TW202016428A - Wind-based power generating system characterized in that the operation controlling portion can improve blade aerodynamic performance by considering the wing deformation and suppress the reduction of the power generating efficiency by considering the real machine operation - Google Patents

Abstract

The present invention provides a wind-based power generating system equipped with an operation controlling portion. The operation controlling portion can improve blade aerodynamic performance by considering the wing deformation and suppress the reduction of the power generating efficiency by considering the real machine operation. The present wind-based power generating system is characterized by being equipped with: a plurality of blades capable of changing the thread pitch angle by a thread pitch angle driving device; a rotor which rotates by the blades receiving wind; and a power generator using the rotational energy of the rotor to generate power. The wind-based power generating system is also equipped with a deformation considering thread pitch angle command value calculating portion which calculates the thread pitch angle command value of considering the blade deformation, and endows the thread pitch angle driving device with the thread pitch angle command value of the deformation considering thread pitch angle command value calculating portion, so as to change the thread pitch angle.

Description

Wind power generation system

The present invention relates to a wind power generation system, and in particular to a wind power generation system that reduces the reduction in power generation efficiency caused by an increase in blade deformation.

In recent years, global warming caused by increased carbon dioxide emissions and energy shortages caused by the depletion of fossil fuels have been regarded as problems. Therefore, as a power generation system that reduces carbon dioxide emissions and does not use fossil fuels, the introduction of power generation systems that utilize renewable energy sources such as wind or sunlight to obtain from nature has attracted attention.

In the power generation system using renewable energy, it is generally a solar power generation system, but since the output directly changes according to the sunshine, the output varies greatly, and the power cannot be generated at night. On the other hand, the wind power generation system can be installed by selecting locations where wind conditions such as wind speed and wind direction are stable, so as to realize relatively stable power generation day and night.

In general large-scale wind power generation systems, there are a pitch drive device for adjusting the pitch angle and a power converter for adjusting the generator torque. By adjusting the pitch angle and the generator torque, it can be used in any wind speed area. Implement control in a way that maximizes power generation.

As one of the above control methods, for example, there is Patent Literature 1. In Patent Document 1: measurement information input processing step, which inputs at least measurement information of wind speed and direction of wind flowing into the windmill wing; torque calculation processing step, which is based on the measurement information input and used in the above measurement information input processing step Use the pre-memory data to calculate the torque to calculate the actual generated torque in each wing element and the optimal torque set in each wing element calculated by the product of the radius position, weight and angular velocity of each wing element ; And a torque comparison processing step, which compares the generated torque calculated in the above torque calculation processing step with the optimal torque; it can be based on the result of the comparison in the above torque comparison processing step to reduce the generated torque The method of difference from the optimal torque corresponds to the characteristics of the airflow generating device, the pitch angle driving mechanism, and the yaw angle driving mechanism, and controls each individually. In this way, the technology for improving the power generation efficiency of the wind power generation system is disclosed. [Prior Technical Literature] [Patent Literature]

[Patent Document 1] Patent No. 5323133

[Problems to be solved by the invention]

In recent years, wind power generation systems have required increased power generation power brought about by large-scale, with the continuous development of blades with long wings. Especially in downwind-type windmills, blades are installed on the downwind side, and the blades are bent downwind due to wind load, so there is less risk of hitting the tower. Therefore, compared with the upwind-type windmill, the blades can be elongated by flexibly structuring the blades (hereinafter referred to as long-wing blades).

By applying the technology disclosed in Patent Document 1, it is possible to operate at a pitch angle considering the aerodynamic performance of the blades according to the wind speed, the rotation speed of the rotor or the generator, and improve the power generation efficiency.

However, in long-wing blades that can improve power generation efficiency, there is a possibility that the amount of blade deformation becomes large. The aerodynamic performance of the blade is determined by the inflow relative wind speed and the original pitch angle of the blade, which is determined according to the wind speed, rotation speed, and wing deformation. In the case of a newly constructed blade, there is no amount of wing deformation, so when the wind speed and rotation speed are timed, the pitch angle that maximizes the aerodynamic performance is uniquely determined. However, in the case of flexible blade construction, considering the influence of the amount of wing deformation (bending and twisting) in the relative wind speed, it is necessary to derive the pitch angle that maximizes the aerodynamic performance. Therefore, even if the inflow relative wind speed is the same, the pitch angle of the rigid blade and the flexible blade to maximize the aerodynamic performance is also different.

Therefore, when applying Patent Document 1 to the case of long-wing blades, it is impossible to consider that the pitch angle that maximizes the aerodynamic performance changes due to the amount of wing deformation compared to the case of rigid-structured blades with less wing deformation . Therefore, there is a possibility of adjusting the pitch angle to reduce the power generation efficiency. In addition, because the azimuth angle is not used for the input value, the effect of wind shear at higher wind speeds at higher altitudes cannot be considered. Furthermore, the effect of the reduced tower shadow effect near the tower due to the influence of the tower cannot be considered in the downwind windmill.

Therefore, there is a possibility that the influence of changes in wind speed in one rotation period cannot be considered, resulting in a decrease in power generation efficiency and an increase in wing vibration. In addition, when the error between the numerical analysis data of the analysis model of the wind power generation system and the characteristics of the actual machine is large, there is a possibility that the power generation efficiency may be reduced.

In view of the above, an object of the present invention is to provide a wind power generation system including an operation control unit that improves blade aerodynamic performance by considering the amount of blade deformation, and suppresses a reduction in power generation efficiency by considering actual machine operation. [Technical means to solve the problem]

In order to solve the above-mentioned problems, the present invention is constituted as follows: "A wind power generation system is characterized by comprising: a plurality of blades, etc., which can be changed by a pitch angle driving device; and a rotor, which rotates when the blade receives wind ; And a generator, which uses the rotational energy of the rotor to generate electricity; and is provided with a deformation amount considering a pitch angle command value calculation unit that takes into account a pitch angle command value of the blade deformation amount, and a deformation amount is given to the pitch angle drive device to consider the pitch angle The pitch value command value of the command value calculation unit changes the pitch angle". [Effect of invention]

According to the present invention, it is possible to provide a wind power generation system equipped with a control device that realizes power generation efficiency by correcting the pitch angle command value based on the actual machine operating data by considering the pitch angle command value of the blade deformation Increase or decrease in load.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In addition, in each drawing, the same components are marked with the same symbols, and detailed descriptions of the overlapping parts are omitted. [Example 1]

In the description of Embodiment 1, here, first, the configuration and control method of the previous general wind power generation equipment will be described using FIGS. 1 to 4, and then Embodiment 1 will be described using FIGS. 5 to 14.

