Method and apparatus for controlling power of wind farm

Patent metadata

Field	Value
Publication number	US-12416287-B2
Application number	US-202218563578-A
Country	US
Kind code	B2
Filing date	Mar 25, 2022
Priority date	Nov 29, 2021
Publication date	Sep 16, 2025
Grant date	Sep 16, 2025

Abstract

The present disclosure provides a method and an apparatus for controlling power of a wind farm. The method includes: acquiring a historical equivalent fatigue load of a target component of each of a plurality of wind turbines; acquiring an entire fatigue load, an equivalent fatigue load of the target component and a pay per unit power generation of each of the plurality of wind turbines at a current moment; determining an optimal power value for at least one of the wind turbines according to the historical equivalent fatigue load, as well as the entire fatigue load, the equivalent fatigue load, and the pay per unit power generation; and adjusting a power value of the at least one of the wind turbines to the optimal power value for the wind turbine.

First claim

What is claimed is:

1. A method for controlling power of a wind farm, wherein the method for controlling power comprises:
acquiring a historical equivalent fatigue load of a target component of each of a plurality of wind turbines in the wind farm during a target historical period;
acquiring, in response to entering a limited power cycle, an entire fatigue load, an equivalent fatigue load of the target component and a pay per unit power generation of each of the plurality of wind turbines at a current moment, wherein the pay per unit power generation of a wind turbine is obtained based on the equivalent fatigue load and power generation of the wind turbine;
determining an optimal power value of at least one of the wind turbines according to the historical equivalent fatigue load as well as the entire fatigue load, the equivalent fatigue load and the pay per unit power generation, to ensure that a sum of power values of the plurality of wind turbines is less than or equal to a limited power value in the limited power cycle; and
sending a control instruction to the at least one of the wind turbines to adjust a power value of the at least one of the wind turbines to the optimal power value for the wind turbine.

2. The method for controlling power according to claim 1 , wherein the determining the optimal power value of the at least one of the wind turbines according to the historical equivalent fatigue load as well as the entire fatigue load, the equivalent fatigue load and the pay per unit power generation, to ensure that a sum of power values of the plurality of wind turbines is less than or equal to a limited power value in the limited power cycle comprises:
determining, for the plurality of wind turbines, the optimal power value according to the historical equivalent fatigue load as well as the entire fatigue load, the equivalent fatigue load and the pay per unit power generation on a basis of a wind turbine and in an order of the entire fatigue load from largest to smallest, until the sum of the power values of the plurality of wind turbines is less than or equal to the limited power value.

3. The method for controlling power according to claim 1 , wherein before the determining the optimal power value according to the historical equivalent fatigue load as well as the entire fatigue load, the equivalent fatigue load and the pay per unit power generation, the method further comprises:
acquiring a set of preset power values for a target wind turbine among the plurality of wind turbines; and
traversing the set of preset power values, re-determining an equivalent fatigue load and power generation of the target wind turbine at the current moment according to the respective preset power values, and re-determining the pay per unit power generation according to the equivalent fatigue load and the power generation.

4. The method for controlling power according to claim 3 , wherein the determining the optimal power value according to the historical equivalent fatigue load as well as the entire fatigue load, the equivalent fatigue load and the pay per unit power generation comprises:
determining reference indexes corresponding to the respective preset power values at the current moment, wherein the reference indexes comprises the entire fatigue load, the pay per unit power generation, and a difference between a consumed fatigue load and an average value of consumed fatigue loads of the plurality of wind turbines, and wherein the consumed fatigue load is a sum of the historical equivalent fatigue load, the equivalent fatigue load and the entire fatigue load; and
selecting, based on the reference indexes, one of the set of preset power values as the optimal power value, such that the reference indexes corresponding to the optimal power value are all less than the respective reference indexes corresponding to at least a part of other preset power values among the set of preset power values.

5. The method for controlling power according to claim 4 , wherein the selecting, based on the reference indexes, one of the set of preset power values as the optimal power value comprises:
determining an index product and an index sum of the reference indexes corresponding to each of the preset power values, and determining a ratio of the index product to the index sum as a comprehensive index corresponding to the preset power value; and
determining a preset power value with a minimum comprehensive index as the optimal power value, and
wherein the reference indexes each further comprises a sum of the consumed fatigue loads of the plurality of wind turbines.

6. The method for controlling power according to claim 3 , wherein
the set of preset power values comprises a plurality of preset power values which are obtained from a full power value of the target wind turbine subtracted gradually by a preset value, wherein a minimum one of the set of preset power values is greater than or equal to 0.

7. The method for controlling power according to claim 1 , wherein the historical equivalent fatigue load and the equivalent fatigue load are obtained by inputting operation data of each of the plurality of wind turbines during a corresponding period into a preset model trained by machine learning.

8. The method for controlling power according to claim 1 , wherein the entire fatigue load is obtained based on a wind resource parameter at the current moment, and
the wind resource parameter comprises statistical values of turbulence intensity under a plurality of respective wind speeds, probability densities of the plurality of respective wind speeds, and probability densities of a plurality of wind directions corresponding to the plurality of respective wind speeds, and wherein the entire fatigue load is obtained by accumulating the statistical values of turbulence intensity with consideration of the probability densities of the plurality of respective wind speeds and the probability densities of a plurality of wind directions corresponding to the plurality of respective wind speeds.

9. The method for controlling power according to claim 8 , wherein the statistical value of each turbulence intensity is a standard deviation of turbulence intensity, and wherein the entire fatigue load is obtained by: determining, for each of the wind turbines and for each of the wind speeds, a product of an m-th power of a standard deviation of turbulence intensity, a probability density of the wind speed, and a probability density of a corresponding wind direction, and determining a sum of the products as a reference load, wherein m is a material Wolter coefficient of the wind turbine;
determining a (1/m)-th power of the reference load as the entire fatigue load of the wind turbine.

10. The method for controlling power according to claim 1 , wherein the method for controlling power further comprises:
determining, during the limited power cycle, a sum of the historical equivalent fatigue load and the equivalent fatigue load as an updated historical equivalent fatigue load;
repeating the step of acquiring the entire fatigue load, the equivalent fatigue load of the target component and the pay per unit power generation of each of the plurality of wind turbines at a current moment to update the optimal power value.

11. A non-transitory computer readable storage medium having instructions which, when executed by at least one processor, cause the at least one processor to perform the method for controlling power according to claim 1 .

12. An apparatus for controlling power of a wind farm, wherein the apparatus for controlling power comprises:
at least one processor;
at least one memory storing computer-executable instructions,

Assignees

• Beijing Goldwind Science & Creation Windpower Equipment Co Ltd

Inventors

• Su Huili
• Zhang Pengfei
• Mi Yuxia
• Zhao Shipei

Classifications

• H02J2101/28

  Wind energy · CPC title

• H02J3/46 Primary

  Controlling the sharing of generated power between the generators, sources or networks · CPC title

• F05B2270/20

  to optimise the performance of a machine · CPC title

• F05B2270/1033

  Power (if explicitly mentioned) · CPC title

• F03D7/048

  controlling wind farms · CPC title