WO2022153525A1 - Wind direction correction device, model generation device, correction method, model generation method, and program - Google Patents

Wind direction correction device, model generation device, correction method, model generation method, and program Download PDF

Info

Publication number
WO2022153525A1
WO2022153525A1 PCT/JP2021/001429 JP2021001429W WO2022153525A1 WO 2022153525 A1 WO2022153525 A1 WO 2022153525A1 JP 2021001429 W JP2021001429 W JP 2021001429W WO 2022153525 A1 WO2022153525 A1 WO 2022153525A1
Authority
WO
WIPO (PCT)
Prior art keywords
wind
model
wind direction
correction
nacelle
Prior art date
Application number
PCT/JP2021/001429
Other languages
French (fr)
Japanese (ja)
Inventor
健太郎 犬童
Original Assignee
株式会社ユーラステクニカルサービス
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 株式会社ユーラステクニカルサービス filed Critical 株式会社ユーラステクニカルサービス
Priority to PCT/JP2021/001429 priority Critical patent/WO2022153525A1/en
Priority to JP2022575029A priority patent/JP7408846B2/en
Publication of WO2022153525A1 publication Critical patent/WO2022153525A1/en

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
    • F03D—WIND MOTORS
    • F03D7/00—Controlling wind motors 
    • F03D7/02—Controlling wind motors  the wind motors having rotation axis substantially parallel to the air flow entering the rotor
    • F03D7/04—Automatic control; Regulation
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01P—MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION, OR SHOCK; INDICATING PRESENCE, ABSENCE, OR DIRECTION, OF MOVEMENT
    • G01P13/00—Indicating or recording presence, absence, or direction, of movement
    • G01P13/02—Indicating direction only, e.g. by weather vane
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/70—Wind energy
    • Y02E10/72—Wind turbines with rotation axis in wind direction