FIG. 1 shows an example of a general configuration of a general wind power generation system to which the present invention can be applied.

The wind power generation system 1 of FIG. 1 includes a rotor 4 composed of a plurality of blades 2 and a hub 3 connecting the plurality of blades 2. The rotor 4 is connected to the nacelle 5 via a rotating shaft (omitted in FIG. 1 ), and the position of the blade 2 can be changed by rotation. The nacelle 5 supports the rotor 4 and enables it to rotate. When the blade 2 receives the wind and the rotor 4 rotates, the rotational force of the rotor 4 rotates the generator 6 in the nacelle 5, thereby generating electric power. Furthermore, the nacelle 5 is provided with a wind direction and wind speed sensor 7 for measuring wind direction and wind speed.

Each blade 2 is provided with a pitch angle driving device 8 capable of adjusting the angle (pitch angle) of the blade 2 with respect to the wind. By using the pitch angle driving device 8 to change the pitch angle, the wind force (air volume) received by the blade 2 can be adjusted, thereby changing the rotational energy of the rotor 4 relative to the wind. Thereby, the rotation speed and generated power can be controlled in a wide wind speed area.

In the wind power generation system 1, the nacelle 5 is installed on the tower 9 and has a mechanism (omitted in the figure) that can rotate relative to the tower 9. The tower 9 supports the load of the blade 2 via the hub 2 and the nacelle 5 and is fixed to a base (omitted in the figure) provided at a specific position on the ground, at sea, or in a floating body.

The generator 6 can control the torque generated by the generator (hereinafter referred to as generator torque) by the power converter 10 provided in the tower 9, thereby controlling the rotational torque of the rotor 4.

In addition, the wind power generation system 1 includes a controller 11. Based on the rotation speed output from the rotation speed sensor 12 that measures the rotation speed of the generator 6 and the generator torque of the generator 6, the controller 11 is used to adjust the generator 6 and The pitch angle driving device 8 adjusts the generated power or the rotation speed of the wind power generation system 1.

Furthermore, the controller 11 uses, for example, a control panel or SCADA (Supervisory Control And Data Acquisition, monitoring and data acquisition). In addition, the controller 11 uses a processor such as a CPU (Central Processing Unit) not shown, a ROM (Read Only Memory) that stores various programs, and a RAM that temporarily stores data of the calculation process (Random Access Memory, random access memory), external memory devices and other memory devices are realized, and CPU and other processors read and execute various programs stored in ROM, and store the calculation results as execution results in RAM or external memory devices .

The blade 2 may be, for example, a rotor with a diameter of 100 m or more. In addition, when the rotor diameter is set to 180 m or more, the effect corresponding to the control of the flexible structure is large. In addition, although the blade 2 has an initial twist in shape according to the rotation surface, it is possible to design the web of the blade 2 in such a way that the twist applied by the wind during the power generation operation generates a twist of 0.2° or more Or spar caps, etc. In addition, in the case of a flexible structure blade with a twist of 0.5° or more during power generation operation, the control of this embodiment can obtain a particularly remarkable effect.

FIG. 2 shows a block diagram of the operation control unit installed in the controller 11. The variable speed control unit 21 as the operation control unit shown in FIG. 2 includes a pitch angle control unit 22 based on the deviation between the target value of the generator torque and the measured value and the target value of the generator rotational speed and The deviation of the measured value is determined by the feedback control of the pitch angle command value. In addition, the variable speed control unit 21 includes a generator torque control unit 23 that determines the generator torque command value by feedback control based on the deviation between the target value of the generator rotation speed and the measured value. Furthermore, the pitch angle command value is given to the pitch angle driving device 8 of FIG. 1 provided for each blade 2 to adjust the respective pitch angle, and the generator torque command value is given to the generator 6 of FIG. 1 (here To include the power converter 10), the generator torque is adjusted.

FIG. 3 is a block diagram showing the outline of the pitch angle control unit 22 in the variable speed control unit 21. The pitch angle control unit 22 includes a rotation speed control unit 22a and a torque control unit 22b. The rotation speed control unit 22a determines the pitch angle command value by feedback control based on the deviation between the target value and the measured value of the generator rotation speed. Furthermore, the torque control unit 22b determines the pitch angle command value by feedback control based on the deviation between the target value of the generator torque and the measured value. By adding these two values, the final pitch angle command value of the pitch angle control unit 22 is determined.

In FIG. 4, the characteristics of the wind power generation system 1 obtained by the operation control unit installed in the controller 11 shown in FIGS. 2 and 3 are shown.

FIG. 4 shows the relationship between the generated power with respect to the wind speed, the rotation speed of the generator, the generator torque, and the pitch angle. The horizontal axis of each graph represents the wind speed, and the wind speed is faster toward the right. In addition, the vertical axis of each graph indicates that the higher the value of the generated power, the rotation speed, and the generator torque, the higher the value. Regarding the pitch angle, the upper side indicates the feather (wind) side, and the lower side indicates the feather (wind) side.

Power generation is performed within the range of the cut-in wind speed Vin that starts the rotation of the rotor 4 to the cut-out wind speed Vout that stops rotation, so that the power generation value increases as the wind speed increases to the wind speed Vd, but the power is generated at the wind speed above it Be a fixed way to control.

The controller 11 controls the generator torque such that the rotation speed from the cut-in wind speed Vin to the wind speed Va becomes fixed (Wlow), and controls the generator rotation from the wind speed Va to the wind speed Vb such as a rotation speed proportional to the wind speed Moment. In the stage up to this point, the generator torque control unit 23 of FIG. 2 is used to give the generator torque command value, thereby performing control.

