As in this embodiment, by using the pitch angle deviation 205 for wind turbine control, compared to the case where the wind direction is detected by an anemometer, control according to the wind condition received on the entire rotor surface of the wind turbine is possible. It becomes possible. For example, in the case of controlling the Yaw axis using an anemometer, only the wind direction at the place where the anemometer is installed can be observed, so the Yaw axis is controlled by the local wind direction. In this case, the control is not performed according to the state of the wind direction at the other part of the vast rotor surface. For example, if the anemometer is installed on the nacelle, it is difficult to control the wind direction in the vicinity of the upper end of the rotor surface away from the nacelle. In addition, when a large number of wind turbines are installed relatively close to each other like a wind farm, a situation occurs in which the wake of the wind turbine continues to be applied to a part of the rotor surface other than the vicinity of the nacelle of the wind turbine. sell. The situation on the left cannot be detected by an anemometer. On the other hand, in the method of the present invention, the pitch angle deviation varies depending on the wind conditions on the entire rotor surface, so that the above-described state may be detected. In the state where the wake continues to be applied to a part of the rotor surface, unlike the case of excessive Yaw error, the wind direction and other conditions do not change with a delay after the deviation 205 increases. Therefore, if the above condition is correct, it can be determined that turbulent flow is flowing into part of the rotor surface due to wake or topographical factors. In this case, it is possible to prevent accumulation of fatigue in the wind turbine by performing operation to reduce the maximum output. On the other hand, regarding the detection of Yaw errors, it is possible to take measures other than shutdown by utilizing the property that the change in wind direction starts on the sky side on average. For example, if the control of the Yaw axis is started when a sign of a change in the wind direction due to an increase in the pitch angle deviation 205 is detected, it is not necessary to perform a shutdown operation due to an excessive Yaw error. Such control may not be in time after waiting for the start of wind direction change by the anemometer. With the above control, power generation can be continued without going through a series of sequences that require time for shutdown and startup. In addition, it is possible to prevent the accumulation of fatigue in the wind turbine due to cross wind. In addition, if the shutdown and start-up can be omitted, it is possible to omit the application release and reapplication of the thrust to the rotor, and in a floating wind turbine, it is possible to prevent the accumulation of fatigue due to the swinging of the tower. In addition to shutting down due to excessive yaw error, the wind turbine device control includes, for example, offset correction of yaw error, replacement of wind direction measurement values by an anemometer, or compensation (or correction) of accuracy of wind direction measurement values. Can be mentioned.

  Next, FIG. 10 is used to show an example of changing the threshold depending on the degree of turbulence. In the configuration of FIG. 8, a turbulent flow index calculation means 331 is added to adjust the pitch angle deviation determination threshold 303 based on the calculated turbulent flow index.

  The turbulent flow index calculation means 331 determines the degree of turbulent flow using the degree of change of the wind speed value in time series and the measured value of the wind direction by the anemometer 411 depending on the case. When the degree of turbulence is large in a normal state, the determination threshold 303 is increased. Conversely, when the degree of turbulence is small, the determination threshold 303 is decreased. This method can reduce false alarms due to excessive Yaw errors, such as when turbulence occurs. Conversely, when the degree of turbulence is small, the determination time of excessive Yaw error can be shortened by lowering the threshold value. As an index of the degree of turbulent flow in the turbulent flow index calculating means 331, there is a method of taking an RMS (Root Mean Square) value relating to a fluctuation component of wind speed. In addition, the RMS value of wind direction fluctuation may be used as an index of turbulence. This is because the degree of turbulence may change depending on the topography around the windmill and the arrangement of obstacles. For example, in the case of wind from a certain direction, there are terrain and artifacts that disturb mountains and other winds in the windward direction, and if it is known in advance that a highly disturbed wind will flow in, it is possible to Alarms can be reduced. The same applies when a tendency of a large degree of turbulence is known from past observation results. On the other hand, when the sea surface continues far and the wind turbulence is small, the determination threshold 303 can be lowered. If the judgment threshold value 303 is lowered, it is possible to improve the detection sensitivity of a Yaw error in a windmill installed in an environment where the wind is stable during normal times. From the left, it is possible to quickly determine whether a cold front has passed or a gust front has occurred. On the other hand, when the wind is flowing in from the direction of the wind where the turbulence is large, increasing the determination threshold 303 can reduce the erroneous determination rate.