Definitions

  • the present invention relates to a wind direction correction device, a model generation device, a correction method, a model generation method, and a program.
  • Wind power generation is one of the renewable energies.
  • Patent Document 1 describes that the yaw turning mechanism of the nacelle is controlled by using the detection result of the wind direction and speed sensor. Further, Patent Document 1 also describes that the degree of wind turbulence is calculated using the detection result of the wind direction and speed sensor, and the yaw turning mechanism of the nacelle is controlled by using this degree of turbulence.
  • the wind direction measurement result by the weather vane contains errors due to various causes. Therefore, it is difficult to accurately control the direction of the nacelle according to the wind direction.
  • An example of an object of the present invention is to accurately control the direction of the nacelle according to the wind direction.
  • the present invention is a correction device for correcting the measurement result of the weather vane provided in the wind power generation device.
  • the measurement items of the weather vane include the relative wind direction, which is the direction of the wind with respect to the wind power generator, and the wind speed.
  • a correction amount determining unit that determines a correction parameter for correcting at least one of the relative wind direction and the absolute wind direction using the air density parameter, the wind speed, and the direction of the nacelle.
  • the correction amount determining unit provides a wind direction correction device that determines the correction parameters using a conversion model that converts input data including the air density parameter, the wind speed, and the direction of the nacelle into the correction parameters.
  • a model generation device that generates the above-mentioned conversion model.
  • a base model creation unit that creates a base model using the output power of the wind power generator as output data, with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the input data.
  • a transformation model generator that generates the transformation model using the base model, A model generator is provided.
  • the wind direction correction method performed by the wind direction correction device described above the model generation method performed by the model generation device described above, the program for realizing the wind direction correction device described above, and the model generation device described above.
  • a program is also provided to realize this.
  • the direction of the nacelle can be accurately controlled according to the wind direction in front of the wind turbine.
  • (A) and (B) are diagrams for explaining the reason why an error occurs in the relative wind direction measured by the weather vane meter. It is a figure which shows an example of the functional structure of a model generator. It is a figure explaining the machine learning for generating a base model. It is a figure for demonstrating an example of the method of generating a transformation model from a base model. It is a figure for demonstrating an example of the method of generating a transformation model from a base model. It is a figure for demonstrating an example of the constant calculation method used by a wind direction correction apparatus.
  • FIG. 1 is a diagram for explaining a usage environment of the wind direction correction device 50 according to the present embodiment.
  • the wind direction correction device 50 corrects the measurement result of the wind direction meter 30 attached to the wind power generation device 10.
  • the wind direction meter 30 measures the wind direction (hereinafter referred to as relative wind direction) and the wind speed with reference to the wind power generation device 10. Then, the yaw control device 60 of the wind power generation device 10 controls the direction of the nacelle 70 in the yaw direction so that the nacelle 70 faces the wind by using the relative wind direction measured by the wind direction meter 30. This is to increase the power generation efficiency of the power generation device 72 of the nacelle 70.
  • the wind direction correction device 50 performs a process for bringing the relative wind direction closer to an appropriate value. Specifically, the wind direction correction device 50 acquires the air density parameter (for example, outside temperature) related to the air density around the wind power generation device 10, the relative wind direction measured by the wind direction correction device 50, and the direction of the nacelle 70. , These data are used to determine the correction parameters for correcting at least one of the relative wind direction and the absolute wind direction.
  • the air density parameter for example, outside temperature
  • the wind direction correction device 50 uses a model (hereinafter referred to as a conversion model) that converts input data into correction parameters when determining correction parameters.
  • the input data includes air density parameters, wind speed, and nacelle direction.
  • the conversion model is generated by the model generation device 20.
  • the wind direction correction device 50 acquires the conversion model and / or the update information of the conversion model from the model generation device 20.
  • the model generation device 20 generates a conversion model using, for example, a model generated by machine learning (hereinafter referred to as a base model). When there are a plurality of wind power generation devices 10, the base model and the conversion model are generated for each of the plurality of wind power generation devices 10.
  • an example of the air density parameter is air temperature (outside air temperature). This air temperature is measured by a thermometer 40 attached to the outer surface of the wind power generator 10. Then, the wind direction correction device 50 acquires measurement data from the thermometer 40.
  • the air density parameter may further include at least one of humidity and air pressure.
  • the thermometer 40 also has a function of measuring humidity and / or atmospheric pressure.
  • the wind direction correction device 50 is mounted on the nacelle 70 of the wind power generation device 10.
  • the wind direction correction device 50 may be mounted on a portion of the wind power generation device 10 other than the nacelle 70, or may be provided outside the wind power generation device 10.
  • FIG. 2 is a diagram for explaining the reason why the relative wind direction measured by the weather vane 30 is not an appropriate value.
  • the wind direction meter 30 is attached to the nacelle 70 so that the rotating body axis (rotor) of the wind power generation device 10 is parallel to the wind direction in front of the wind power generation device 10 and the relative wind direction is 0 °.
  • the yaw control device 60 shown in FIG. 1 controls the yaw direction so that the relative wind direction measured by the weather vane 30 becomes 0 °.
  • the relative wind direction may be 0 ° when the rotor of the wind power generation device 10 and the wind direction in front of the wind power generation device 10 are not parallel.
  • misalignment when the weathercock 30 is attached to the nacelle 70.
  • the rotor of the wind power generation device 10 has a blade 74.
  • the weather vane 30 is arranged behind the blade 74. Therefore, the wind may be affected by the blade 74 before reaching the weathercock 30. This effect contributes to an error between the relative wind direction measured by the wind direction meter 30 and the relative wind direction in front of the wind power generator 10.
  • wind direction correction device 50 can simultaneously correct the error caused by the above two reasons.
  • FIG. 3 is a diagram showing an example of the functional configuration of the model generator 20.
  • the model generation device 20 includes a training data acquisition unit 220, a base model generation unit 230, a conversion model generation unit 250, and a model transmission unit 260.
  • the training data acquisition unit 220 acquires a plurality of training data.
  • Each of the plurality of training data includes wind speed, relative wind direction, air density parameter, nacelle direction, and output power of the generator 72.
  • the training data acquisition unit 220 acquires training data from the training data storage unit 210.
  • the training data storage unit 210 uses the actual data of the wind power generation device 10 (that is, the data obtained while the wind power generation device 10 is generating power) as the training data.
  • the training data storage unit 210 stores training data for each of the plurality of wind power generation devices 10. This training data is preferably data when the wind speed is in a predetermined range (for example, a range in which linear approximation can be performed in FIG. 7 described later).
  • the base model generation unit 230 generates a base model by performing machine learning using a plurality of training data acquired by the training data acquisition unit 220. As described with reference to FIG. 1, the base model is used in generating the transformation model. As shown in FIG. 4, in this machine learning, the explanatory variables include the wind speed (v), the air density parameter (T), the relative wind direction ( ⁇ ), and the nacelle direction ( ⁇ ), and the objective variable is the output. Power (P). Then, the base model outputs the output power (P) when the wind speed (v), the air density parameter (T), the relative wind direction ( ⁇ ), and the direction of the nacelle ( ⁇ ) are input. The base model generation unit 230 stores the generated base model in the model storage unit 240.
  • the base model generation unit 230 may generate a base model by using a method other than machine learning, for example, a physical model or a model that approximately represents a phenomenon.
  • the conversion model generation unit 250 uses the base model generated by the base model generation unit 230 to generate a conversion model or data for updating the conversion model (hereinafter referred to as update data). Details of the processing performed by the conversion model generation unit 250 will be described later with reference to other figures.
  • the conversion model generation unit 250 stores the generated conversion model and update data in the model storage unit 240. When there are a plurality of wind power generation devices 10, the conversion model generation unit 250 stores the base model, conversion model, and update data to be used in the wind power generation device 10 for each of the plurality of wind power generation devices 10.
  • the model transmission unit 260 transmits the conversion model or update data generated by the model generation device 20 to the wind direction correction device 50.
  • the model transmission unit 260 reads out the conversion model and the update data from the model storage unit 240 and transmits the conversion model and the update data to the wind direction correction device 50.
  • the training data storage unit 210 and the model storage unit 240 are a part of the model generation device 20. However, the training data storage unit 210 and the model storage unit 240 may be located outside the model generation device 20.
  • FIG. 5 and 6 are diagrams for explaining an example of a method of generating a conversion model from a base model (that is, a process performed by the conversion model generation unit 250).
  • the base model outputs the output power (P) when the wind speed (v), the air density parameter (T), the relative wind direction ( ⁇ ), and the direction of the nacelle ( ⁇ ) are input. Therefore, in this base model, if the wind speed (v), air density parameter (T), and nacelle direction ( ⁇ ) are fixed and the relative wind direction ( ⁇ ) is changed, the wind speed (v) and air density parameter are changed.
  • a function of output power (P) with the relative wind direction ( ⁇ ) as a variable in (T) and the direction of the nacelle ( ⁇ ) can be obtained.
  • FIG. 5 is an example of this function.
  • ⁇ when this function takes a maximum value is a value different from 0. Then, ⁇ at this time indicates this error ( ⁇ ).
  • the conversion model generation unit 250 performs the above processing while changing each of the wind speed (v), the air density parameter (T), and the direction of the nacelle ( ⁇ ), thereby performing the wind speed (v) and the air density parameter.
  • ⁇ can be calculated for each combination of (T) and the direction ( ⁇ ) of the nacelle.
  • This result is stored in the model storage unit 240 as a conversion model.
  • the wind speed (v), the air density parameter (T), and the direction of the nacelle ( ⁇ ) can be converted to ⁇ .
  • FIG. 6 shows the relationship between the wind speed (v) and ⁇ when the air density parameter (temperature) and the direction of the nacelle are fixed.
  • the wind direction correction device 50 uses a constant as a correction parameter when a predetermined condition is satisfied. An example of a predetermined condition will be described later.
  • the base model generation unit 230 of the model generation device 20 also calculates this constant.
  • 7 and 8 are diagrams for explaining an example of a method for calculating this constant.
  • FIG. 7 is a diagram showing the relationship between the wind speed (v) and the output power (P) of the power generation device 72.
  • the output power (P) increases as the wind speed increases, but has some variation.
  • One of the factors of this variation is that the distribution is generated in the relative wind direction ( ⁇ ).
  • the reason for this distribution is that the relative wind direction changes frequently.
  • the base model generation unit 230 obtains an average value (P v_avg ) of the output voltage at each wind speed (v) using, for example, the training data stored in the training data storage unit 210. Then, for each training data, the difference ( ⁇ P) between the output power P v ( ⁇ ) of the training data and the average value (P v_avg ) of the output power at the wind speed of the training data is calculated.
  • ⁇ P can also be regarded as a function of the relative wind direction ( ⁇ ).
  • FIG. 8 is a diagram showing the relationship between the value ( ⁇ ( ⁇ )) indicating the magnitude of ⁇ P and the relative wind direction ( ⁇ ).
  • ⁇ ( ⁇ ) is defined as (P v ( ⁇ ) / (P v_avg ) -1) ⁇ 100).
  • the base model generation unit 230 generates data in which x is the relative wind direction ( ⁇ ) and y is ⁇ ( ⁇ ) for each training data, and by processing these data, ⁇ (for each relative wind direction ( ⁇ )). Calculate the average value of ⁇ ). Then, the base model generation unit 230 sets the relative wind direction ( ⁇ ) when ⁇ ( ⁇ ) is maximized as a constant used as a correction parameter.
  • the constant calculated by the base model generation unit 230 is stored in the model storage unit 240. Then, the conversion model generation unit 250 transmits this constant to the wind direction correction device 50 together with the conversion model or its update data.
  • FIG. 9 is a diagram showing an example of the functional configuration of the wind direction correction device 50.
  • the wind direction correction device 50 includes a performance acquisition unit 530 and a correction amount determination unit 540.
  • the achievement acquisition unit 530 acquires the achievement data.
  • the actual data is data on the wind power generator 10 at that time, and includes the above-mentioned air density parameters (including temperature, for example), the wind speed measured by the weather vane 30, and the direction of the nacelle 70.
  • the air density parameter is generated, for example, by the thermometer 40, and the direction of the nacelle 70 is generated, for example, by the yaw controller 60.
  • the correction amount determination unit 540 determines the above-mentioned correction parameters using the air density parameter, the wind speed, and the direction of the nacelle acquired by the performance acquisition unit 530.
  • the correction amount determination unit 540 uses the conversion model generated by the model generation device 20 when determining the correction parameters. This conversion model is stored in the model storage unit 520.
  • the correction amount determination unit 540 outputs the correction parameter to the yaw control device 60.
  • the yaw control device 60 corrects the relative wind direction generated by the wind direction meter 30 using this correction parameter, and controls the direction of the nacelle 70 using the corrected relative wind direction.
  • the correction amount determination unit 540 uses a constant as a correction parameter when at least one of the air density parameter, the relative wind direction, and the wind speed does not meet the standard.
  • the criteria used here are set for each of the air density parameters, the relative wind direction, and the wind speed, and these are in the range estimated to be the values in the normal state. That is, the correction amount determination unit 540 uses a constant as a correction parameter when at least one of the parameters input to the conversion model becomes an abnormal value.
  • This constant is also generated by the model generator 20 as described above. And this constant is stored in the model storage unit 520.
  • An example of such a state is a case where a sensor for generating a parameter fails.
  • the wind direction correction device 50 includes a model update unit 510.
  • the model update unit 510 acquires a conversion model and constants as correction parameters from the model generation device 20, and stores these data in the model storage unit 520. Further, the model update unit 510 acquires update data for updating the conversion model from the model generation device 20, and updates the conversion model using the update data.
  • the model update unit 510 acquires the update value of the constant as the correction parameter from the model generation device 20, the model update unit 510 rewrites the constant to this update value.
  • FIG. 10 is a diagram showing a hardware configuration example of the model generator 20.
  • the model generator 20 includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.
  • the bus 1010 is a data transmission path for the processor 1020, the memory 1030, the storage device 1040, the input / output interface 1050, and the network interface 1060 to transmit and receive data to and from each other.
  • the method of connecting the processors 1020 and the like to each other is not limited to the bus connection.
  • the processor 1020 is a processor realized by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.
  • the memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.
  • the storage device 1040 is an auxiliary storage device realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a memory card, a ROM (Read Only Memory), or the like.
  • the storage device 1040 stores a program module that realizes each function of the model generation device 20 (for example, a training data acquisition unit 220, a base model generation unit 230, a conversion model generation unit 250, and a model transmission unit 260).
  • the processor 1020 reads each of these program modules into the memory 1030 and executes them, each function corresponding to the program module is realized.
  • the storage device 1040 also functions as a training data storage unit 210 and a model storage unit 240.
  • the input / output interface 1050 is an interface for connecting the model generator 20 and various input / output devices.
  • the network interface 1060 is an interface for connecting the model generator 20 to the network.
  • This network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network).
  • the method of connecting the network interface 1060 to the network may be a wireless connection or a wired connection.
  • the model generation device 20 may communicate with the wind direction correction device 50 via the network interface 1060.
  • the hardware configuration of the wind direction correction device 50 is also the same as the hardware configuration of the model generation device 20 shown in FIG.
  • the storage device 1040 stores a program module that realizes each function of the wind direction correction device 50 (for example, a model update unit 510, a result acquisition unit 530, and a correction amount determination unit 540).
  • the storage device 1040 also functions as a model storage unit 520.
  • FIG. 11 is a flowchart for explaining the generation process of the transformation model performed by the model generation device 20.
  • the model generation device 20 performs the processing shown in this figure for each wind power generation device 10. Further, the training data stored in the training data storage unit 210 of the model generation device 20 is periodically updated. Therefore, the model generator 20 periodically performs the processes shown in this figure.
  • the training data acquisition unit 220 of the model generation device 20 reads the training data from the training data storage unit 210 (step S10).
  • the base model generation unit 230 generates a base model using this training data (step S20), and stores the base model in the model storage unit 240.
  • the conversion model generation unit 250 generates a conversion model (or update data) using the base model generated in step S20 (step S30), and transfers the generated conversion model (or update data) to the model storage unit 240. It is memorized (step S40).
  • the model transmission unit 260 transmits the conversion model (or update data) to the wind direction correction device 50 at an appropriate timing.
  • FIG. 12 is a flowchart for explaining an example of the processing performed by the wind direction correction device 50. While the wind power generation device 10 is generating power, the wind direction correction device 50 repeats the process shown in this figure in real time.
  • the performance acquisition unit 530 of the wind direction correction device 50 acquires the performance data at that time.
  • the actual data is the data related to the wind power generator 10 at that time, and the above-mentioned air density parameters (including temperature, for example), the wind speed measured by the wind direction meter 30, and the direction of the nacelle 70.
  • the correction amount determination unit 540 converts the actual data into a correction parameter for the relative wind direction using the conversion model stored in the model storage unit 520 (step S120), and outputs this correction parameter to the yaw control device 60.
  • the yaw control device 60 corrects the relative wind direction generated by the wind direction meter 30 using this correction parameter, and controls the direction of the nacelle 70 using the corrected relative wind direction.
  • FIG. 13 is a diagram showing a first modification of FIG. 12. The example shown in this figure is the same as that in FIG. 12 except for the following points.
  • the correction amount determination unit 540 When the actual data acquired by the actual acquisition unit 530 meets the criteria (step S122: Yes), the correction amount determination unit 540 generates correction parameters using the conversion model (step S124). .. On the other hand, when the correction amount determination unit 540 does not satisfy the standard (step S122: No), the correction amount determination unit 540 sets the constant stored in the model storage unit 520 as the correction parameter (step S126).
  • FIG. 14 is a diagram showing a second modification of FIG. 12.
  • the correction amount determination unit 540 corrects the relative wind direction using the correction parameter (step S132) and outputs the corrected relative wind direction to the yaw control device 60 (step S142). It is the same as FIG. Then, the yaw control device 60 controls the direction of the nacelle 70 by using the relative wind direction after the correction amount determination unit 540 corrects.
  • the correction amount determining unit 540 may correct the relative wind direction as shown in FIG.
  • the wind direction correction device 50 generates a correction parameter for correcting the relative wind direction measured by the wind direction meter 30. Therefore, the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
  • FIG. 15 is a diagram showing an example of the functional configuration of the wind direction correction device 50 according to the present embodiment.
  • the wind direction correction device 50 has a function of generating a conversion model.
  • the wind direction correction device 50 has a training data storage unit 550, a training data acquisition unit 560, a base model generation unit 570, and a conversion model generation unit 580.
  • the training data storage unit 550, the training data acquisition unit 560, the base model generation unit 570, and the conversion model generation unit 580 are the training data storage unit 210, the training data acquisition unit 220, the base model generation unit 230, and the conversion, respectively. It has the same function as the model generation unit 250.
  • the model storage unit 520 also has the function of the model storage unit 240 of FIG. 3 in addition to the functions described in the first embodiment.
  • the correction amount determination unit 540 is a constant as a conversion parameter. Is used. This constant is as described with reference to FIGS. 7 and 8.
  • the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
  • FIG. 16 is a diagram for explaining the wind direction correction device 50 according to the present embodiment.
  • the yaw control device 60 also serves as a wind direction correction device 50.
  • the wind direction correction device 50 also has a function of controlling the yaw direction of the nacelle 70 by using the relative wind direction and the correction parameters.
  • the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
  • FIG. 17 is a diagram for explaining the wind direction correction device 50 according to the present embodiment.
  • the wind direction correction device 50 is a part of the wind direction meter 30.
  • the wind direction meter 30 outputs the corrected relative wind direction.
  • the yaw control device 60 controls the direction of the nacelle 70 according to the wind direction by using the corrected relative wind direction.
  • the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
  • Wind power generation device 20
  • Model generation device 30
  • Wind direction meter 40
  • Thermometer 50
  • Wind direction correction device 60
  • Yaw control device 70
  • Nacelle 72
  • Power generation device 74
  • Blade 210
  • Training data storage unit 220
  • Training data acquisition unit 230
  • Base model generation unit 240
  • Model storage unit 250
  • Conversion Model generation unit 260
  • Model transmission unit 510
  • Model update unit 520
  • Model storage unit 530
  • Achievement acquisition unit 540
  • Correction amount determination unit 550
  • Training data storage unit 560
  • Training data acquisition unit 570
  • Base model generation unit 580 Conversion model generation unit

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Sustainable Development (AREA)
  • Sustainable Energy (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Mechanical Engineering (AREA)
  • General Engineering & Computer Science (AREA)
  • Wind Motors (AREA)

Abstract

This wind direction correction device (50) corrects a result of measurement by an anemoscope provided to a wind power generation apparatus. The items measured by the anemoscope include relative wind direction, which is the direction of wind with reference to the wind power generation apparatus, and wind speed . A past record acquisition unit (530) acquires an air density parameter related to the air density in the surroundings of the wind power generation apparatus, the wind speed measured by the anemoscope, and the direction of a nacelle of the wind power generation apparatus. A correction amount determination unit (540) determines a correction parameter for correcting the relative wind direction and/or the absolute wind direction by using the air density parameter, the wind speed, and the direction of the nacelle. When doing so, the correction amount determination unit (540) uses a conversion model in which input data including the air density parameter, the wind speed, and the direction of the nacelle is converted to the correction parameter.