When the wind speed Vb is reached and the rotation speed reaches the rated rotation speed Wrat, the generator torque and the pitch angle are controlled in such a manner that the rated rotation speed Wrat is maintained in the wind speed state above the wind speed Vb. At this stage, the pitch angle control unit 22 of FIG. 2 and the generator torque control unit 23 operate in cooperation. For example, as a result of the operation of the pitch angle control unit 22, the pitch angle command value changes, and the rotation speed of the rotor 4 is barely a generator rotation speed change, and the generator torque control unit 23 executes the change of the generator rotation speed to make the rotation speed fixed. Torque command value and other programs, thereby controlling the generator torque and pitch angle in a manner to maintain the rated rotation speed Wrat.

In the control of the controller 11 described above, basically, the control of the generator torque is performed to ensure the generated power. In the control of generator torque, within the range from wind speed Vb to wind speed Vd, the generator torque is changed according to the wind speed until it becomes the rated generator torque Qrat, and within the range from wind speed Vd to cut-out wind speed Vout, Maintain the rated generator torque Qrat.

In the control of the pitch angle, the pitch angle is maintained at the reverse propeller angle θmin up to the wind speed Vc, and within the range from the wind speed Vc to the cut-out wind speed Vout, the pitch angle is changed from the propeller side θmin to the feather side θmax according to the wind speed . However, in the example of FIG. 4, the control of the generator torque and the pitch angle is overlapped in the range from the wind speed Vc to the wind speed Vd, but it can also be set to Vc=Vd to eliminate the overlap and execute the generator rotation independently Moment control and pitch angle control.

The present invention is based on the premise of the wind power generation system described in FIGS. 1 to 4 above, by adjusting the pitch angle to prevent a reduction in power generation efficiency caused by the deformation of the blade 2 or an increase in wing vibration.

In Embodiment 1 of the present invention, by adjusting the pitch angle, the reduction of the power generation efficiency or the wings caused by the deformation of the blade in the wind speed region from the cut-in wind speed Vin to the wind speed Vd, which is particularly required to improve the power generation efficiency, is prevented Increase in vibration.

FIG. 5 shows a block diagram of the deformation amount considering pitch angle command value calculation unit 100 according to the first embodiment of the present invention. As shown in FIG. 5, the deformation amount considering pitch angle command value calculation unit 100 includes an analysis model 101 and a learner 102. In addition, the wind speed is measured by the wind speed measurement unit 103, the rotation speed of the generator is measured by the rotation speed measurement unit (rotation speed measurement unit) 104 of the rotor or generator, and the deviation from the wind direction is measured by the yaw error measurement unit 105 Aerospace error, the cabin tilt angle is measured by the cabin tilt angle measurement unit 106, the azimuth angle is measured by the azimuth angle measurement unit 107, the generated power is measured by the generated power measurement unit 108, and the wing deformation is measured by the wing deformation measurement unit 109 , And input the measured measurement signals 110 to the deformation amount considering pitch angle command value calculation unit 100.

The analysis model 101 given the measurement signal 110 outputs a command value of the pitch angle in consideration of the amount of deformation calculated in advance by simulation. In the learner 102, based on the measurement signal 110, a function for outputting a pitch angle command value is learned during operation, and the deformation amount of the function is output to consider the pitch angle command value.

In FIG. 5, the deformation amount-considered pitch angle command value calculation unit 100 according to Embodiment 1 of the present invention is a calculation unit in which the deformation amount-considered pitch angle command value is calculated. It includes two calculation units (analysis model 101 and learner 102). It is possible to modify the relationship between the characteristics of the initially prepared analysis model 101 according to the experience of the learner 102 as shown in the condition S10, or add the output reflecting the experience of the learner 102 to the initial as shown in the condition S11 The output of the state analysis model 101 is set to the optimal pitch angle command value suitable for the current operating state.

In addition, the pitch angle command value calculated by the pitch angle command value calculation unit 100 in consideration of the deformation amount is obtained for each blade, and is directly given to the pitch angle driving device 8 of each blade, or added separately to FIG. 2 The pitch angle command value of each blade given by the variable speed control unit 21 is given to the pitch angle drive device 8 of FIG. 1.

FIG. 6 illustrates the relationship between the analysis model 101 and its input using square line curves. As shown in FIG. 6, the measurement signal 110 is composed of a wind speed measurement unit 103, a rotation speed measurement unit (rotation speed measurement unit) 104 of a rotor or a generator, a yaw error measurement unit 105, a cabin inclination angle measurement unit 106, and an azimuth The measurement signal measured by the angle measuring unit 107 is input to the analysis model 101 as a state signal 111, and outputs a pitch angle command value that takes into account the wing deformation amount stored in advance (the deformation amount considers the pitch angle command value).

FIG. 7 is a flow chart of an analysis model used to prepare the output deformation amount in consideration of the pitch angle command value in advance. In the processing step S100, input the parameters of wind speed, rotation speed, yaw error, cabin tilt angle, and azimuth angle.

In the processing step S101, input the initial value of the pitch angle. In processing step S102, based on the input values of processing step S100 and processing step S101, the amount of deformation and aerodynamic performance of each wing element are calculated by the aerodynamic and physical models of the blade.

In processing step S103, it is determined whether the pitch angle that maximizes the aerodynamic performance is calculated. In the case where it is not calculated, after changing the value of the pitch angle in processing step S101, the processing of processing step S102 is performed again. With this, the pitch angle that maximizes the aerodynamic performance considering the amount of deformation is explored for the same parameter.

In the processing step S104, it is determined whether the available measurement information is included. In the case of not being covered, the processing of the processing steps S100-S103 is executed again. Based on this, we explored the pitch angle that maximizes the aerodynamic performance considering the amount of deformation for the parameters of the operating state of the windmill.

Through the processing step S105, an analysis model of the pitch angle that maximizes the aerodynamic performance considering the amount of deformation is prepared for the parameters that include the operating state of the windmill.

Furthermore, if the maximum aerodynamic performance in the wind speed region from the cut-in wind speed Vin to the wind speed Vd is required to increase power generation efficiency, it means that the torque or force applied to the rotation direction of the entire blade is maximized, and the entire blade is The quotient of the force applied in the direction of rotation and the force applied in the direction of the thrust of the entire blade maximizes, maximizes the overall lift of the blade, or maximizes the quotient of the overall lift and resistance of the blade.

