Systems and methods for controlling a wind turbine

Patent metadata

Field	Value
Publication number	US-12560150-B2
Application number	US-202118699019-A
Country	US
Kind code	B2
Filing date	Oct 7, 2021
Priority date	Oct 7, 2021
Publication date	Feb 24, 2026
Grant date	Feb 24, 2026

Abstract

Systems and methods are provided for the control of a wind turbine. Accordingly, a wind classification module of a controller determines a current aerodynamic state of the wind resource based, at least in part, on a current operational data set of the wind turbine. The current operational data set is indicative of a current operation of the wind turbine. A configuration intelligence module of the controller then generates an estimated configuration for a turbine estimator module and a predictive control configuration for a predictive control module based, at least in part, on the current aerodynamic state. An operation of the wind turbine is emulated via the turbine estimator module to generate a control initial state for the predictive control module. The predictive control module then determines a predicted performance of the wind turbine over a predictive interval based on the control initial state and the predictive control configuration. The predictive control module generates a set point for at least one actuator of the wind turbine based on the predicted performance, and an operating state of the wind turbine is affected via the at least one actuator in accordance with the setpoint.

First claim

What is claimed is:

1 . A method for controlling a wind turbine, the wind turbine including a controller, the method comprising:
determining, via a wind classification module of the controller, a current aerodynamic state of a wind resource based, at least in part, on a current operational data set of the wind turbine indicative of a current operation of the wind turbine; generating, via a configuration intelligence module of the controller, an estimator configuration for a turbine estimator module based, at least in part on the current aerodynamic state; emulating, via the turbine estimator module of the controller, an operation of the wind turbine so as to generate a control initial state for a predictive control module, wherein the control initial state comprises a modeled current operating state of a plurality of components of the wind turbine; generating, via the configuration intelligence module, a predictive control configuration for the predictive control module based, at least in part, on the current aerodynamic state, wherein generating the estimator configuration and the predictive control configuration further comprises selecting, via the configuration intelligence module, a gain tuning of a plurality of gain tunings based on the current aerodynamic state, wherein each gain tuning is configured to modify at least one of the estimator configuration and the predictive control configuration; determining, via the predictive control module of the controller, a predicted performance of the wind turbine over a predictive interval based on the control initial state and the predictive control configuration; generating, via the predictive control module, a setpoint for at least one actuator of the wind turbine based on the predicted performance; and affecting an operating state of the wind turbine via the at least one actuator in accordance with the setpoint.

2 . The method of claim 1 , wherein the wind classification module comprises a first plurality of aeroelastic estimators, wherein each aeroelastic estimator is tuned to a different presumptive wind condition, and wherein determining the current aerodynamic state further comprises:
generating, via the wind classification module, a plurality of wind descriptive parameters from each aeroelastic estimator based, at least in part, on the current operational data set; and determining, via the wind classification module, the current aerodynamic state based on a designated portion of the pluralities of wind descriptive parameters.

3 . The method of claim 2 , further comprising determining the designated portion of the pluralities of wind descriptive parameters by:
a) modelling, via the controller, a projected operational response of the wind turbine to one potential aerodynamic state of a plurality of potential aerodynamic states of the wind resource to generate a projected operational data set; b) generating, via the each aeroelastic estimator of the first plurality of aeroelastic estimators, a plurality of projected wind descriptive parameters based on the projected operational data set; c) determining a portion of the pluralities of projected wind descriptive parameters that replicates the one potential aerodynamic state; d) correlating the portion of the pluralities of projected wind descriptive parameters to the projected operational data set corresponding to the potential aerodynamic state; e) repeating steps a)-d) for a remainder of the plurality of potential aerodynamic states, wherein the plurality of potential aerodynamic states corresponds to an environmental operating envelope of the wind turbine; f) generating, via the controller, an operational-response signature data set comprising the projected operational data set at each potential aerodynamic state of the plurality of potential aerodynamic states and the portion of the pluralities of projected wind descriptive parameters that corresponds; and determining, via the wind classification module, the designated portion of the pluralities of wind descriptive parameters based on the operational-response signature data set, the pluralities of wind descriptive parameters, and the current operational data set.

4 . The method of claim 3 , wherein determining the designated portion of the pluralities of wind descriptive parameters further comprises:
implementing at least one machine learning algorithm within the wind classification module to generate the operational-response signature data set; implementing the at least one machine learning algorithm to determine the designated portion of the pluralities of wind descriptive parameters corresponding to the current operational data set; and implementing the at least one machine learning algorithm to determine the current aerodynamic state based on the designated portion of the pluralities of wind descriptive parameters.

5 . The method of claim 2 , wherein each aeroelastic estimator of the first plurality of aeroelastic estimators comprises at least one aeroelastic model and at least one filtering algorithm, wherein the at least one aeroelastic model of each aeroelastic estimator is configured to model a behavior of the wind turbine as a multibody system of flexible structures, and wherein generating the pluralities of wind descriptive parameters further comprises:
deriving, via the at least one aeroelastic model, a resultant aerodynamic state of the wind resource that develops the current operational data set in the presence of the presumptive wind condition corresponding to the tuning of the at least one aerodynamic model; and determining the plurality of wind descriptive parameters corresponding to the resultant aerodynamic state derived by the at least one aeroelastic model of each aeroelastic estimator of the first plurality of aeroelastic estimators.

6 . The method of claim 1 , wherein generating the estimator configuration and the predictive control configuration further comprises:
determining, via an operating condition module of the controller, a current operating condition of the wind turbine based, at least in part, on the current operational data set, wherein the operational data set further comprises a plurality of output signals from a sensor system of the wind turbine and at least one control signal from the controller; generating, via the configuration intelligence module of the controller, the estimator configuration for the turbine estimator module based, at least in part, on the current aerodynamic state and the current operating condition; and generating, via the configuration intelligence module, the predictive control configuration for the predictive control module based, at least in part, on the current aerodynamic state and the current operating condition.

7 . The method of claim 6 , wherein the operating condition module comprises a second plurality of aeroelastic estimators, wherein each aeroelastic estimator is tuned to a different presumptive fault condition of the wind turbine, and wherein determining the current operating condition further comprises:
generating, via the operating condition module, a plurality of descriptive operating parameters from the second plurality of aeroelastic estimators based, at least in part, on the current operational data set; and determining, via the operating condition module, the current operating condition of the wind turbine based on a designated portion of the plurality of descriptive operating parameters.

8 . The method of claim 7 , wherein determining the designated portion of the plurality of descriptive operating parameters comprises:
a) modelling, via the controller, a projected plurality of sensor outputs from the sensor system under one potential fault condition of a plurality of potential fau

Assignees

• General Electric Renovables Espana Sl
• Ge Vernova Renovables Espana S L

Inventors

• D'Amato Fernando Javier
• Achanta Hema Kumari
• Abbaszadeh Masoud
• Singal Kalpesh
• Dokucu Mustafa Tekin
• Fu Xu

Classifications

• F03D7/046

  with learning or adaptive control, e.g. self-tuning, fuzzy logic or neural network · CPC title

• F05B2260/821

  Parameter estimation or prediction · CPC title

• F05B2270/404

  active, predictive, or anticipative · CPC title

• F05B2260/84

  Modelling or simulation · CPC title

• Y02E10/72

  Wind turbines with rotation axis in wind direction · CPC title