Description

風向補正装置、モデル生成装置、補正方法、モデル生成方法、及びプログラムWind direction correction device, model generation device, correction method, model generation method, and program
　本発明は、風向補正装置、モデル生成装置、補正方法、モデル生成方法、及びプログラムに関する。 The present invention relates to a wind direction correction device, a model generation device, a correction method, a model generation method, and a program.
　再生エネルギーの一つに、風力発電がある。風力発電装置の発電効率を上げるためには、ナセルの向きを風向に合わせることは重要である。例えば特許文献１には、風向風速センサの検出結果を用いてナセルのヨ―旋回機構を制御することが記載されている。また、特許文献１には、風向風速センサの検出結果を用いて風の乱れ度を算出し、この乱れ度を用いてナセルのヨ―旋回機構を制御することも記載されている。 Wind power generation is one of the renewable energies. In order to improve the power generation efficiency of wind power generation equipment, it is important to match the direction of the nacelle with the wind direction. For example, Patent Document 1 describes that the yaw turning mechanism of the nacelle is controlled by using the detection result of the wind direction and speed sensor. Further, Patent Document 1 also describes that the degree of wind turbulence is calculated using the detection result of the wind direction and speed sensor, and the yaw turning mechanism of the nacelle is controlled by using this degree of turbulence.
特開２０２０－２０２６４号公報Japanese Unexamined Patent Publication No. 2020-20264
　風向計による風向の測定結果は、様々な原因に起因した誤差を含んでいる。このため、ナセルの向きを風向に合わせて精度良く制御することは難しい。 The wind direction measurement result by the weather vane contains errors due to various causes. Therefore, it is difficult to accurately control the direction of the nacelle according to the wind direction.
　本発明の目的の一例は、ナセルの向きを風向に合わせて精度良く制御することにある。 An example of an object of the present invention is to accurately control the direction of the nacelle according to the wind direction.
　本発明によれば、風力発電装置に設けられた風向計の測定結果を補正する補正装置であって、
　前記風向計の測定項目は、前記風力発電装置を基準とした風の向きである相対風向、及び風速を含んでおり、
　前記風力発電装置の周囲における空気密度に関連する空気密度パラメータ、前記風向計が測定した前記風速、及び前記風力発電装置のナセルの方向を取得する実績取得部と、
　前記空気密度パラメータ、前記風速、及び前記ナセルの方向を用いて、前記相対風向及び絶対風向の少なくとも一方を補正するための補正パラメータを決定する補正量決定部と、
を備え、
　前記補正量決定部は、前記空気密度パラメータ、前記風速、及び前記ナセルの方向を含む入力データを前記補正パラメータに変換する変換モデルを用いて、前記補正パラメータを決定する、風向補正装置が提供される。 According to the present invention, it is a correction device for correcting the measurement result of the weather vane provided in the wind power generation device.
The measurement items of the weather vane include the relative wind direction, which is the direction of the wind with respect to the wind power generator, and the wind speed.
An air density parameter related to the air density around the wind power generator, the wind speed measured by the wind direction meter, and a performance acquisition unit for acquiring the direction of the nacelle of the wind power generator.
A correction amount determining unit that determines a correction parameter for correcting at least one of the relative wind direction and the absolute wind direction using the air density parameter, the wind speed, and the direction of the nacelle.
With
The correction amount determining unit provides a wind direction correction device that determines the correction parameters using a conversion model that converts input data including the air density parameter, the wind speed, and the direction of the nacelle into the correction parameters. To.
　本発明によれば、上記した変換モデルを生成するモデル生成装置であって、
　前記風速、前記相対風向、前記空気密度パラメータ、及び前記ナセルの方向を入力データの少なくとも一部として、前記風力発電装置の出力電力を出力データとするベースモデルを作成するベースモデル作成部と、
　前記ベースモデルを用いて前記変換モデルを生成する変換モデル生成部と、
を備えるモデル生成装置が提供される。 According to the present invention, it is a model generation device that generates the above-mentioned conversion model.
A base model creation unit that creates a base model using the output power of the wind power generator as output data, with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the input data.
A transformation model generator that generates the transformation model using the base model,
A model generator is provided.
　また、本発明によれば、上記した風向補正装置によって行われる風向補正方法、上記したモデル生成装置によって行われるモデル生成方法、上記した風向補正装置を実現するためのプログラム、及び上記したモデル生成装置を実現するためのプログラムも提供される。 Further, according to the present invention, the wind direction correction method performed by the wind direction correction device described above, the model generation method performed by the model generation device described above, the program for realizing the wind direction correction device described above, and the model generation device described above. A program is also provided to realize this.
　本発明によれば、ナセルの向きを風車前方の風向に合わせて精度良く制御できる。 According to the present invention, the direction of the nacelle can be accurately controlled according to the wind direction in front of the wind turbine.
第１実施形態に係る風向補正装置の使用環境を説明するための図である。It is a figure for demonstrating the use environment of the wind direction correction apparatus which concerns on 1st Embodiment. （Ａ）及び（Ｂ）は、風向計が測定した相対風向に誤差が生じる理由を説明するための図である。(A) and (B) are diagrams for explaining the reason why an error occurs in the relative wind direction measured by the weather vane meter. モデル生成装置の機能構成の一例を示す図である。It is a figure which shows an example of the functional structure of a model generator. ベースモデルを生成するための機械学習を説明する図である。It is a figure explaining the machine learning for generating a base model. ベースモデルから変換モデルを生成する方法の一例を説明するための図である。It is a figure for demonstrating an example of the method of generating a transformation model from a base model. ベースモデルから変換モデルを生成する方法の一例を説明するための図である。It is a figure for demonstrating an example of the method of generating a transformation model from a base model. 風向補正装置が用いる定数の算出方法の一例を説明するための図である。It is a figure for demonstrating an example of the constant calculation method used by a wind direction correction apparatus. ΔＰの大きさを示す値（ε（φ））と相対風向（φ）の関係を示す図である。It is a figure which shows the relationship between the value (ε (φ)) which shows the magnitude of ΔP, and the relative wind direction (φ). 風向補正装置の機能構成の一例を示す図である。It is a figure which shows an example of the functional structure of the wind direction correction device. モデル生成装置のハードウェア構成例を示す図である。It is a figure which shows the hardware configuration example of the model generator. モデル生成装置が行う変換モデルの生成処理を説明するためのフローチャートである。It is a flowchart for demonstrating the generation process of the transformation model performed by the model generation apparatus. 風向補正装置が行う処理の一例を説明するためのフローチャートである。It is a flowchart for demonstrating an example of a process performed by a wind direction correction apparatus. 図１２の第１変形例を示す図である。It is a figure which shows the 1st modification of FIG. 図１２の第２変形例を示す図である。It is a figure which shows the 2nd modification of FIG. 第２実施形態に係る風向補正装置の機能構成の一例を示す図である。It is a figure which shows an example of the functional structure of the wind direction correction device which concerns on 2nd Embodiment. 第３実施形態に係る風向補正装置を説明するための図である。It is a figure for demonstrating the wind direction correction apparatus which concerns on 3rd Embodiment. 第４実施形態に係る風向補正装置を説明するための図である。It is a figure for demonstrating the wind direction correction apparatus which concerns on 4th Embodiment.
　以下、本発明の実施の形態について、図面を用いて説明する。尚、すべての図面において、同様な構成要素には同様の符号を付し、適宜説明を省略する。 Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all drawings, similar components are designated by the same reference numerals, and description thereof will be omitted as appropriate.
（第１実施形態）
　図１は、本実施形態に係る風向補正装置５０の使用環境を説明するための図である。風向補正装置５０は、風力発電装置１０に取り付けられた風向計３０の測定結果を補正する。 (First Embodiment)
FIG. 1 is a diagram for explaining a usage environment of the wind direction correction device 50 according to the present embodiment. The wind direction correction device 50 corrects the measurement result of the wind direction meter 30 attached to the wind power generation device 10.
　具体的には、風向計３０は、風力発電装置１０を基準とした風向（以下、相対風向と記載）、及び風速を測定する。そして風力発電装置１０のヨー制御装置６０は、風向計３０が測定した相対風向を用いて、ナセル７０が風に正対するように、ナセル７０のヨー方向の向きを制御する。これは、ナセル７０が有する発電装置７２の発電効率を上げるためである。 Specifically, the wind direction meter 30 measures the wind direction (hereinafter referred to as relative wind direction) and the wind speed with reference to the wind power generation device 10. Then, the yaw control device 60 of the wind power generation device 10 controls the direction of the nacelle 70 in the yaw direction so that the nacelle 70 faces the wind by using the relative wind direction measured by the wind direction meter 30. This is to increase the power generation efficiency of the power generation device 72 of the nacelle 70.
　風向計３０が測定した相対風向が適正でない場合、発電装置７２の発電効率は低下する。風向補正装置５０は、相対風向を適正値に近づけるための処理を行う。具体的には、風向補正装置５０は、風力発電装置１０の周囲における空気密度に関連する空気密度パラメータ（例えば外気温）、風向補正装置５０が測定した相対風向、及びナセル７０の方向を取得し、これらのデータを用いて、相対風向及び絶対風向の少なくとも一方を補正するための補正パラメータを決定する。 If the relative wind direction measured by the wind direction meter 30 is not appropriate, the power generation efficiency of the power generation device 72 will decrease. The wind direction correction device 50 performs a process for bringing the relative wind direction closer to an appropriate value. Specifically, the wind direction correction device 50 acquires the air density parameter (for example, outside temperature) related to the air density around the wind power generation device 10, the relative wind direction measured by the wind direction correction device 50, and the direction of the nacelle 70. , These data are used to determine the correction parameters for correcting at least one of the relative wind direction and the absolute wind direction.
　風向補正装置５０は、補正パラメータを決定する際に、入力データを補正パラメータに変換するモデル（以下、変換モデルと記載）を用いる。入力データは、空気密度パラメータ、風速、及びナセルの方向を含んでいる。また、変換モデルは、モデル生成装置２０によって生成されている。風向補正装置５０は、モデル生成装置２０から変換モデル及び／又は変換モデルの更新情報を取得する。なお、モデル生成装置２０は、例えば機械学習によって生成されたモデル（以下、ベースモデルと記載）を用いて、変換モデルを生成する。風力発電装置１０が複数ある場合、ベースモデル及び変換モデルは、複数の風力発電装置１０別に生成される。 The wind direction correction device 50 uses a model (hereinafter referred to as a conversion model) that converts input data into correction parameters when determining correction parameters. The input data includes air density parameters, wind speed, and nacelle direction. Further, the conversion model is generated by the model generation device 20. The wind direction correction device 50 acquires the conversion model and / or the update information of the conversion model from the model generation device 20. The model generation device 20 generates a conversion model using, for example, a model generated by machine learning (hereinafter referred to as a base model). When there are a plurality of wind power generation devices 10, the base model and the conversion model are generated for each of the plurality of wind power generation devices 10.
　上記したように、空気密度パラメータの一例は気温（外気温）である。この気温は、風力発電装置１０の外面に取り付けられた温度計４０によって測定される。そして風向補正装置５０は、温度計４０から測定データを取得する。ただし、空気密度パラメータは、さらに湿度及び気圧の少なくとも一方を含んでいてもよい。この場合、温度計４０は湿度及び／又は気圧を測定する機能も有している。 As mentioned above, an example of the air density parameter is air temperature (outside air temperature). This air temperature is measured by a thermometer 40 attached to the outer surface of the wind power generator 10. Then, the wind direction correction device 50 acquires measurement data from the thermometer 40. However, the air density parameter may further include at least one of humidity and air pressure. In this case, the thermometer 40 also has a function of measuring humidity and / or atmospheric pressure.