In addition, if the maximum aerodynamic performance during power generation operation that requires reducing the blade load wind speed Vd or more or during power generation standby during storms is maximized, it means that the force applied to the rotation direction of the entire blade and the thrust direction to the entire blade Maximize the quotient of the applied force, minimize the force applied in the thrust direction of the entire blade, maximize the quotient of the overall lift and resistance of the blade, or minimize the resistance of the entire blade.

Therefore, based on the measurement conditions, the deformation amount referenced by the self-analysis model can also be changed according to the wind speed area in consideration of the pitch angle.

The realization form of the analysis model 101 may be a function form and a table reference type. The function is made by fitting methods such as interpolation and extrapolation, or machine learning, using analysis and past action data. With this, the amount of information of the analysis model can be reduced. However, there is a possibility of being accompanied by approximation errors. The table reference type can reduce the approximation error by storing multiple data. However, the amount of information in the analysis model increases.

In the case where the analysis model 101 is composed of a table reference type, in the example of FIG. 6, five of wind speed, rotation speed, yaw error, cabin tilt angle, and azimuth angle are used as input parameters, and the pitch angle command value is used as Five-dimensional output table. Furthermore, the pitch angle command value may not only be a fixed value, but also a value with a specific amplitude. The analysis model 101 in the form of a five-dimensional table is constructed by executing the flow of FIG. 7 in advance, and in actual use, outputs a pitch angle command value determined by five input parameters input at the same time. In addition, when there is no one value determined by the five parameters on the table, the following processing can be performed: the pitch angle command value is obtained by appropriate interpolation processing or selecting an approximate value determined by multiple input parameters.

As described above, the analysis model 101 in the situation shown in FIGS. 6 and 7 is preset with the model characteristics of the analysis model 101 by the method of FIG. 7, the measurement signal 110 corresponding to the current status is input, and the deformation is considered with reference to the analysis model 101 The value of the pitch angle command value.

FIG. 8 is a block diagram showing the relationship between the input of the learner 102 of the deformation amount considering the pitch angle command value calculation unit 100 of the first embodiment of the present invention and its input.

As shown in FIG. 8, the learner 102 is the same as the analysis model 101. The wind speed is measured by the wind speed measuring section 103, and the rotational speed of the generator is measured by the rotational speed measuring section (rotation speed measuring section) 104 of the rotor or generator. The yaw error measurement section 105 measures the yaw error relative to the wind direction, the cabin tilt angle measurement section 106 measures the cabin tilt angle, the azimuth angle measurement section 107 measures the azimuth angle, and inputs these measurement signals as the status signal 111 .

In addition, in order to obtain the evaluation signal 112 in the evaluation signal determination unit 114, the azimuth angle measured by the azimuth angle measurement unit 107, the generated power measured by the generated power measurement unit 108, and the measurement by the wing deformation measurement unit 109 are used. Measurement signal of wing deformation.

During the calculation of the evaluation signal 112, the wing vibration speed calculation unit 113 derives the wing vibration speed based on the measurement signal measured by the wing deformation measurement unit 109. In the wing vibration speed calculation unit 109, for example, the speed can be derived by differentiating the measured strain or displacement. Alternatively, the acceleration can be derived from the second-order differential of the measured strain or displacement. Alternatively, the measured strain or displacement can also be used directly.

Due to the influence of wind shear and tower shadow effect, the wing vibration speed may increase depending on the azimuth. In addition, during actual operation, the wind speed is accompanied by turbulence. Therefore, due to the influence of the wind speed fluctuation, the wing vibration speed may be further increased. Especially in long-wing blades, there is a possibility that the increase of the wing vibration speed is more significant.

Therefore, the evaluation signal determination unit 114 shown in FIG. 8 has a function of deciding whether to use the generated power and/or the wing vibration speed as the evaluation signal based on the azimuth angle and/or the wing vibration speed. For example, under the azimuth angle near the tower, the wing vibration speed can be used as the evaluation signal 112. In addition, when the wing vibration speed is above an arbitrary threshold, the wing vibration speed can be used as the evaluation signal 112, and if it is below the threshold, only the generated power can be used as the evaluation signal 112. In this way, in the learner 102, the influence of the blade vibration speed can be considered, and besides the power generation power can be increased, the load of the blade can be optimized.

Based on the state signal 111 and the evaluation signal 112, in the learner 102, through reinforcement learning (Reinforcement Learning), a function of outputting the state signal 111 and improving the pitch angle command value of the evaluation signal 112 is learned. Reinforcement learning is, for example, a learning method that uses a state signal obtained from an environment such as a control object such as a wind power generation system, and generates an operation signal to the environment in such a manner that the expected value of the evaluation signal obtained from the current state to the future becomes maximum. In addition, for reinforcement learning algorithms, for example, well-known techniques such as Q learning, deep reinforcement learning, and actor-critic can be used.

In addition, when the analysis model 101 is in the form of a function, the function of the analysis model 101 and/or the function learned by reinforcement learning in the learner 102 may also be updated. In addition, when the analysis model 101 is a table reference type, the function learned by reinforcement learning in the learner 102 is updated, and the pitch angle command value output from the learner is added as a correction value to the self-analysis model 101 output The pitch angle command value is calculated as the final pitch angle command value. Thus, even when the analysis model 101 is accompanied by a numerical error caused by analysis, the pitch angle command value can be corrected by the learner 102 based on actual measurement data, thereby controlling the pitch angle to be considered for deformation.

In summary, the amount of deformation of the first embodiment of the present invention shown in FIG. 8 takes into account the five of the wind speed, rotation speed, yaw error, cabin tilt angle, and azimuth angle as the learner 102 of the pitch angle command value calculation unit 100 The parameters are basically composed of the model of each time (referred to as a time-sharing model) by the processing of FIG. 7 using five input parameters at each time, obtaining the amount of deformation considering the pitch angle command value, and on the other hand, using the evaluation The evaluation signal 112 determined by the signal determination unit 114 is selected to select the time-sharing model. After that, a learning model corresponding to the analysis model 101 is formed according to the selected time-sharing model group deemed appropriate. This series of processing is realized using the so-called reinforcement learning thinking method.

