LEARNING-BASED BACKUP CONTROLLER FOR A WIND TURBINE

Abstract

A method for providing backup control for a supervisory controller of at least one wind turbine includes observing, via a learning-based backup controller of the at least one wind turbine, at least one operating parameter of the supervisory controller under normal operation. The method also includes learning, via the learning-based backup controller, one or more control actions of the at least one wind turbine based on the operating parameter(s). Further, the method includes receiving, via the learning-based backup controller, an indication that the supervisory controller is unavailable to continue the normal operation. Upon receipt of the indication, the method includes controlling, via the learning-based backup controller, the wind turbine(s) using the learned one or more control actions until the supervisory controller becomes available again. Moreover, the control action(s) defines a delta that one or more setpoints of the wind turbine(s) should be adjusted by to achieve a desired outcome.

Description

Claims

1. A method for providing backup control for a supervisory controller of at least one wind turbine, the method comprising: observing, via a learning-based backup controller of the at least one wind turbine, at least one operating parameter of the supervisory controller under normal operation;learning, via the learning-based backup controller, one or more control actions of the at least one wind turbine based on the at least one operating parameter;receiving, via the learning-based backup controller, an indication that the supervisory controller is unavailable to continue the normal operation; andupon receipt of the indication, controlling, via the learning-based backup controller, the at least one wind turbine using the learned one or more control actions until the supervisory controller becomes available again,wherein the one or more control actions defines a delta that one or more setpoints of the at least one wind turbine should be adjusted by to achieve a desired outcome.

2. The method of claim 1, wherein observing the at least one operating parameter of the supervisory controller under normal operation further comprises: observing a plurality of operating parameters of the supervisory controller under normal operation.

3. The method of claim 2, wherein the plurality of operating parameters comprises at least one of power output, pitch angle, generator speed, measured wind speed, or estimated wind speed.

4. The method of claim 2, wherein the plurality of operating parameters comprises a combination of measured and estimated operating parameters.

5. The method of claim 2, wherein learning, via the learning-based backup controller, the one or more control actions of the wind turbine based on the at least one operating parameter further comprises: generating a plurality of input/output tuple combinations based on the plurality of operating parameters; andtraining, via a machine learning algorithm of the learning-based backup controller, the one or more control actions of the wind turbine based on the plurality of input/output tuple combinations.

6. The method of claim 5, wherein the machine learning algorithm comprises at least one of a deep neural network, a recurrent neural network, or a convolutional neural network, or an extreme learning machine.

7. The method of claim 1, wherein controlling, via the learning-based backup controller, the at least one wind turbine using the learned one or more control actions until the supervisory controller becomes available again further comprises: sending, via the learning-based backup controller, the learned one or more control actions to the at least one wind turbine during a time period in which the supervisory controller is unavailable; andcontinuously training the learned one or more control actions based on additional operating parameters received during the time period.wherein the one or more control actions defines a delta that one or more setpoints of the wind turbine should be adjusted by to achieve a desired outcome.

8. The method of claim 1, wherein the delta that one or more setpoints of the at least one wind turbine should be adjusted by to achieve the desired outcome comprises at least one of a pitch angle delta or a torque delta.

9. The method of claim 1, wherein the supervisory controller is a model-based controller.

10. The method of claim 1, wherein the supervisory controller is a turbine controller of the at least one wind turbine or a farm-level controller of a wind farm containing the at least one wind turbine.

11. The method of claim 1, wherein the supervisory controller is unavailable due to at least one of a cyberattack, a hardware failure, or a system fault.

12. A system for providing backup control for a supervisory controller of at least one wind turbine, the system comprising: a supervisory controller;a learning-based backup controller communicatively coupled to the supervisory controller, the learning-based backup controller comprising at least one processor configured to perform a plurality of operations, the plurality of operations comprising: observing at least one operating parameter of the supervisory controller under normal operation;learning one or more control actions of the at least one wind turbine based on the at least one operating parameter; andcontrolling the at least one wind turbine using the learned one or more control actions when the supervisory controller is unavailable,wherein the one or more control actions defines a delta that one or more setpoints of the at least one wind turbine should be adjusted by to achieve a desired outcome.

13. The system of claim 12, wherein observing the at least one operating parameter of the supervisory controller under normal operation further comprises: observing a plurality of operating parameters of the supervisory controller under normal operation.

14. The system of claim 13, wherein the plurality of operating parameters comprises at least one of power output, pitch angle, generator speed, measured wind speed, or estimated wind speed.

15. The system of claim 13, wherein the plurality of operating parameters comprises a combination of measured and estimated operating parameters.

16. The system of claim 13, wherein learning, via the learning-based backup controller, the one or more control actions of the wind turbine based on the at least one operating parameter further comprises: generating a plurality of input/output tuple combinations based on the plurality of operating parameters; andtraining, via a machine learning algorithm of the learning-based backup controller, the one or more control actions of the wind turbine based on the plurality of input/output tuple combinations.

17. The system of claim 16, wherein the machine learning algorithm comprises at least one of a deep neural network, a recurrent neural network, or a convolutional neural network, or an extreme learning machine.

18. The system of claim 12, wherein controlling, via the learning-based backup controller, the at least one wind turbine using the learned one or more control actions until the supervisory controller becomes available again further comprises: sending, via the learning-based backup controller, the learned one or more control actions to the at least one wind turbine during a time period in which the supervisory controller is unavailable; andcontinuously training the learned one or more control actions based on additional operating parameters received during the time period.wherein the one or more control actions defines a delta that one or more setpoints of the wind turbine should be adjusted by to achieve a desired outcome.

19. The system of claim 12, wherein the delta that one or more setpoints of the at least one wind turbine should be adjusted by to achieve the desired outcome comprises at least one of a pitch angle delta or a torque delta.

20. The system of claim 12, wherein the supervisory controller is a model-based controller, wherein the model-based controller is a turbine controller of the at least one wind turbine or a farm-level controller of a wind farm containing the at least one wind turbine.