　本図に示す例において、風向補正装置５０は風力発電装置１０のナセル７０に搭載されている。ただし、風向補正装置５０は風力発電装置１０のうちナセル７０以外の部分に搭載されていてもよいし、風力発電装置１０の外部に設けられていてもよい。 In the example shown in this figure, the wind direction correction device 50 is mounted on the nacelle 70 of the wind power generation device 10. However, the wind direction correction device 50 may be mounted on a portion of the wind power generation device 10 other than the nacelle 70, or may be provided outside the wind power generation device 10.
　図２の各図は、風向計３０が測定した相対風向が適正値でない状態が生じる理由を説明するための図である。 Each figure of FIG. 2 is a diagram for explaining the reason why the relative wind direction measured by the weather vane 30 is not an appropriate value.
　風向計３０は、風力発電装置１０が有する回転体軸（ローター）が風力発電装置１０の前方の風向と平行な状態で相対風向が０°になるように、ナセル７０に取り付けられている。そして図１に示したヨー制御装置６０は、風向計３０が測定した相対風向が０°になるように、ヨー方向を制御する。しかし、図２（Ａ）に示すように、風力発電装置１０のローターと風力発電装置１０の前方の風向が平行でない状態で相対風向が０°となる場合がある。この理由の一つは、風向計３０をナセル７０に取り付けるときのミスアライメントである。 The wind direction meter 30 is attached to the nacelle 70 so that the rotating body axis (rotor) of the wind power generation device 10 is parallel to the wind direction in front of the wind power generation device 10 and the relative wind direction is 0 °. Then, the yaw control device 60 shown in FIG. 1 controls the yaw direction so that the relative wind direction measured by the weather vane 30 becomes 0 °. However, as shown in FIG. 2A, the relative wind direction may be 0 ° when the rotor of the wind power generation device 10 and the wind direction in front of the wind power generation device 10 are not parallel. One of the reasons for this is misalignment when the weathercock 30 is attached to the nacelle 70.
　また、風力発電装置１０のローターはブレード７４を有している。そして、風向計３０はブレード７４の後方に配置されている。このため、風は、風向計３０に届く前にブレード７４から影響を受けることがある。この影響は、風向計３０が測定した相対風向と風力発電装置１０の前方の相対風向に誤差が生じる一因となる。 Further, the rotor of the wind power generation device 10 has a blade 74. The weather vane 30 is arranged behind the blade 74. Therefore, the wind may be affected by the blade 74 before reaching the weathercock 30. This effect contributes to an error between the relative wind direction measured by the wind direction meter 30 and the relative wind direction in front of the wind power generator 10.
　そして風向補正装置５０は、上記した２つの理由に起因した誤差を同時に補正することができる。 And the wind direction correction device 50 can simultaneously correct the error caused by the above two reasons.
　図３は、モデル生成装置２０の機能構成の一例を示す図である。モデル生成装置２０は、訓練データ取得部２２０、ベースモデル生成部２３０、変換モデル生成部２５０、及びモデル送信部２６０を有している。 FIG. 3 is a diagram showing an example of the functional configuration of the model generator 20. The model generation device 20 includes a training data acquisition unit 220, a base model generation unit 230, a conversion model generation unit 250, and a model transmission unit 260.
　訓練データ取得部２２０は、複数の訓練データを取得する。複数の訓練データのそれぞれは、風速、相対風向、空気密度パラメータ、ナセルの方向、及び発電装置７２の出力電力を含んでいる。本図に示す例において、訓練データ取得部２２０は訓練データ記憶部２１０から訓練データを取得する。訓練データ記憶部２１０は、訓練データとして風力発電装置１０の実績データ（すなわち風力発電装置１０が発電している間に得られたデータ）を用いる。風力発電装置１０が複数ある場合、訓練データ記憶部２１０は、複数の風力発電装置１０別に訓練データを記憶している。この訓練データは、風速が所定の範囲の時のデータ（例えば後述する図７において直線近似ができる範囲）であるのが好ましい。 The training data acquisition unit 220 acquires a plurality of training data. Each of the plurality of training data includes wind speed, relative wind direction, air density parameter, nacelle direction, and output power of the generator 72. In the example shown in this figure, the training data acquisition unit 220 acquires training data from the training data storage unit 210. The training data storage unit 210 uses the actual data of the wind power generation device 10 (that is, the data obtained while the wind power generation device 10 is generating power) as the training data. When there are a plurality of wind power generation devices 10, the training data storage unit 210 stores training data for each of the plurality of wind power generation devices 10. This training data is preferably data when the wind speed is in a predetermined range (for example, a range in which linear approximation can be performed in FIG. 7 described later).
　ベースモデル生成部２３０は、訓練データ取得部２２０が取得した複数の訓練データを用いて機械学習を行うことにより、ベースモデルを生成する。図１を用いて説明したように、ベースモデルは変換モデルを生成する際に用いられる。図４に示すように、この機械学習において、説明変数は、風速（ｖ）、空気密度パラメータ（Ｔ）、相対風向（φ）、及びナセルの方向（ω）を含んでおり、目的変数は出力電力（Ｐ）である。そしてベースモデルは、風速（ｖ）、空気密度パラメータ（Ｔ）、相対風向（φ）、及びナセルの方向（ω）が入力されると、出力電力（Ｐ）を出力する。ベースモデル生成部２３０は、生成したベースモデルをモデル記憶部２４０に記憶させる。 The base model generation unit 230 generates a base model by performing machine learning using a plurality of training data acquired by the training data acquisition unit 220. As described with reference to FIG. 1, the base model is used in generating the transformation model. As shown in FIG. 4, in this machine learning, the explanatory variables include the wind speed (v), the air density parameter (T), the relative wind direction (φ), and the nacelle direction (ω), and the objective variable is the output. Power (P). Then, the base model outputs the output power (P) when the wind speed (v), the air density parameter (T), the relative wind direction (φ), and the direction of the nacelle (ω) are input. The base model generation unit 230 stores the generated base model in the model storage unit 240.
　なお、ベースモデル生成部２３０は、機械学習以外の方法、例えば、物理的なモデル、または現象を近似的に表すモデルを用いて、ベースモデルを生成してもよい。 The base model generation unit 230 may generate a base model by using a method other than machine learning, for example, a physical model or a model that approximately represents a phenomenon.
　変換モデル生成部２５０は、ベースモデル生成部２３０が生成したベースモデルを用いて、変換モデル、又は変換モデルを更新するためのデータ（以下、更新データと記載）を生成する。変換モデル生成部２５０が行う処理の詳細については、他の図を用いて後述する。変換モデル生成部２５０は、生成した変換モデル及び更新データを、モデル記憶部２４０に記憶させる。なお、風力発電装置１０が複数ある場合、変換モデル生成部２５０は、複数の風力発電装置１０別に、その風力発電装置１０で用いられるべきベースモデル、変換モデル、及び更新データを記憶している。 The conversion model generation unit 250 uses the base model generated by the base model generation unit 230 to generate a conversion model or data for updating the conversion model (hereinafter referred to as update data). Details of the processing performed by the conversion model generation unit 250 will be described later with reference to other figures. The conversion model generation unit 250 stores the generated conversion model and update data in the model storage unit 240. When there are a plurality of wind power generation devices 10, the conversion model generation unit 250 stores the base model, conversion model, and update data to be used in the wind power generation device 10 for each of the plurality of wind power generation devices 10.
　モデル送信部２６０は、モデル生成装置２０が生成した変換モデル又は更新データを風向補正装置５０に送信する。本図に示す例において、モデル送信部２６０は、モデル記憶部２４０から変換モデル及び更新データを読み出して風向補正装置５０に送信する。 The model transmission unit 260 transmits the conversion model or update data generated by the model generation device 20 to the wind direction correction device 50. In the example shown in this figure, the model transmission unit 260 reads out the conversion model and the update data from the model storage unit 240 and transmits the conversion model and the update data to the wind direction correction device 50.
　なお、図３に示した例において、訓練データ記憶部２１０及びモデル記憶部２４０はモデル生成装置２０の一部となっている。ただし、訓練データ記憶部２１０及びモデル記憶部２４０はモデル生成装置２０の外部に位置していてもよい。 In the example shown in FIG. 3, the training data storage unit 210 and the model storage unit 240 are a part of the model generation device 20. However, the training data storage unit 210 and the model storage unit 240 may be located outside the model generation device 20.
　図５及び図６は、ベースモデルから変換モデルを生成する方法（すなわち変換モデル生成部２５０が行う処理）の一例を説明するための図である。上記したように、ベースモデルは、風速（ｖ）、空気密度パラメータ（Ｔ）、相対風向（φ）、及びナセルの方向（ω）が入力されると、出力電力（Ｐ）を出力する。このため、このベースモデルにおいて、風速（ｖ）、空気密度パラメータ（Ｔ）、及びナセルの方向（ω）を固定して、相対風向（φ）を変更すると、その風速（ｖ）、空気密度パラメータ（Ｔ）、及びナセルの方向（ω）における、相対風向（φ）を変数とした出力電力（Ｐ）の関数が得られる。 5 and 6 are diagrams for explaining an example of a method of generating a conversion model from a base model (that is, a process performed by the conversion model generation unit 250). As described above, the base model outputs the output power (P) when the wind speed (v), the air density parameter (T), the relative wind direction (φ), and the direction of the nacelle (ω) are input. Therefore, in this base model, if the wind speed (v), air density parameter (T), and nacelle direction (ω) are fixed and the relative wind direction (φ) is changed, the wind speed (v) and air density parameter are changed. A function of output power (P) with the relative wind direction (φ) as a variable in (T) and the direction of the nacelle (ω) can be obtained.
　図５は、この関数の一例である。風力発電装置１０の前方の風向に風向計３０の向きが調整されている場合、相対風向（φ）＝０のときにこの関数は極大値をとる。しかし、風力発電装置１０の前方の風向に風向計３０の向きが調整されていない場合、この関数が極大値をとるときのφは０とは異なる値となる。そして、この時のφは、この誤差（Δφ）を示している。 FIG. 5 is an example of this function. When the direction of the wind direction meter 30 is adjusted to the wind direction in front of the wind power generation device 10, this function takes a maximum value when the relative wind direction (φ) = 0. However, when the direction of the wind direction meter 30 is not adjusted to the wind direction in front of the wind power generation device 10, φ when this function takes a maximum value is a value different from 0. Then, φ at this time indicates this error (Δφ).
　そして、変換モデル生成部２５０は、風速（ｖ）、空気密度パラメータ（Ｔ）、及びナセルの方向（ω）のそれぞれを変えつつ、上記した処理を行うことにより、風速（ｖ）、空気密度パラメータ（Ｔ）、及びナセルの方向（ω）の組み合わせ毎の、Δφを算出できる。この結果は変換モデルとしてモデル記憶部２４０に記憶される。そしてこの変換モデルを用いることにより、風速（ｖ）、空気密度パラメータ（Ｔ）、及びナセルの方向（ω）をΔφに変換できる。なお、図６は、空気密度パラメータ（温度）及びナセルの方向を固定した時の、風速（ｖ）とΔφの関係を示している。 Then, the conversion model generation unit 250 performs the above processing while changing each of the wind speed (v), the air density parameter (T), and the direction of the nacelle (ω), thereby performing the wind speed (v) and the air density parameter. Δφ can be calculated for each combination of (T) and the direction (ω) of the nacelle. This result is stored in the model storage unit 240 as a conversion model. Then, by using this conversion model, the wind speed (v), the air density parameter (T), and the direction of the nacelle (ω) can be converted to Δφ. Note that FIG. 6 shows the relationship between the wind speed (v) and Δφ when the air density parameter (temperature) and the direction of the nacelle are fixed.
　なお、風向補正装置５０は、所定の条件を満たしたときに、補正パラメータとして定数を用いる。所定の条件の例については後述する。モデル生成装置２０のベースモデル生成部２３０は、この定数も算出する。図７及び図８は、この定数の算出方法の一例を説明するための図である。 The wind direction correction device 50 uses a constant as a correction parameter when a predetermined condition is satisfied. An example of a predetermined condition will be described later. The base model generation unit 230 of the model generation device 20 also calculates this constant. 7 and 8 are diagrams for explaining an example of a method for calculating this constant.