In FIG. 20, a series of processes related to the learner 102 are described in the form of a flowchart. In the processing step S200 which is the initial processing of FIG. 20, the parameters of wind speed, rotation speed, yaw error, cabin tilt angle, and azimuth angle are input. Then, in the processing step S201, the evaluation signal decided by the evaluation signal decision unit 114 is input. In addition, as its premise, the calculation processing of the input of the azimuth angle, the generated power, the wing deformation amount, and the wing vibration speed is completed to complete the determination signal evaluation process.

In the processing step S202, the operation method of the pitch angle is learned, but the processing content here is equivalent to the previously described "taking five of wind speed, rotation speed, yaw error, cabin tilt angle, and azimuth angle as input parameters, basically The process of FIG. 7 using five input parameters at each time constitutes a model (called a time-sharing model) at each time, obtains the amount of deformation considering the pitch angle command value, and, on the other hand, uses the evaluation signal determination unit The evaluation signal 112 decided by 114 chooses the above time-sharing model. After that, a learning model corresponding to the analysis model 101 is formed according to the selected time-sharing model group deemed appropriate." Therefore, according to this processing, correction information S10 corresponding to the difference between the pitch angle command value obtained by learning at the same input and the pitch angle command value obtained by the analysis model 101 is obtained.

In the processing step S203, it is determined whether the analysis model 101 is in the form of a function, and the processing moves to the processing of the processing step S204 when the function is not, and the processing to the processing step S207 when it is not the function.

In the processing step S204, the function of the analysis model 101 is updated with the correction information S10 corresponding to the difference between the pitch angle command value obtained by learning and the pitch angle command value obtained by the analysis model 101, in the processing step The pitch angle command value obtained by the analysis model 101 is output in S205, and as a result, the pitch angle is controlled in the processing step S206.

In the processing step S207, the model is constructed by the learner 102, and the pitch angle command value is output from the model in the learner 102 in the processing step S208 (the difference between the output of the learner and the analysis model is output in this case), in the processing step The pitch angle command value obtained by the analysis model 101 is added in S209 and output, and as a result, the pitch angle is controlled in processing step S206.

In addition, if the operation is performed only in a low wind speed region from the wind speed Vin to the wind speed Vd, the pitch angle command value calculation unit 100 can adjust the pitch angle by considering the deformation amount of FIG. 5 only. In addition, when considering the operation in a wider wind speed region, as shown in FIG. 9, by adding the command value calculated by the deformation-considered pitch angle command value calculation unit 100 to the wind power generation system 1 The command value of the pitch angle control unit 22 determines the final pitch angle command value. Here, the pitch angle control unit 22 may add the command value from the rotation speed control unit 22a and the command value from the torque control unit 22b to calculate the pitch angle command value, or it may be calculated based only on the rotation speed control unit 22a Pitch angle command value. 9 is a block diagram showing the outline of the processing of the operation control unit installed in the wider wind speed region of the controller 11 of the first embodiment of the present invention.

As described above, the learner 102 in the case shown in FIG. 8 obtains the pitch angle command value considering the amount of deformation by learning, and its use method can also be used to correct the analysis model 101 as shown in the condition S10 as described above The model characteristic method can also be used in the following form as shown in condition S11, that is, the difference between the output of the analysis model 101 in the initial state and the output of the analysis model 101 is used as the learner 102 after learning Output.

FIG. 10 is a flowchart for preparing an analysis model 101 of the generator torque corresponding to the deformation considering the pitch angle in advance.

In FIG. 10, parameters of wind speed, rotation speed, yaw error, cabin tilt angle, azimuth angle, and pitch angle are input in processing step S106. In the processing step S107, the generated power and the generator torque are calculated.

In the processing step S108, it is determined whether the available measurement information is included. In the case of not being covered, the processing of the processing step S106 and the processing step S107 is executed again. In this way, the generator torque is explored for the parameters that include the operating state of the windmill. In the processing step S109, the generator torque corresponding to the deformation amount considering the pitch angle may be additionally stored in the analysis model storing the above-mentioned pitch angle for the parameter including the operating state of the windmill, thereby creating the analysis model 108.

FIG. 11 is a diagram for explaining the relationship between the angle of attack, the pitch angle, and the initial twist angle of the relative wind speed in the wind power generation system 1 relative to the air element flowing into a certain blade profile. As shown in FIG. 11, the rotation speed ω generated by the rotation of the wing 115 and the relative wind speed W caused by the wind speed V flow into the wing 115. The angle formed by the rotation surface of the wing 115 and the chord length is the sum of the pitch angle θp and the initial twist angle θs of the blade (θp+θs). The angle formed by the relative wind speed W and the chord length of the wing becomes the angle of attack α0 of the wing.

，翼115之迎角變化為α1。In addition, FIG. 12 is a diagram illustrating the relationship between the amount of deformation in FIG. 11. When the amount of deformation occurs in the wing 115, as shown in FIG. 12, the change in relative wind speed is added, and the angle formed by the rotation surface and the chord length of the wing is added with twist

, The angle of attack of the wing 115 changes to α1.

FIG. 13 is a graph showing the relationship between the angle of attack of a blade profile and the aerodynamic performance (profile aerodynamic performance) in the case where the amount of deformation occurs in the wind power generation system 1. Therefore, as shown in FIG. 13, the aerodynamic performance of the profile of the wing 115 is reduced because it changes from α0 to α1. Here, the profile aerodynamic performance is the quotient of the lifting force of the wing 115, the torque and force applied in the direction of rotation, or the force applied in the direction of rotation of the wing 115 and the force applied in the direction of thrust.

FIG. 14 is a diagram showing an example of the relationship between the wind speed and the force applied in the direction of rotation when the amount of deformation of the first embodiment of the present invention is applied considering the pitch angle command value calculation unit 100 and the case where it is not applied. As shown in FIG. 14, the force 117 applied in the rotation direction after the database application can be increased by about 10% on average compared to the force 116 applied in the rotation direction before the database application.

In the case of applying the first embodiment of the present invention, it is possible to control the pitch angle to be considered as the amount of deformation, and it is possible to update or correct the pitch angle command value by utilizing the actual machine operation data to achieve an increase in generated power or a reduction in load.

The physical significance and effects of the analysis model 101 or the learner 102 that constitute the analysis model 101 or the learner 102 in the pitch angle command value calculation unit 100 as described above are summarized as follows.