  In the above example, the turbulent flow index based on the measurement value from the anemometer or the anemometer is used to change the determination threshold 303. However, an index (not shown) that focuses on the vertical component of the wind may be used. The deviation 205 between the pitch angle A and the pitch angle B shows a good correlation with the Yaw error when the wind direction fluctuates with a component in the horizontal plane. On the other hand, the actual wind direction component includes not only the horizontal plane but also the vertical component. Therefore, by using a means for detecting the vertical component of wind (not shown) (wind vertical component detecting means) and taking the above measures, it is possible to achieve high accuracy with respect to detection of an excessive Yaw error state. Specifically, when the deviation 205 between the pitch angle A and the pitch angle B mainly depends on the vertical component of the wind, the determination threshold 303 is adjusted. The adjustment described on the left is performed based on the vertical component information from the means for detecting the vertical component of the wind. When there are many vertical components, the determination threshold 303 is increased to prevent erroneous determination that it is determined that the Yaw error is excessive. When detecting the vertical component of the wind, the Yaw error judgment is not completely suppressed, but maintaining the judgment threshold 303 while increasing it may be related to the degree of fatigue damage of the wind turbine even though the vertical component of the wind direction is maintained. It is. However, unlike the case where a large angle error can occur continuously, such as a Yaw error that occurs in the horizontal direction, the probability of continuously taking a high value for the vertical component of the wind is considered to be relatively small. This is because the wind in the vertical direction has at least one surface that is the ground or the water surface with respect to the traveling direction of the wind and has no escape. Therefore, depending on the size and strength of the wind turbine, there is a method in which a gate such as a logical sum is provided between the threshold determination 302 and the logical sum operation 442 to suppress the determination result 305 itself using the pitch angle deviation. In this way, by making the pitch angle deviation determination threshold 303 variable according to the environment around the windmill, it is possible to make a quick determination while reducing false alarms.

  Next, an example in which the present embodiment is applied to a floating offshore wind power generator will be described with reference to FIG. Among the offshore wind power generators, in the case of a floating type that is not firmly fixed to the seabed, the floating body itself causes translational motion and rotational motion due to the action accompanying the motion of the rotating body such as wind and rotor. For example, the rotational motion of the floating body around the Yaw axis is observed by being superimposed on the Yaw error calculated from the wind direction observed on the nacelle and the deviation between the pitch angle A and the pitch angle B. Therefore, in this embodiment, the floating body translation / rotation detection sensor 342 detects rotation about the Yaw axis, corrects the superimposed error, and determines the yaw error. The correction is made for both the anemometer output and the Yaw error conversion value from the pitch angle. In the latter case, the deviation 205 between the pitch angle A and the pitch angle B is performed even if the pitch angle A203 is used. Also good. The detection of the rotation of the floating body may be performed by a gyro sensor or an orientation sensor. When the gyro sensor is used, there is little influence of magnetization of the wind turbine and the floating structure that are likely to be magnetic bodies. When the azimuth sensor is used, there is no accumulation of errors that are likely to occur when an inexpensive rate gyro is used. Therefore, even if correction is continued for a long time, an operation such as azimuth resetting is unnecessary. Since both the gyro sensor and the orientation sensor have advantages and disadvantages, the characteristics may be improved by combining the two. Although the example of the correction is rotation around the Yaw axis, the inclination of the tower that supports the nacelle, and the attitude and movement of the nacelle due to the inclination may be corrected. In the correction related to the rotation including the axis other than the yaw axis of the floating body, the rotation about the pitch axis has the greatest influence on the wind speed value. The pitch axis is a general rotation axis related to the attitude of ships and airplanes, not the rotor blade pitch. Specifically, it is a rotation axis in the direction in which the rotor surface falls. In a floating wind power generator, the rotor surface is tilted to a considerable extent by the thrust force received by the wind. For example, when the state changes, such as when power generation is stopped, the nacelle is displaced considerably by the autonomous recovery of the tower inclination by removing the thrust force. In the situation on the left, a rotation occurs that significantly affects the wind speed observed on the nacelle. The detection of these rotations may be performed using a gyro sensor installed on the floating body or using an acceleration sensor installed on the nacelle. Moreover, you may use the inclination sensor installed in the tower. A three-dimensional magnetic sensor may be used. Since these sensor groups have advantages and disadvantages as described above, not only a single sensor but also information of a plurality of sensors may be combined to appropriately correct the accumulated error.