　より詳細には、図７は、風速（ｖ）と発電装置７２の出力電力（Ｐ）の関係を示す図である。出力電力（Ｐ）は、風速が上がるにつれて上がるが、多少のばらつきを有している。このばらつきの要因の一つは、相対風向に分布が生じること（Δφ）である。この分布が生じる理由は、相対風向が頻繁に変化することにある。一方、風向計３０の向きが風力発電装置１０の前方の風向にあっていると、Δφ＝０で出力電力（Ｐ）は極大値をとる。 More specifically, FIG. 7 is a diagram showing the relationship between the wind speed (v) and the output power (P) of the power generation device 72. The output power (P) increases as the wind speed increases, but has some variation. One of the factors of this variation is that the distribution is generated in the relative wind direction (Δφ). The reason for this distribution is that the relative wind direction changes frequently. On the other hand, when the direction of the wind direction meter 30 is in the direction of the wind in front of the wind power generation device 10, the output power (P) takes a maximum value at Δφ = 0.
　ここでベースモデル生成部２３０は、例えば訓練データ記憶部２１０が記憶している訓練データを用いて、各風速（ｖ）における出力電圧の平均値（Ｐｖ＿ａｖｇ）を求める。そして、各訓練データについて、その訓練データの出力電力Ｐｖ（φ）と、その訓練データの風速における出力電力の平均値（Ｐｖ＿ａｖｇ）と、の差（ΔＰ）を算出する。ΔＰは、相対風向（φ）の関数とみなすこともできる。 Here, the base model generation unit 230 obtains an average value (P v_avg ) of the output voltage at each wind speed (v) using, for example, the training data stored in the training data storage unit 210. Then, for each training data, the difference (ΔP) between the output power P v (φ) of the training data and the average value (P v_avg ) of the output power at the wind speed of the training data is calculated. ΔP can also be regarded as a function of the relative wind direction (φ).
　図８は、ΔＰの大きさを示す値（ε（φ））と相対風向（φ）の関係を示す図である。本図において、ε（φ）は、（Ｐｖ（φ）／（Ｐｖ＿ａｖｇ）－１）×１００）と定義されている。そしてベースモデル生成部２３０は、各訓練データについて、ｘを相対風向（φ）としてｙをε（φ）としたデータを生成し、これらデータを処理することにより、相対風向（φ）別にε（φ）の平均値を算出する。そしてベースモデル生成部２３０は、ε（φ）が最大となったときの相対風向（φ）を、補正パラメータとして用いられる定数にする。 FIG. 8 is a diagram showing the relationship between the value (ε (φ)) indicating the magnitude of ΔP and the relative wind direction (φ). In this figure, ε (φ) is defined as (P v (φ) / (P v_avg ) -1) × 100). Then, the base model generation unit 230 generates data in which x is the relative wind direction (φ) and y is ε (φ) for each training data, and by processing these data, ε (for each relative wind direction (φ)). Calculate the average value of φ). Then, the base model generation unit 230 sets the relative wind direction (φ) when ε (φ) is maximized as a constant used as a correction parameter.
　なお、ベースモデル生成部２３０が算出した定数は、モデル記憶部２４０に記憶される。そして変換モデル生成部２５０は、変換モデル又はその更新データとともに、この定数も風向補正装置５０に送信する。 The constant calculated by the base model generation unit 230 is stored in the model storage unit 240. Then, the conversion model generation unit 250 transmits this constant to the wind direction correction device 50 together with the conversion model or its update data.
　図９は、風向補正装置５０の機能構成の一例を示す図である。風向補正装置５０は、実績取得部５３０及び補正量決定部５４０を備える。 FIG. 9 is a diagram showing an example of the functional configuration of the wind direction correction device 50. The wind direction correction device 50 includes a performance acquisition unit 530 and a correction amount determination unit 540.
　実績取得部５３０は、実績データを取得する。実績データは、その時の風力発電装置１０に関するデータであり、上記した空気密度パラメータ（例えば温度を含む）、風向計３０が測定した風速、及びナセル７０の方向を含んでいる。空気密度パラメータは、例えば温度計４０によって生成され、ナセル７０の方向は、例えばヨー制御装置６０によって生成される。 The achievement acquisition unit 530 acquires the achievement data. The actual data is data on the wind power generator 10 at that time, and includes the above-mentioned air density parameters (including temperature, for example), the wind speed measured by the weather vane 30, and the direction of the nacelle 70. The air density parameter is generated, for example, by the thermometer 40, and the direction of the nacelle 70 is generated, for example, by the yaw controller 60.
　補正量決定部５４０は、実績取得部５３０が取得した空気密度パラメータ、風速、及びナセルの方向を用いて、上記した補正パラメータを決定する。補正量決定部５４０は、補正パラメータを決定する際、モデル生成装置２０が生成した変換モデルを用いる。この変換モデルは、モデル記憶部５２０に記憶されている。 The correction amount determination unit 540 determines the above-mentioned correction parameters using the air density parameter, the wind speed, and the direction of the nacelle acquired by the performance acquisition unit 530. The correction amount determination unit 540 uses the conversion model generated by the model generation device 20 when determining the correction parameters. This conversion model is stored in the model storage unit 520.
　本図に示す例において、補正量決定部５４０は、補正パラメータをヨー制御装置６０に出力する。ヨー制御装置６０は、この補正パラメータを用いて風向計３０が生成した相対風向を補正し、補正後の相対風向を用いてナセル７０の方向を制御する。 In the example shown in this figure, the correction amount determination unit 540 outputs the correction parameter to the yaw control device 60. The yaw control device 60 corrects the relative wind direction generated by the wind direction meter 30 using this correction parameter, and controls the direction of the nacelle 70 using the corrected relative wind direction.
　また補正量決定部５４０は、空気密度パラメータ、相対風向、及び風速の少なくとも一つが基準を満たさなかったとき、補正パラメータとして定数を用いる。ここで用いられる基準は、空気密度パラメータ、相対風向、及び風速別に設定されており、これらが正常な状態の値である、と推定される範囲になっている。すなわち補正量決定部５４０は、変換モデルに入力されるパラメータの少なくとも一つが異常な値になったとき、補正パラメータとして定数を用いる。この定数も、上記したようにモデル生成装置２０が生成している。そしてこの定数は、モデル記憶部５２０に記憶されている。なお、このような状態になるときの一例としては、パラメータを生成するためのセンサに故障が生じた場合である。 Further, the correction amount determination unit 540 uses a constant as a correction parameter when at least one of the air density parameter, the relative wind direction, and the wind speed does not meet the standard. The criteria used here are set for each of the air density parameters, the relative wind direction, and the wind speed, and these are in the range estimated to be the values in the normal state. That is, the correction amount determination unit 540 uses a constant as a correction parameter when at least one of the parameters input to the conversion model becomes an abnormal value. This constant is also generated by the model generator 20 as described above. And this constant is stored in the model storage unit 520. An example of such a state is a case where a sensor for generating a parameter fails.
　本図に示す例において、風向補正装置５０はモデル更新部５１０を備えている。モデル更新部５１０は、モデル生成装置２０から変換モデル及び補正パラメータとしての定数を取得し、これらのデータをモデル記憶部５２０に記憶させる。またモデル更新部５１０は、モデル生成装置２０から変換モデルを更新するための更新データを取得し、この更新データを用いて変換モデルを更新する。なお、モデル更新部５１０は、モデル生成装置２０から補正パラメータとしての定数の更新値を取得すると、定数をこの更新値に書き換える。 In the example shown in this figure, the wind direction correction device 50 includes a model update unit 510. The model update unit 510 acquires a conversion model and constants as correction parameters from the model generation device 20, and stores these data in the model storage unit 520. Further, the model update unit 510 acquires update data for updating the conversion model from the model generation device 20, and updates the conversion model using the update data. When the model update unit 510 acquires the update value of the constant as the correction parameter from the model generation device 20, the model update unit 510 rewrites the constant to this update value.
　図１０は、モデル生成装置２０のハードウェア構成例を示す図である。モデル生成装置２０は、バス１０１０、プロセッサ１０２０、メモリ１０３０、ストレージデバイス１０４０、入出力インタフェース１０５０、及びネットワークインタフェース１０６０を有する。 FIG. 10 is a diagram showing a hardware configuration example of the model generator 20. The model generator 20 includes a bus 1010, a processor 1020, a memory 1030, a storage device 1040, an input / output interface 1050, and a network interface 1060.
　バス１０１０は、プロセッサ１０２０、メモリ１０３０、ストレージデバイス１０４０、入出力インタフェース１０５０、及びネットワークインタフェース１０６０が、相互にデータを送受信するためのデータ伝送路である。ただし、プロセッサ１０２０などを互いに接続する方法は、バス接続に限定されない。 The bus 1010 is a data transmission path for the processor 1020, the memory 1030, the storage device 1040, the input / output interface 1050, and the network interface 1060 to transmit and receive data to and from each other. However, the method of connecting the processors 1020 and the like to each other is not limited to the bus connection.
　プロセッサ１０２０は、ＣＰＵ（Central Processing Unit） やＧＰＵ（Graphics Processing Unit）などで実現されるプロセッサである。 The processor 1020 is a processor realized by a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or the like.
　メモリ１０３０は、ＲＡＭ（Random Access Memory）などで実現される主記憶装置である。 The memory 1030 is a main storage device realized by a RAM (Random Access Memory) or the like.
　ストレージデバイス１０４０は、ＨＤＤ（Hard Disk Drive）、ＳＳＤ（Solid State Drive）、メモリカード、又はＲＯＭ（Read Only Memory）などで実現される補助記憶装置である。ストレージデバイス１０４０はモデル生成装置２０の各機能（例えば訓練データ取得部２２０、ベースモデル生成部２３０、変換モデル生成部２５０、及びモデル送信部２６０）を実現するプログラムモジュールを記憶している。プロセッサ１０２０がこれら各プログラムモジュールをメモリ１０３０上に読み込んで実行することで、そのプログラムモジュールに対応する各機能が実現される。また、ストレージデバイス１０４０は訓練データ記憶部２１０及びモデル記憶部２４０としても機能する。 The storage device 1040 is an auxiliary storage device realized by an HDD (Hard Disk Drive), an SSD (Solid State Drive), a memory card, a ROM (Read Only Memory), or the like. The storage device 1040 stores a program module that realizes each function of the model generation device 20 (for example, a training data acquisition unit 220, a base model generation unit 230, a conversion model generation unit 250, and a model transmission unit 260). When the processor 1020 reads each of these program modules into the memory 1030 and executes them, each function corresponding to the program module is realized. The storage device 1040 also functions as a training data storage unit 210 and a model storage unit 240.
　入出力インタフェース１０５０は、モデル生成装置２０と各種入出力機器とを接続するためのインタフェースである。 The input / output interface 1050 is an interface for connecting the model generator 20 and various input / output devices.
　ネットワークインタフェース１０６０は、モデル生成装置２０をネットワークに接続するためのインタフェースである。このネットワークは、例えばＬＡＮ（Local Area Network）やＷＡＮ（Wide Area Network）である。ネットワークインタフェース１０６０がネットワークに接続する方法は、無線接続であってもよいし、有線接続であってもよい。モデル生成装置２０は、ネットワークインタフェース１０６０を介して風向補正装置５０と通信してもよい。 The network interface 1060 is an interface for connecting the model generator 20 to the network. This network is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network). The method of connecting the network interface 1060 to the network may be a wireless connection or a wired connection. The model generation device 20 may communicate with the wind direction correction device 50 via the network interface 1060.
　なお、風向補正装置５０のハードウェア構成も、図１０に示したモデル生成装置２０のハードウェア構成と同様である。そしてストレージデバイス１０４０は風向補正装置５０の各機能（例えばモデル更新部５１０、実績取得部５３０、及び補正量決定部５４０）を実現するプログラムモジュールを記憶している。またストレージデバイス１０４０は、モデル記憶部５２０としても機能する。 The hardware configuration of the wind direction correction device 50 is also the same as the hardware configuration of the model generation device 20 shown in FIG. The storage device 1040 stores a program module that realizes each function of the wind direction correction device 50 (for example, a model update unit 510, a result acquisition unit 530, and a correction amount determination unit 540). The storage device 1040 also functions as a model storage unit 520.