By using the output of the wind speed measuring unit 103 and the rotation speed measuring unit 104, the pitch angle can be adjusted to maximize the aerodynamic performance in consideration of the amount of deformation (the wing deformation considers the pitch angle).

Furthermore, by utilizing the yaw error measurement unit 105, it is possible to adjust the pitch angle in consideration of the wing deformation generated when the wind power generation system 1 is not facing the wind direction.

In addition, by using the nacelle inclination angle measuring unit 106, when the wind power generation system 1 is installed on a floating body, it is possible to adjust the wing deformation generated when the rotor 4 is tilted forward and backward in consideration of the pitch angle.

Since the wind speed measuring section 103 measures the wind speed near the nacelle 5, it is impossible to consider the effect of higher wind speed at higher altitudes called wind shear. Furthermore, the effect of reducing the wind speed in the vicinity of the tower after passing through the tower, which is called the tower shadow effect, in the downwind mode windmill cannot be considered. Therefore, by utilizing the azimuth measurement unit 107, it is possible to consider the change in wind speed in one rotation period caused by wind shear and the tower shadow effect. Therefore, it can be adjusted to the wing deformation in one rotation cycle considering the pitch angle. In this case, it is possible to perform independent pitch control on each blade.

The wing deformation amount measuring unit 109 can be measured by, for example, an optical fiber sensor or a strain gauge. As a specific example of the optical fiber sensor, the technology described in Japanese Patent Laid-Open No. 2018-145899 can be used. The light is irradiated from the light source, and in the optical fiber sensor disposed in the blade, light having a wavelength corresponding to the change in the strain of the blade is reflected to the detector via the optical cable. The detector detects the wavelength of the reflected light transmitted, and the detected reflected light can be converted into a strain corresponding to the wavelength by converting the light intensity into strain.

In addition, here, the wind speed measurement unit 103, the rotation speed measurement unit 104, the yaw error measurement unit 105, the cabin inclination angle measurement unit 106, the generated power measurement unit 108, and the wing deformation measurement unit 109 are input to the deformation amount considering the pitch The value of the angle command value calculation unit 100 may coincide with the output signal of each measurement unit, or it may be a value that is implemented with a filter process that sets a specific time constant.

Furthermore, when configuring the operation control unit of the present invention, it is actually more practical to use a computing system and realize it by software. Therefore, the function of the variable speed control unit 21 (pitch angle control unit 22, generator torque control unit 23) in the controller 11 of FIG. 2 or the deformation amount of FIG. 5 considers the function of the pitch angle command value calculation unit 100, etc. By computer system, using software. In the embodiment, the software about a part of the main functions is illustrated in FIG. 7, FIG. 8, FIG. 9, FIG. 10, and so on. Although not all exemplified, of course, the parts not exemplified here are also implemented by software. In addition, the part of the measuring device or the conversion function that processes its output can also be processed by software. [Example 2]

Using FIG. 15, a wind power generation system according to Embodiment 2 of the present invention will be described. In addition, with regard to points overlapping with the first embodiment, detailed description is omitted.

15 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 2 of the present invention. Unlike the first embodiment, the measurement signal 110 includes an azimuth angle correction value generation unit 200, and adopts a configuration in which the azimuth angle correction value is input to the deformation amount consideration pitch angle command value calculation unit 100 to calculate the pitch angle command value.

The azimuth correction value generating unit 200 calculates the self-pitch by adding the value obtained by multiplying the time constant of the pitch angle driving device by the rotation speed measured by the rotation speed measurement unit 104 to the value from the azimuth angle measurement unit 107 The angle command value is calculated as the azimuth angle where the influence of the phase advance of the azimuth angle until it actually reaches the pitch angle command value is corrected. At this time, the value of 2 times or 3 times the time constant can be multiplied by the rotation speed.

In the case of applying the second embodiment of the present invention, by correcting the deviation between the pitch angle command value at the azimuth angle measurement time and the pitch angle command value calculated at the time when the actual pitch angle command value is reached, the power generation efficiency can be suppressed Of reduction. [Example 3]

Embodiment 3 of the present invention will be described using FIG. 16. In addition, with regard to points overlapping with Example 1 and Example 2, detailed description is omitted.

16 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 3 of the present invention. Unlike Embodiment 1 and Embodiment 2, it has an inverse model 300 of a pitch angle driving device. By adding the pitch angle command value corrected by the inverse model to the pitch angle command value in Embodiment 1 and Embodiment 2, Determine the final pitch angle command value. The inverse model is made by analyzing and machine learning based on the data of past movements to find the inverse transfer function of the pitch angle driving device.

In the case of applying the third embodiment of the present invention, the inverse model is used to correct the calculation time from the pitch angle command value until the time when the pitch angle command value is actually reached due to the change of the measurement information, and the measurement information according to the arrival time The difference between the calculated pitch angle command value and the actual pitch angle at the time of arrival can suppress the decrease in power generation efficiency. [Example 4]

The operation control unit of the wind power generation system according to Embodiment 4 of the present invention will be described using FIG. 17. The pitch angle operation control unit of the fourth embodiment of the present invention is the same as that of the first to third embodiments, so the description is omitted.

In Embodiment 4, different from Embodiments 1 to 3, it includes a pitch angle measurement unit 400. In addition to the above measurement units, based on the measurement information of the pitch angle measurement unit 400, the generator torque command is calculated from the analysis model 101 value.

In the learner 102, in addition to the measurement unit, based on the measurement signal measured by the pitch angle measurement unit 400 and the evaluation signal, a function of outputting the generator torque command value is updated to correct the generator torque command value.

Furthermore, the generator torque command value may use only the value calculated from the self-analysis model 101, or may add the value calculated from the self-analysis model 101 to the value calculated from the generator torque control unit 23 to determine the final Generator torque command value.

In the case of applying the fourth embodiment of the present invention, by adjusting the generator torque corresponding to the operation at a pitch angle command value in consideration of the deformation amount, it is possible to suppress the decrease in power generation efficiency. [Example 5]

The operation control unit of the wind power generation system according to Embodiment 5 of the present invention will be described using FIG. 18. Different from Embodiment 1 to Embodiment 4, it has a mode switching function 500 capable of switching between the power generation operation mode and the power generation standby mode, thereby changing the amount of deformation and considering the pitch angle based on the measurement information based on the measurement information during the power generation operation and the power generation standby .

