Control of a wind turbine based on predicting amplitude modulation (AM) noise

1 . A method, performed by an electronic device, for enabling control of operation of a wind turbine, the method comprising:

obtaining wind turbine data associated with the wind turbine, wherein the wind turbine data is indicative of conditions of operation of the wind turbine;

obtaining a measured Amplitude Modulation (AM) noise parameter indicative of measured AM noise in a region at a distance from the wind turbine;

predicting an AM noise parameter indicative of AM noise within the region at the distance from the wind turbine by applying a machine learning model to the wind turbine data;

comparing the measured AM noise parameter and the predicted AM noise parameter;

determining, based on the comparison between the measured AM noise parameter and the predicted AM noise parameter, a source of the AM noise near the wind turbine;

generating, based on the AM noise parameter, control data indicative of a control operation of the wind turbine, wherein the control data comprises:

determining whether the measured AM noise parameter exceeds a first threshold;

determining whether the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies a second threshold; and

only after determining that the measured AM noise parameter exceeds the first threshold and that the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies the second threshold, generating the control data indicating that the control operation is to be triggered; and

controlling the wind turbine in accordance with the control data.

2 . The method according to claim 1 , wherein the machine learning model is trained using previous wind turbine data and measured AM noise data.

3 . The method according to claim 1 , wherein obtaining the measured AM noise parameter comprises:

filtering measurements; and

obtaining, based on the filtered measurements, the measured AM noise parameter.

4 . The method according to claim 1 , wherein generating, based on the AM noise parameter, the control data comprises:

upon determining that the measured AM noise parameter does not exceed the first threshold or that the comparison between the measured AM noise parameter and the predicted AM noise parameter does not satisfy the second threshold, generating the control data indicating that the control operation is not to be triggered.

5 . The method according to claim 1 , the method comprising ranking, based on the comparison, the predicted AM noise parameter; wherein generating, based on the AM noise parameter, the control data comprises:

generating, based on the ranking, the control data.

6 . The method according to claim 5 , the method comprising:

monitoring measurements of AM noise after execution of the control operation; and

determining, based on the monitoring, a reduction parameter indicative of a reduction of the AM noise.

7 . The method according to claim 6 , comprising storing reduction data, wherein the reduction data comprises one or more of: the control operation, the reduction parameter associated with the control operation, and one or more environmental condition parameters indicative of environmental conditions during execution of the control operation.

8 . The method according to claim 7 , comprising scoring the reduction data based on the reduction parameter.

9 . The method according to claim 7 comprising updating, based on the reduction data, the ranking of the predicted AM noise parameter.

10 . The method according to claim 7 , wherein generating, based on the AM noise parameter, the control data comprises:

generating, based on the stored reduction data, a target reduction parameter indicative of a target reduction of the AM noise; and

including the target reduction parameter in the control data.

11 . The method according to claim 1 comprising:

upon determining that the comparison between the measured AM noise parameter and the predicted AM noise parameter does not satisfy the second threshold, re-training the machine learning model based on the measured AM noise parameter.

12 . An electronic device configured to perform an operation controlling a wind turbine, the operation comprising:

obtaining wind turbine data associated with the wind turbine, wherein the wind turbine data is indicative of conditions of operation of the wind turbine;

obtaining a measured Amplitude Modulation (AM) noise parameter indicative of measured AM noise in a region at a distance from the wind turbine;

predicting an AM noise parameter indicative of AM noise within the region at the distance from the wind turbine by applying a machine learning model to the wind turbine data;

comparing the measured AM noise parameter and the predicted AM noise parameter;

determining, based on the comparison between the measured AM noise parameter and the predicted AM noise parameter, a source of the AM noise near the wind turbine;

generating, based on the AM noise parameter, control data indicative of a control operation of the wind turbine, wherein the control data comprises:

determining whether the measured AM noise parameter exceeds a first threshold;

determining whether the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies a second threshold; and

only after determining that the measured AM noise parameter exceeds the first threshold and that the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies the second threshold, generating the control data indicating that the control operation is to be triggered; and

controlling the wind turbine in accordance with the control data.

13 . A wind turbine system, comprising:

a tower;

a nacelle disposed on the tower;

a rotor extending from the nacelle and having a plurality of blades at one end thereof; and

an electronic device configured to control an operation, comprising:

obtaining wind turbine data associated with the wind turbine system, wherein the wind turbine data is indicative of conditions of operation of the wind turbine system;

obtaining a measured Amplitude Modulation (AM) noise parameter indicative of measured AM noise in a region at a distance from the wind turbine system;

predicting an AM noise parameter indicative of AM noise within the region at the distance from the wind turbine system by applying a machine learning model to the wind turbine data;

comparing the measured AM noise parameter and the predicted AM noise parameter;

determining, based on the comparison between the measured AM noise parameter and the predicted AM noise parameter, a source of the AM noise near the wind turbine system;

generating, based on the AM noise parameter, control data indicative of a control operation of the wind turbine system, wherein the control data comprises:

determining whether the measured AM noise parameter exceeds a first threshold;

determining whether the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies a second threshold; and

only after determining that the measured AM noise parameter exceeds the first threshold and that the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies the second threshold, generating the control data indicating that the control operation is to be triggered; and

controlling the wind turbine system in accordance with the control data.

14 . A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed on any combination of one or more processors, carry out an operation controlling a wind turbine, the operation comprising:

obtaining wind turbine data associated with the wind turbine, wherein the wind turbine data is indicative of conditions of operation of the wind turbine;

obtaining a measured Amplitude Modulation (AM) noise parameter indicative of measured AM noise in a region at a distance from the wind turbine;

predicting an AM noise parameter indicative of AM noise within the region at the distance from the wind turbine by applying a machine learning model to the wind turbine data;

comparing the measured AM noise parameter and the predicted AM noise parameter;

determining, based on the comparison between the measured AM noise parameter and the predicted AM noise parameter, a source of the AM noise near the wind turbine;

generating, based on the AM noise parameter, control data indicative of a control operation of the wind turbine, wherein the control data comprises:

determining whether the measured AM noise parameter exceeds a first threshold;

determining whether the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies a second threshold; and

only after determining that the measured AM noise parameter exceeds the first threshold and that the comparison between the measured AM noise parameter and the predicted AM noise parameter satisfies the second threshold, generating the control data indicating that the control operation is to be triggered; and

controlling the wind turbine in accordance with the control data.