　図１１は、モデル生成装置２０が行う変換モデルの生成処理を説明するためのフローチャートである。モデル生成装置２０は、風力発電装置１０毎に本図に示す処理を行う。またモデル生成装置２０の訓練データ記憶部２１０が記憶している訓練データは定期的に更新される。このため、モデル生成装置２０は、定期的に本図に示す処理を行う。 FIG. 11 is a flowchart for explaining the generation process of the transformation model performed by the model generation device 20. The model generation device 20 performs the processing shown in this figure for each wind power generation device 10. Further, the training data stored in the training data storage unit 210 of the model generation device 20 is periodically updated. Therefore, the model generator 20 periodically performs the processes shown in this figure.
　まずモデル生成装置２０の訓練データ取得部２２０は、訓練データ記憶部２１０から訓練データを読み出す（ステップＳ１０）。次いでベースモデル生成部２３０は、この訓練データを用いてベースモデルを生成し（ステップＳ２０）、このベースモデルをモデル記憶部２４０に記憶させる。次いで変換モデル生成部２５０は、ステップＳ２０で生成されたベースモデルを用いて、変換モデル（又は更新データ）を生成し（ステップＳ３０）、生成した変換モデル（又は更新データ）をモデル記憶部２４０に記憶させる（ステップＳ４０）。 First, the training data acquisition unit 220 of the model generation device 20 reads the training data from the training data storage unit 210 (step S10). Next, the base model generation unit 230 generates a base model using this training data (step S20), and stores the base model in the model storage unit 240. Next, the conversion model generation unit 250 generates a conversion model (or update data) using the base model generated in step S20 (step S30), and transfers the generated conversion model (or update data) to the model storage unit 240. It is memorized (step S40).
　その後、モデル送信部２６０は、適切なタイミングで、変換モデル（又は更新データ）を風向補正装置５０に送信する。 After that, the model transmission unit 260 transmits the conversion model (or update data) to the wind direction correction device 50 at an appropriate timing.
　図１２は、風向補正装置５０が行う処理の一例を説明するためのフローチャートである。風力発電装置１０が発電を行っている間、風向補正装置５０は、リアルタイムで本図に示す処理を繰り返し行う。 FIG. 12 is a flowchart for explaining an example of the processing performed by the wind direction correction device 50. While the wind power generation device 10 is generating power, the wind direction correction device 50 repeats the process shown in this figure in real time.
　風向補正装置５０の実績取得部５３０は、その時の実績データを取得する。実績データは、図９を用いて説明したように、その時の風力発電装置１０に関するデータであり、上記した空気密度パラメータ（例えば温度を含む）、風向計３０が測定した風速、及びナセル７０の方向を含んでいる（ステップＳ１１０）。次いで補正量決定部５４０は、モデル記憶部５２０が記憶している変換モデルを用いて、実績データを相対風向の補正パラメータに変換し（ステップＳ１２０）、この補正パラメータをヨー制御装置６０に出力する（ステップＳ１３０）。そしてヨー制御装置６０は、この補正パラメータを用いて風向計３０が生成した相対風向を補正し、補正後の相対風向を用いてナセル７０の方向を制御する。 The performance acquisition unit 530 of the wind direction correction device 50 acquires the performance data at that time. As explained with reference to FIG. 9, the actual data is the data related to the wind power generator 10 at that time, and the above-mentioned air density parameters (including temperature, for example), the wind speed measured by the wind direction meter 30, and the direction of the nacelle 70. (Step S110). Next, the correction amount determination unit 540 converts the actual data into a correction parameter for the relative wind direction using the conversion model stored in the model storage unit 520 (step S120), and outputs this correction parameter to the yaw control device 60. (Step S130). Then, the yaw control device 60 corrects the relative wind direction generated by the wind direction meter 30 using this correction parameter, and controls the direction of the nacelle 70 using the corrected relative wind direction.
　図１３は、図１２の第１変形例を示す図である。本図に示す例は、以下の点を除いて図１２と同様である。補正量決定部５４０は、実績取得部５３０が取得した実績データが基準を満たしていた場合（ステップＳ１２２：Ｙｅｓ）、補正量決定部５４０は変換モデルを用いて補正パラメータを生成する（ステップＳ１２４）。一方、補正量決定部５４０は、この実績データが基準を満たしていなかった場合（ステップＳ１２２：Ｎｏ）、モデル記憶部５２０が記憶している定数を補正パラメータにする（ステップＳ１２６）。 FIG. 13 is a diagram showing a first modification of FIG. 12. The example shown in this figure is the same as that in FIG. 12 except for the following points. When the actual data acquired by the actual acquisition unit 530 meets the criteria (step S122: Yes), the correction amount determination unit 540 generates correction parameters using the conversion model (step S124). .. On the other hand, when the correction amount determination unit 540 does not satisfy the standard (step S122: No), the correction amount determination unit 540 sets the constant stored in the model storage unit 520 as the correction parameter (step S126).
　図１４は、図１２の第２変形例を示す図である。本図に示す例は、補正量決定部５４０が補正パラメータを用いて相対風向を補正し（ステップＳ１３２）、補正後の相対風向をヨー制御装置６０に出力する（ステップＳ１４２）点を除いて、図１２と同様である。そしてヨー制御装置６０は、補正量決定部５４０が補正した後の相対風向を用いてナセル７０の方向を制御する。 FIG. 14 is a diagram showing a second modification of FIG. 12. In the example shown in this figure, except that the correction amount determination unit 540 corrects the relative wind direction using the correction parameter (step S132) and outputs the corrected relative wind direction to the yaw control device 60 (step S142). It is the same as FIG. Then, the yaw control device 60 controls the direction of the nacelle 70 by using the relative wind direction after the correction amount determination unit 540 corrects.
　なお図１３に示す例において、図１４に示したように補正量決定部５４０が相対風向を補正してもよい。 In the example shown in FIG. 13, the correction amount determining unit 540 may correct the relative wind direction as shown in FIG.
　以上、本実施形態によれば、風向補正装置５０は、風向計３０が測定した相対風向を補正するための補正パラメータを生成する。このため、ヨー制御装置６０は、ナセル７０の向きを風向に合わせて精度良く制御できる。 As described above, according to the present embodiment, the wind direction correction device 50 generates a correction parameter for correcting the relative wind direction measured by the wind direction meter 30. Therefore, the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
（第２実施形態）
　図１５は、本実施形態に係る風向補正装置５０の機能構成の一例を示す図である。本図に示す例において、風向補正装置５０は、変換モデルを生成する機能を有している。 (Second Embodiment)
FIG. 15 is a diagram showing an example of the functional configuration of the wind direction correction device 50 according to the present embodiment. In the example shown in this figure, the wind direction correction device 50 has a function of generating a conversion model.
　具体的には、風向補正装置５０は、訓練データ記憶部５５０、訓練データ取得部５６０、ベースモデル生成部５７０、及び変換モデル生成部５８０を有している。訓練データ記憶部５５０、訓練データ取得部５６０、ベースモデル生成部５７０、及び変換モデル生成部５８０は、それぞれ図３の訓練データ記憶部２１０、訓練データ取得部２２０、ベースモデル生成部２３０、及び変換モデル生成部２５０と同様の機能を有している。また、モデル記憶部５２０は、第１実施形態で説明した機能に加えて、図３のモデル記憶部２４０の機能も有している。 Specifically, the wind direction correction device 50 has a training data storage unit 550, a training data acquisition unit 560, a base model generation unit 570, and a conversion model generation unit 580. The training data storage unit 550, the training data acquisition unit 560, the base model generation unit 570, and the conversion model generation unit 580 are the training data storage unit 210, the training data acquisition unit 220, the base model generation unit 230, and the conversion, respectively. It has the same function as the model generation unit 250. Further, the model storage unit 520 also has the function of the model storage unit 240 of FIG. 3 in addition to the functions described in the first embodiment.
　また本実施形態において、補正量決定部５４０は、訓練データ記憶部５５０が記憶している訓練データの数が基準以下であり、変換モデルの精度が十分でないと予想されるときには、変換パラメータとして定数を用いる。この定数は、図７及び図８を用いて説明した通りである。 Further, in the present embodiment, when the number of training data stored in the training data storage unit 550 is less than the standard and the accuracy of the conversion model is expected to be insufficient, the correction amount determination unit 540 is a constant as a conversion parameter. Is used. This constant is as described with reference to FIGS. 7 and 8.
　本実施形態によっても、ヨー制御装置６０は、ナセル７０の向きを風向に合わせて精度良く制御できる。 Also in this embodiment, the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
（第３実施形態）
　図１６は、本実施形態に係る風向補正装置５０を説明するための図である。本図に示す例において、ヨー制御装置６０は風向補正装置５０を兼ねている。言い換えると、風向補正装置５０は、相対風向及び補正パラメータを用いてナセル７０のヨー方向を制御する機能も有している。 (Third Embodiment)
FIG. 16 is a diagram for explaining the wind direction correction device 50 according to the present embodiment. In the example shown in this figure, the yaw control device 60 also serves as a wind direction correction device 50. In other words, the wind direction correction device 50 also has a function of controlling the yaw direction of the nacelle 70 by using the relative wind direction and the correction parameters.
　本実施形態によっても、ヨー制御装置６０は、ナセル７０の向きを風向に合わせて精度良く制御できる。 Also in this embodiment, the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
（第４実施形態）
　図１７は、本実施形態に係る風向補正装置５０を説明するための図である。本図に示す例において、風向補正装置５０は、風向計３０の一部となっている。そして風向計３０は、補正後の相対風向を出力する。ヨー制御装置６０は、この補正後の相対風向を用いてナセル７０の向きを風向に合わせて制御する。 (Fourth Embodiment)
FIG. 17 is a diagram for explaining the wind direction correction device 50 according to the present embodiment. In the example shown in this figure, the wind direction correction device 50 is a part of the wind direction meter 30. Then, the wind direction meter 30 outputs the corrected relative wind direction. The yaw control device 60 controls the direction of the nacelle 70 according to the wind direction by using the corrected relative wind direction.
　本実施形態によっても、ヨー制御装置６０は、ナセル７０の向きを風向に合わせて精度良く制御できる。 Also in this embodiment, the yaw control device 60 can accurately control the direction of the nacelle 70 according to the wind direction.
　以上、図面を参照して本発明の実施形態について述べたが、これらは本発明の例示であり、上記以外の様々な構成を採用することもできる。 Although the embodiments of the present invention have been described above with reference to the drawings, these are examples of the present invention, and various configurations other than the above can be adopted.
　また、上述の説明で用いた複数のフローチャートでは、複数の工程（処理）が順番に記載されているが、各実施形態で実行される工程の実行順序は、その記載の順番に制限されない。各実施形態では、図示される工程の順番を内容的に支障のない範囲で変更することができる。また、上述の各実施形態は、内容が相反しない範囲で組み合わせることができる。 Further, in the plurality of flowcharts used in the above description, a plurality of steps (processes) are described in order, but the execution order of the steps executed in each embodiment is not limited to the order of description. In each embodiment, the order of the illustrated steps can be changed within a range that does not hinder the contents. In addition, the above-described embodiments can be combined as long as the contents do not conflict with each other.
１０　　　　風力発電装置
２０　　　　モデル生成装置
３０　　　　風向計
４０　　　　温度計
５０　　　　風向補正装置
６０　　　　ヨー制御装置
７０　　　　ナセル
７２　　　　発電装置
７４　　　　ブレード
２１０　　　　訓練データ記憶部
２２０　　　　訓練データ取得部
２３０　　　　ベースモデル生成部
２４０　　　　モデル記憶部
２５０　　　　変換モデル生成部
２６０　　　　モデル送信部
５１０　　　　モデル更新部
５２０　　　　モデル記憶部
５３０　　　　実績取得部
５４０　　　　補正量決定部
５５０　　　　訓練データ記憶部
５６０　　　　訓練データ取得部
５７０　　　　ベースモデル生成部
５８０　　　　変換モデル生成部 10 Wind power generation device 20 Model generation device 30 Wind direction meter 40 Thermometer 50 Wind direction correction device 60 Yaw control device 70 Nacelle 72 Power generation device 74 Blade 210 Training data storage unit 220 Training data acquisition unit 230 Base model generation unit 240 Model storage unit 250 Conversion Model generation unit 260 Model transmission unit 510 Model update unit 520 Model storage unit 530 Achievement acquisition unit 540 Correction amount determination unit 550 Training data storage unit 560 Training data acquisition unit 570 Base model generation unit 580 Conversion model generation unit