In addition, the rotation of the rotor or the generator is stopped during the standby of power generation, and therefore, the rotation speed measurement unit and the generated power measurement unit are not necessary for the measurement information.

During power generation standby, the pitch angle that minimizes the force applied in the thrust direction is referenced according to the analysis model. In the learner 102, a function of outputting a pitch angle command value for improving the wing vibration speed is updated based on the state information of the measurement information and the wing vibration speed evaluation signal.

In the case of applying the fifth embodiment of the present invention, the load applied to the blade can be reduced by controlling the pitch angle in consideration of the deformation amount during power generation standby. [Example 6]

The operation control unit of the wind power generation system according to Embodiment 6 of the present invention will be described using FIG. 19. In addition, the points that overlap with Examples 1 to 5 will not be described in detail.

In the sixth embodiment, the addition unit of the command value from the deformation-considered pitch angle command value calculation unit 100 and the command value of the pitch angle control unit 22 includes a weighted calculation unit 600 to perform weighted calculation. This can control the pitch angle in the vicinity of the rated operation where the wind speed at which the input of energy generated by the wind is adjusted to be higher is near the pitch angle in consideration of the deformation, thereby suppressing the reduction of the generated power. The addition method at this time is determined by formula (1), for example.

[Number 1]

Pitch angle command Here, k is a weighting coefficient, θ _{ip _ dem} i-th blade to the final decision value, θ _{p _ conv} pitch-angle command to the control unit of value, θ _{ip _ opt} for the pitch angle from a consideration of the amount of deformation The command value of the command value calculation unit 100. k is the wind speed V ₁ that starts to increase using the measured wind speed V, the pitch angle, and the increase rate with respect to the generated power generated by the conventional control, and is the minimum wind speed V ₂ that is derived by the following formula. Furthermore, V ₁ and V ₂ are calculated based on the performance evaluation result in advance.

[Number 2]

With the derived k, the pitch angle is controlled to be deformed when the wind speed V ₁ is not reached. In addition, when the wind speed V _{1 is} higher than the wind speed V ₂ , the weight k self-deformation is added to the existing command value in consideration of the pitch angle, and when the wind speed is V ₂ or higher, the control is performed to the existing command value.

Through the above embodiments, in the present invention, it is possible to realize "a wind power generation system characterized by including a deformation amount considering a pitch angle command value calculation unit that calculates a pitch angle command value considering the blade deformation amount; The amount of deformation is given to the pitch angle drive device to change the pitch angle in consideration of the pitch angle command value of the pitch angle command value calculation unit."

Furthermore, the estimation method of the pitch angle can realize "based on the wind speed, the rotation speed of the rotor or generator, the azimuth state signal, the azimuth angle and/or the wing vibration speed, the determined power generation and/or wing vibration speed can be evaluated The signal is input to the learner, and the learner estimates and improves the pitch angle command value of the evaluation signal relative to the status signal, and controls the pitch angle based on the estimated pitch angle command value."

In addition, as a method of using the learner, a method of correcting the model characteristics of the analysis model set to the initial state, a method of correcting the difference from the analysis model, etc. can be applied.

In addition, the method of using the deformation angle determined by a learner or the like in consideration of the pitch angle command value may be directly given to the pitch angle driving device 8 of the blade 2 or may be given in the form of being added to the output of the controller 11.

In addition, when considering the pitch angle command value calculation unit 100 for the structural deformation, various configurations can be adopted, and various correspondences can be adopted according to the current situation.

1: Wind power generation system 2: blade 3: wheels 4: rotor 5: cabin 6: Generator 7: Wind direction and speed sensor 8: pitch angle drive device 9: Tower 10: Power converter 11: Controller 12: Rotation speed sensor 21: Variable speed control section 22: Pitch angle control unit 22a: Rotation speed control section 22b: Torque Control Department 23: Generator Torque Control Department 100: Deformation considering the pitch angle command value calculation unit 101: Analysis model 102: Learner 103: Wind Speed Measurement Department 104: Rotation speed measurement section 105: Yaw Error Measurement Department 106: Engine room tilt angle measurement department 107: Azimuth measurement department 108: Power generation measurement department 109: Wing deformation measurement section 110: measuring signal 111: status signal 112: Evaluation signal 113: Wing vibration speed calculation unit 114: Evaluation signal decision department 115: Wing 116: Wing deformation considers the force applied in the direction of rotation before the application of the pitch angle command value calculation unit 117: The wing deformation considers the force applied in the direction of rotation after the application of the pitch angle command value calculation unit 200: Azimuth correction value generation unit 300: Inverse model 400: Pitch angle measuring section 500: Mode switching section 600: Weighted operation section Qrat: rated generator torque S10: Conditions S11: Conditions S101～S109: Procedure S200～S206: Steps V: wind speed Va: wind speed Vb: wind speed Vc: wind speed Vd: wind speed Vin: cut into the wind speed Vout: Cut out the wind speed W: relative wind speed Wrat: Rated rotation speed α0: angle of attack α1: angle of attack θmax: feather side θmin: reverse propeller side θp: pitch angle θs: initial twist angle ω: rotation speed

