Control of a wind turbine taking noise into account

• US 10,697,431 B2
• Filed: 04/03/2017
• Issued: 06/30/2020
• Est. Priority Date: 04/07/2016
• Status: Active Grant

• Current Assignee: Vestas Wind Systems A/S
• Original Assignee: Vestas Wind Systems A/S
• Inventors: Hammerum, Keld, Hovgaard, Tobias Gybel, Sloth, Erik
• Primary Examiner(s): Patel, Ramesh B
• Application Number:
  US16/091,365
• Publication Number:
  US 20190154000A1
• Time in Patent Office: 1,184 Days
• Field of Search: None
• CPC Class Codes: F03D 7/0296 to prevent, counteract or reduce noise emissions
• CPC Class Codes: F03D 7/045 with model-based controls
• CPC Class Codes: F05B 2270/333 Noise or sound levels
• CPC Class Codes: F05B 2270/404 active, predictive, or anticipative
• CPC Class Codes: Y02E 10/72 Wind turbines with rotation axis in wind direction

Abstract

The present invention relates to control of wind turbines where a noise measure is taken into account. Control of a wind turbine is described where a control trajectory is calculated based on noise measure, the noise measure being determined from a predicted operational trajectory. In embodiments the predicted operational trajectories are calculated by using a model predictive control (MPC) routine.

Claims

1. A method of controlling a wind turbine, the method comprising:

calculating at least one predicted operational trajectory based on a current operational state of the wind turbine, wherein the at least one predicted operational trajectory comprises a time series of at least one variable;

calculating at least one predicted noise measure based at least on the at least one predicted operational trajectory; and

controlling the wind turbine according to at least one control trajectory calculated based at least on the at least one predicted noise measure.

2. The method according to claim 1, wherein calculating the at least one predicted operational trajectory comprises:

optimizing at least one cost function that includes the at least one predicted noise measure.

3. The method according to claim 1, wherein calculating that at least one predicted operational trajectory comprises:

optimizing at least one cost function,

wherein the at least one predicted noise measure is included as an optimization constraint.

4. The method according to claim 1,

wherein the at least one predicted operational trajectory is a receding horizon trajectory with a prediction horizon, and

wherein the at least one predicted noise measure is determined for the prediction horizon.

5. The method according to claim 1, wherein the at least one predicted operational trajectory is calculated by using a model predictive control (MPC) routine.

6. The method according to claim 1, wherein the at least one predicted noise measure is calculated by a wind turbine noise emission model that predicts noise of the wind turbine as a function of one or more operational parameters over a prediction horizon.

7. The method according to claim 6, wherein the wind turbine noise emission model is configured to calculate a noise level at a specified immission point.

8. The method according to claim 2, wherein the cost function comprises a weighted element including the at least one predicted noise measure.

9. The method according to claim 8, wherein a weight of the weighted element is correlated with one of a measured atmospheric condition and a measured background noise level.

10. The method according to claim 8, wherein a weight of the weighted element is correlated with an estimated noise level determined by a wind turbine noise emission model.

11. The method according to claim 8, wherein a weight of the weighted element is correlated with a function of one or both of a time of day and a date.

12. The method according to claim 3, wherein the optimization constraint comprises maintaining the at least one predicted noise measure less than or equal to a predefined level.

13. The method according to claim 12, wherein the predefined level is received as an input from an external source.

14. The method according to claim 12, wherein the predefined level is correlated with one of:

a function of one or both of a time of day and a date; and

a measured background noise level.

15. The method according to claim 1, wherein the at least one predicted noise measure includes one or more tonal components.

16. A method of controlling a wind power plant comprising a plurality of wind turbines, the method comprising:

calculating, based on a current operational state of a selected wind turbine of the plurality of wind turbines, one or more predicted operational trajectories for the selected wind turbine, wherein the one or more predicted operational trajectories comprise a time series of at least one variable;

calculating at least one predicted noise measure using at least one of the one or more predicted operational trajectories; and

controlling the selected wind turbine according to at least one control trajectory calculated based at least on the at least one predicted noise measure.

17. The method according to claim 16,

wherein the at least one predicted noise measure comprises an aggregate noise measure that is determined for a specified immission point based on a plurality of noise measures determined for individual wind turbines of the plurality of wind turbines.

18. The method according to claim 16, wherein calculating the one or more predicted operational trajectories comprises:

optimizing at least one cost function,

wherein the at least one predicted noise measure is one or both of;

included in the cost function, and

included as an optimization constraint.

19. A control system for a wind turbine, the control system comprising:

a controller configured to;

receive a current operational state of the wind turbine;

calculate, based on the current operational state, one or more predicted operational trajectories comprising a time series of at least one variable;

calculate at least one predicted noise measure based at least on at least one predicted operational trajectory of the one or more predicted operational trajectories;

calculate a control trajectory based at least on the at least one predicted noise measure; and

control the wind turbine based on the control trajectory.