Claims (13)

  1. 　風力発電装置に設けられた風向計の測定結果を補正する風向補正装置であって、
    　前記風向計の測定項目は、前記風力発電装置を基準とした風の向きである相対風向、及び風速を含んでおり、
    　前記風力発電装置の周囲における空気密度に関連する空気密度パラメータ、前記風向計が測定した前記風速、及び前記風力発電装置のナセルの方向を取得する実績取得部と、
    　前記空気密度パラメータ、前記風速、及び前記ナセルの方向を用いて、前記相対風向及び絶対風向の少なくとも一方を補正するための補正パラメータを決定する補正量決定部と、
    を備え、
    　前記補正量決定部は、前記空気密度パラメータ、前記風速、及び前記ナセルの方向を含む入力データを前記補正パラメータに変換する変換モデルを用いて、前記補正パラメータを決定する、風向補正装置。     It is a wind direction correction device that corrects the measurement result of the wind direction meter installed in the wind power generation device.
    The measurement items of the weather vane include the relative wind direction, which is the direction of the wind with respect to the wind power generator, and the wind speed.
    An air density parameter related to the air density around the wind power generator, the wind speed measured by the wind direction meter, and a performance acquisition unit for acquiring the direction of the nacelle of the wind power generator.
    A correction amount determining unit that determines a correction parameter for correcting at least one of the relative wind direction and the absolute wind direction using the air density parameter, the wind speed, and the direction of the nacelle.
    With
    The correction amount determining unit is a wind direction correction device that determines the correction parameters by using a conversion model that converts input data including the air density parameter, the wind speed, and the direction of the nacelle into the correction parameters.
  2. 　請求項１に記載の風向補正装置において、
    　前記空気密度パラメータは気温を含む、風向補正装置。     In the wind direction correction device according to claim 1,
    The air density parameter is a wind direction correction device including air temperature.
  3. 　請求項１又は２に記載の風向補正装置において、
    　前記補正量決定部は、前記空気密度パラメータ、前記相対風向、及び前記風速の少なくとも一つが基準を満たさなかったとき、前記補正パラメータとして定数を用いる風向補正装置。     In the wind direction correction device according to claim 1 or 2.
    The correction amount determining unit is a wind direction correction device that uses a constant as the correction parameter when at least one of the air density parameter, the relative wind direction, and the wind speed does not satisfy the standard.
  4. 　請求項１～３のいずれか一項に記載の風向補正装置において、
    　前記風速、前記相対風向、前記空気密度パラメータ、前記ナセルの方向、及び前記風力発電装置の出力電力を含んでいる複数の訓練データを取得する訓練データ取得部と、
    　前記複数の訓練データを用いて、前記風速、前記相対風向、前記空気密度パラメータ、及び前記ナセルの方向を説明変数の少なくとも一部として、前記出力電力を目的変数とした機械学習を実行してベースモデルを生成するベースモデル生成部と、
    　前記ベースモデルを用いて前記変換モデルを生成する変換モデル生成部と、
    を備える風向補正装置。     In the wind direction correction device according to any one of claims 1 to 3.
    A training data acquisition unit that acquires a plurality of training data including the wind speed, the relative wind direction, the air density parameter, the direction of the nacelle, and the output power of the wind power generator.
    Using the plurality of training data, machine learning is performed with the output power as the objective variable, with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the explanatory variables. The base model generator that generates the model and
    A transformation model generator that generates the transformation model using the base model,
    A wind direction correction device equipped with.
  5. 　請求項４に記載の風向補正装置において、
    　前記補正量決定部は、前記訓練データの数が基準以下であるとき、前記補正パラメータとして定数を用いる風向補正装置。     In the wind direction correction device according to claim 4,
    The correction amount determination unit is a wind direction correction device that uses a constant as the correction parameter when the number of training data is equal to or less than a reference.
  6. 　請求項１～３のいずれか一項に記載の風向補正装置において、
    　通信により、前記変換モデルを更新するための更新データを取得して前記変換モデルを更新するモデル更新部をさらに備える風向補正装置。     In the wind direction correction device according to any one of claims 1 to 3.
    A wind direction correction device further comprising a model update unit that acquires update data for updating the conversion model by communication and updates the conversion model.
  7. 　請求項１～６のいずれか一項に記載の風向補正装置において、
    　前記風向補正装置は前記風力発電装置に搭載される、風向補正装置。     In the wind direction correction device according to any one of claims 1 to 6.
    The wind direction correction device is a wind direction correction device mounted on the wind power generation device.
  8. 　請求項１～７のいずれか一項に記載の風向補正装置において、
    　前記相対風向及び前記補正パラメータを用いて前記ナセルの方向を制御する制御部を備える風向補正装置。     In the wind direction correction device according to any one of claims 1 to 7.
    A wind direction correction device including a control unit that controls the direction of the nacelle using the relative wind direction and the correction parameters.
  9. 　請求項１に記載の前記変換モデルを生成するモデル生成装置であって、
    　前記風速、前記相対風向、前記空気密度パラメータ、及び前記ナセルの方向を入力データの少なくとも一部として、前記風力発電装置の出力電力を出力データとするベースモデルを作成するベースモデル作成部と、
    　前記ベースモデルを用いて前記変換モデルを生成する変換モデル生成部と、
    を備えるモデル生成装置。     A model generator that generates the conversion model according to claim 1.
    A base model creation unit that creates a base model using the output power of the wind power generator as output data, with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the input data.
    A transformation model generator that generates the transformation model using the base model,
    A model generator equipped with.
  10. 　コンピュータが、風力発電装置に設けられた風向計の測定結果を補正する補正方法であって、
    　前記風向計の測定項目は、前記風力発電装置を基準とした風の向きである相対風向、及び風速を含んでおり、
    　前記コンピュータは、
    　　前記風力発電装置の周囲における空気密度に関連する空気密度パラメータ、前記風向計が測定した前記風速、及び前記風力発電装置のナセルの方向を取得する実績取得処理と、
    　　前記空気密度パラメータ、前記風速、及び前記ナセルの方向を用いて、前記相対風向及び絶対風向の少なくとも一方を補正するための補正パラメータを決定する補正決定処理と、
    を行い、
    　前記補正決定処理において、前記コンピュータは、前記空気密度パラメータ、前記風速、及び前記ナセルの方向を含む入力データを前記補正パラメータに変換する変換モデルを用いて、前記補正パラメータを決定する、補正方法。     A computer is a correction method that corrects the measurement result of the weather vane installed in the wind power generator.
    The measurement items of the weather vane include the relative wind direction, which is the direction of the wind with respect to the wind power generator, and the wind speed.
    The computer
    A record acquisition process for acquiring the air density parameters related to the air density around the wind turbine generator, the wind speed measured by the wind turbine meter, and the direction of the nacelle of the wind turbine generator.
    A correction determination process for determining a correction parameter for correcting at least one of the relative wind direction and the absolute wind direction using the air density parameter, the wind speed, and the direction of the nacelle.
    And
    In the correction determination process, the computer determines the correction parameter using a conversion model that converts input data including the air density parameter, the wind speed, and the direction of the nacelle into the correction parameter.
  11. 　コンピュータが、請求項１に記載の前記変換モデルを生成するモデル生成方法であって、
    　前記コンピュータは、
    　　前記風速、前記相対風向、前記空気密度パラメータ、及び前記ナセルの方向を入力データの少なくとも一部として、前記風力発電装置の出力電力を出力データとした出力データとするベースモデルを作成するベースモデル作成処理と、
    　　前記ベースモデルを用いて前記変換モデルを生成する変換モデル生成処理と、
    を行うモデル生成方法。     A model generation method in which a computer generates the transformation model according to claim 1.
    The computer
    Creating a base model that uses the output power of the wind power generator as output data with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the input data. Processing and
    A transformation model generation process for generating the transformation model using the base model,
    Model generation method to do.
  12. 　コンピュータを、風力発電装置に設けられた風向計の測定結果を補正する補正装置として機能させるためのプログラムであって、
    　前記風向計の測定項目は、前記風力発電装置を基準とした風の向きである相対風向、及び風速を含んでおり、
    　前記コンピュータに、
    　　前記風力発電装置の周囲における空気密度に関連する空気密度パラメータ、前記風向計が測定した前記風速、及び前記風力発電装置のナセルの方向を取得する実績取得機能と、
    　　前記空気密度パラメータ、前記風速、及び前記ナセルの方向を用いて、前記相対風向及び絶対風向の少なくとも一方を補正するための補正パラメータを決定する補正決定機能と、
    を持たせ、
    　前記補正決定機能は、前記空気密度パラメータ、前記風速、及び前記ナセルの方向を含む入力データを前記補正パラメータに変換する変換モデルを用いて、前記補正パラメータを決定する、プログラム。     A program for making a computer function as a correction device for correcting the measurement result of a weather vane provided in a wind power generator.
    The measurement items of the weather vane include the relative wind direction, which is the direction of the wind with respect to the wind power generator, and the wind speed.
    On the computer
    An air density parameter related to the air density around the wind turbine generator, a wind speed measured by the wind turbine meter, and a track record acquisition function for acquiring the direction of the nacelle of the wind turbine generator.
    A correction determination function for determining a correction parameter for correcting at least one of the relative wind direction and the absolute wind direction using the air density parameter, the wind speed, and the direction of the nacelle.
    To have
    The correction determination function is a program that determines the correction parameter using a conversion model that converts input data including the air density parameter, the wind speed, and the direction of the nacelle into the correction parameter.
  13. 　コンピュータを、請求項１に記載の前記変換モデルを生成するモデル生成装置として機能させるためのプログラムであって、
    　前記コンピュータに、
    　　前記風速、前記相対風向、前記空気密度パラメータ、及び前記ナセルの方向を入力データの少なくとも一部として、前記風力発電装置の出力電力を出力データとするベースモデルを作成するベースモデル作成機能と、
    　　前記ベースモデルを用いて前記変換モデルを生成する変換モデル生成機能と、
    を持たせるプログラム。     A program for operating a computer as a model generator for generating the conversion model according to claim 1.
    On the computer
    A base model creation function for creating a base model using the output power of the wind power generator as output data, with the wind speed, the relative wind direction, the air density parameter, and the direction of the nacelle as at least a part of the input data.
    A transformation model generation function that generates the transformation model using the base model,
    Program to have.
PCT/JP2021/001429 2021-01-18 2021-01-18 Wind direction correction device, model generation device, correction method, model generation method, and program WO2022153525A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/JP2021/001429 WO2022153525A1 (en) 2021-01-18 2021-01-18 Wind direction correction device, model generation device, correction method, model generation method, and program
JP2022575029A JP7408846B2 (en) 2021-01-18 2021-01-18 Wind direction correction device, model generation device, correction method, model generation method, and program