FIG. 1 is a diagram showing an example of a general configuration of a general wind power generation system to which the present invention can be applied. FIG. 2 is a block diagram showing the outline of the processing of the operation control unit installed in the controller 11 of the wind power generation system 1. FIG. 3 is a block diagram showing the outline of the pitch angle control unit 22 in the variable speed control unit 21. 4 is a schematic diagram showing the relationship among wind speed, generated power, rotation speed, generator torque, and pitch angle of the wind power generation system 1. FIG. 5 is a block diagram of the deformation amount considering the pitch angle command value calculation unit 100 according to Embodiment 1 of the present invention. 6 is a block diagram showing the relationship between the analysis model 101 and its input in the first embodiment of the present invention. FIG. 7 is a diagram showing a flow for preparing the analysis model 101 of the first embodiment of the present invention in advance. FIG. 8 is a block diagram showing the relationship between the input of the learner 102 of the deformation amount considering the pitch angle command value calculation unit 100 of the first embodiment of the present invention and its input. FIG. 9 is a block diagram showing the outline of the processing of the operation control unit installed in the controller 11 of the first embodiment of the present invention in a wider wind speed region. FIG. 10 is a flowchart for preparing an analysis model 101 of the generator torque corresponding to the deformation considering the pitch angle in advance. FIG. 11 is a diagram for explaining the relationship between the angle of attack, the pitch angle, and the initial twist angle of the relative wind speed with respect to the wing element flowing into a blade section in the wind power generation system 1. FIG. FIG. 12 is a diagram for explaining the relationship between the angle of attack, the pitch angle, the initial twist angle, and the wing deformation of the relative wind speed of the wing element flowing into a blade profile in the wind power generation system 1. FIG. FIG. 13 is a diagram for explaining the degradation of aerodynamic performance in a blade section caused by blade deformation in the wind power generation system 1. FIG. 14 is a diagram for explaining the relationship between the wind speed and the force applied in the direction of rotation when the operation control unit of Embodiment 1 of the present invention is applied and when it is not applied. 15 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 2 of the present invention. 16 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 3 of the present invention. 17 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 4 of the present invention. 18 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 5 of the present invention. 19 is a block diagram showing the operation control unit of the wind power generation system according to Embodiment 6 of the present invention. 20 is a flowchart showing a series of processes related to the learner 102.

100: Deformation considering the pitch angle command value calculation unit

101: Analysis model

102: Learner

103: Wind Speed Measurement Department

104: Rotation speed measurement section

105: Yaw Error Measurement Department

106: Engine room tilt angle measurement department

107: Azimuth measurement department

108: Power generation measurement department

109: Wing deformation measurement section

110: measuring signal

S10: Conditions

S11: Conditions

Claims (15)

A wind power generation system characterized by comprising: a plurality of blades, which can be changed by a pitch angle driving device; a rotor, which receives the wind on the blades to rotate; and a generator, which utilizes the rotational energy of the rotor Generate electricity; and A pitch angle command value calculation unit that calculates a pitch angle command value that takes into account the amount of deformation of the blade is provided, and the pitch angle command value of the deformation amount consideration thread pitch command value calculation unit is given to the pitch angle drive device and Change the pitch angle. The wind power generation system according to claim 1, wherein the deformation amount considering pitch angle command value calculation unit includes: an analysis model that models the relationship between the measurement signal in the wind power generation system and the pitch angle command value considering the deformation amount; And a learner, which uses the learning of the measurement signal in the wind power generation system to obtain a pitch angle command value that takes into account the deformation amount. The wind power generation system according to claim 2, wherein the characteristics of the analysis model are corrected according to the learning experience of the learner, and the pitch angle command value is given to the pitch angle driving device, and the pitch angle command value is for the corrected The analysis model gives the pitch angle command value for the above measurement signal in the wind power generation system. The wind power generation system according to claim 2, wherein the pitch angle driving device is given a sum of the pitch angle command value from the analysis model and the pitch angle command value from the learner. The wind power generation system according to any one of claims 1 to 4, which includes a pitch angle control unit that sets a pitch angle command value for controlling the generator torque and the generator rotation speed to their respective target values, The pitch angle command value from the pitch angle control unit and the pitch angle command value from the deformation amount considering pitch angle command value calculation unit are given to the pitch angle driving device to change the pitch angle. The wind power generation system according to any one of claims 2 to 4, wherein the deformation amount takes into account the pitch angle command value calculation unit based on the wind speed, the rotational speed of the rotor or generator, the azimuth angle state signal, the azimuth angle and/or the wing Vibration speed, input the generated power and/or wing vibration speed determined as the evaluation signal to the above learner; The learner estimates to improve the pitch angle command value relative to the evaluation signal of the status signal, and controls the pitch angle based on the estimated pitch angle command value. The wind power generation system according to claim 6, wherein the deformation amount considering pitch angle command value calculation unit outputs the pitch angle control information from the analysis model based on the state signal, and the analysis model memorizes the previously calculated Pitch angle control information; The pitch angle control information of the analysis model is improved or the analysis model is modified by the learner, whereby the pitch angle is controlled based on the calculated pitch angle command value. According to the wind power generation system of claim 6 or 7, wherein the above-mentioned deformation amount takes into account the fact that the pitch angle command value calculation unit is above the first wind speed at which power generation starts and below the second wind speed at rated power generation, based on The pitch angle command value calculated in such a manner as to maximize the force or torque applied in the rotation direction of the blade, controls the pitch angle. The wind power generation system according to any one of claims 6 to 8, which includes a feedback control unit that uses the deviation between the target value of the rotation speed of the rotor or generator and the measured value as an input value to calculate the pitch Angle control information; The pitch angle command value is calculated based on the pitch angle command value calculation section of the pitch angle command value calculation section and the pitch angle control information of the feedback control section, and the pitch angle is controlled by the calculated pitch angle command value. If the wind power generation system according to any one of claims 6 to 9 has a feedback control unit, the feedback control unit uses the deviation between the target value of the generator torque and the measured value as the input value, and calculates the pitch angle control information; The pitch angle is controlled by the calculated pitch angle command value based on the pitch angle command value of the pitch angle command value calculation unit from the deformation amount and the pitch angle control information of the feedback control unit. For the wind power generation system according to any one of claims 6 to 10, the value obtained by multiplying the rotation speed of the rotor or generator by the time constant of the pitch angle drive device is added to the azimuth angle, and the added value is used as the above Analyze the status signals of the model or/and the learner. The wind power generation system according to any one of claims 6 to 11, which adjusts the pitch angle based on the modified pitch angle command value, which is made by the inverse transfer function of the pitch angle driving device It is obtained by correcting the above-mentioned pitch angle command value against the model. The wind power generation system according to claim 9 or 10, which is based on the pitch obtained by calculating the pitch angle control information of the analysis model or/and the learner and the pitch angle control information of the feedback control section by the weighting calculation section The angle command value controls the pitch angle. The wind power generation system according to any one of claims 7 to 13, which has a mode switching section for switching between a power generation operation mode and a power generation standby mode, and controls the pitch angle based on the pitch angle control information, which is included in the above power generation standby In the mode, it is obtained from the analysis model based on the measurement information of the wind speed and calculated in such a way as to minimize the load in the thrust direction of the blade. The wind power generation system according to any one of claims 1 to 14, wherein The above-mentioned blade has a flexible structure.

Images