  In the embodiment described above, an example in which the wind speed value is equal to or higher than the pitch angle control mode is shown, but the same applies to a case where the wind speed is higher than the cut-in wind speed and lower than the wind speed in the pitch angle control mode. In the above wind speed range, a typical wind speed value is calculated from the relationship between the rotor speed and the generator torque, and the deviation from the wind speed value by the anemometer is calculated, thereby realizing the same functions as all the above embodiments. it can.

1 blade
2 Hub
3 Nasser
4 tower
100 Windmill general control
101 windmill
102 Wind turbine body pitch control
103 Wind turbine control value x
104 Deviation calculation
105 Deviation value
111 sensor
112 Sensor value → Control value conversion
113 Estimated value of wind turbine control value x
203 Pitch angle determined as a result of windmill control (pitch angle A)
205 Deviation between pitch angle A and pitch angle B
206 Yaw error direction determination
207 Wind direction correction
208 Wind direction selection
209 Wind direction weight adjustment
210 Selection / adjustment mechanism
211 Anemometer
212 Wind speed → Pitch angle converter
213 Pitch angle (pitch angle B) estimated from anemometer wind speed
214 Anemometer wind speed
215 Yaw error conversion from pitch angle deviation
302 Threshold judgment
303 threshold
305 Threshold judgment result
331 Turbulence index calculation means
342 Floating body translation / rotation detection sensor
411 Anemometer
412 Stop processing start timing with Yaw error threshold 1
413 Stop processing start timing with Yaw error threshold 2
414 Yaw error value by anemometer (absolute value of difference of wind direction with respect to nacelle front)
415 Stop processing start timing based on Yaw error excess judgment using pitch angle deviation
421 average processing
422 threshold
423 Threshold judgment
431 hour averaging
432 threshold
433 Threshold judgment
441 AND operation
442 OR operation
451 hour average processing
452 threshold
453 Threshold judgment
460 Conventional Yaw error overjudgment result
461 Improved Yaw error overjudgment result 1
462 Improved Yaw error overjudgment result 2

Claims (8)

A wind turbine comprising a rotor having blades that rotate by receiving wind, and that generates electric power using rotational energy of the rotor;
The wind turbine control command value determined from a measurement value other than the wind direction or wind speed and the wind direction or wind speed determined by a wind information measuring device that detects the wind direction or wind speed in the wind turbine, and changes depending on the wind direction or wind speed. Comprising an arithmetic device that compares the comparison value ,
Wind power generation system and the control command value, wherein the pitch angle der Rukoto of the blade The wind power generation system according to claim 1,
A wind power generation system that controls wind turbine equipment using a comparison result between the pitch angle and the comparison value The wind power generation system according to claim 2,
The wind turbine device control is any one of a shutdown due to an excessive yaw error, a correction of an offset of a yaw error, a correction of a wind direction measurement value by an anemometer, or a replacement with a wind direction measurement value by an anemometer. The wind power generation system according to any one of claims 1 to 3,
The comparison result is compared with a predetermined threshold value to determine a threshold value,
A wind power generation system characterized by performing a logical operation on the threshold judgment result and the yaw error excess judgment result using an anemometer The wind power generation system according to claim 4,
Comprising turbulent flow index calculating means or wind vertical component detecting means,
A wind power generation system that adjusts the threshold value based on information on a degree of turbulence or a vertical component of wind The wind power generation system according to claim 5,
When the degree of the turbulent flow is large or the vertical component is large, the threshold value is increased. The wind power generation system according to claim 5 or 6,
When the degree of the turbulent flow is small or the vertical component is small, the value of the threshold value is reduced. The wind power generation system according to any one of claims 1 to 7,
The windmill includes a nacelle that rotatably supports the rotor, and a tower that rotatably supports the nacelle,
Furthermore, a floating body that supports the windmill on the water;
A sensor that is disposed on the tower or the floating body and detects at least one of translation, rotation, or inclination of the tower or the floating body;
A wind power generation system using an output of the sensor when determining a yaw error

Images