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/001429 WO2022153525A1 (en) 2021-01-18 2021-01-18 Wind direction correction device, model generation device, correction method, model generation method, and program

Publications (1)

Publication Number Publication Date
WO2022153525A1 true WO2022153525A1 (en) 2022-07-21

Family

ID=82448145

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/001429 WO2022153525A1 (en) 2021-01-18 2021-01-18 Wind direction correction device, model generation device, correction method, model generation method, and program

Country Status (2)

Country Link
JP (1) JP7408846B2 (en)
WO (1) WO2022153525A1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007530926A (en) * 2004-03-26 2007-11-01 ダンマーク テクニスク ユニバーシテ Apparatus and method for determining wind speed and direction experienced by a wind turbine
WO2008146604A1 (en) * 2007-05-25 2008-12-04 Mitsubishi Heavy Industries, Ltd. Wind power generator, wind power generation system, and generation control method of wind power generator
JP2013108462A (en) * 2011-11-22 2013-06-06 Fuji Electric Co Ltd System and program for predicting wind power generated electricity
JP2018097733A (en) * 2016-12-15 2018-06-21 株式会社日立製作所 Operation support system and operation support method
US20190323482A1 (en) * 2018-02-19 2019-10-24 Senvion Gmbh Method and system for determining an alignment correction function
JP2020513492A (en) * 2016-11-15 2020-05-14 ヴォッベン プロパティーズ ゲーエムベーハー Wind power generation facility control method and related wind power generation facility

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007530926A (en) * 2004-03-26 2007-11-01 ダンマーク テクニスク ユニバーシテ Apparatus and method for determining wind speed and direction experienced by a wind turbine
WO2008146604A1 (en) * 2007-05-25 2008-12-04 Mitsubishi Heavy Industries, Ltd. Wind power generator, wind power generation system, and generation control method of wind power generator
JP2013108462A (en) * 2011-11-22 2013-06-06 Fuji Electric Co Ltd System and program for predicting wind power generated electricity
JP2020513492A (en) * 2016-11-15 2020-05-14 ヴォッベン プロパティーズ ゲーエムベーハー Wind power generation facility control method and related wind power generation facility
JP2018097733A (en) * 2016-12-15 2018-06-21 株式会社日立製作所 Operation support system and operation support method
US20190323482A1 (en) * 2018-02-19 2019-10-24 Senvion Gmbh Method and system for determining an alignment correction function

Also Published As

Publication number Publication date
JPWO2022153525A1 (en) 2022-07-21
JP7408846B2 (en) 2024-01-05

Similar Documents

Publication Publication Date Title
EP2693049B1 (en) Control system and method for mitigating rotor imbalance on a wind turbine
US7452185B2 (en) Blade-pitch-angle control device and wind power generator
US9189755B2 (en) Available power estimator
EP3364324A1 (en) Method and device for detecting equivalent load of wind turbine generator system
JP6494514B2 (en) Wind turbine control method using predicted input wind speed
KR20130028923A (en) Control device for a wind power plant
US20200284239A1 (en) System And Method For Augmenting Control Of A Wind Turbine Assembly
CN107709766A (en) The method for calibrating the load cell of wind turbine
CN107110125A (en) For the method and system for the dynamic distortion for determining wind turbine blade
WO2018115423A1 (en) Fatigue load minimization in an operation of a wind farm
CN117212077A (en) Wind wheel fault monitoring method, device and equipment of wind turbine and storage medium
TWI729349B (en) Wind power generation device and wind power generation system
Grundkötter et al. Precision blade deflection measurement system using wireless inertial sensor nodes
WO2022153525A1 (en) Wind direction correction device, model generation device, correction method, model generation method, and program
CN103097725B (en) Be used for the method and apparatus of the valuation of at least one measured value of asking for wind power installation
EP3828408A1 (en) A method and an apparatus for computer-implemented monitoring of a wind turbine
US11920562B2 (en) Temperature estimation in a wind turbine
US20240110546A1 (en) Determining tower to tip clearance for a wind turbine
US10961981B2 (en) Control for a wind turbine
CN114303012A (en) Controlling the power output of a wind turbine below rated wind speed
EP4339453A1 (en) Method for operating a wind turbine, control system and wind turbine
Noppe et al. High frequent scada-based thrust load modeling of wind turbines
US12006914B2 (en) Controlling power output of a wind turbine at below-rated wind speed
JP7237775B2 (en) Plant control system
CN117332602A (en) Primary frequency modulation simulation method and device for wind driven generator

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21919418

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022575029

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21919418

Country of ref document: EP

Kind code of ref document: A1