Load control method and apparatus for wind turbine generator system

Patent metadata

Field	Value
Publication number	US-12123399-B2
Application number	US-202118043662-A
Country	US
Kind code	B2
Filing date	Jun 10, 2021
Priority date	Sep 1, 2020
Publication date	Oct 22, 2024
Grant date	Oct 22, 2024

Abstract

A load control method and a load control apparatus for a wind turbine generator system are provided, and the load control method includes: obtaining feature parameters of the wind turbine generator system for load prediction; obtaining a load estimation value of the wind turbine generator system by inputting the obtained feature parameters into a virtual load sensor; adjusting a control strategy of the wind turbine generator system based on the obtained load estimation value. A controller and a computer readable storage medium storing a computer program are further included. With the load control method and apparatus for the wind turbine generator system, a trained virtual load sensor can be used to realize real-time monitoring of the load of the on-site wind turbine generator system, and a reference for adjusting the control strategy can be provided according to the load.

First claim

What is claimed is:

1. A load control method for a wind turbine generator system, wherein the load control method comprises:
obtaining feature parameters of the wind turbine generator system for load prediction;
obtaining a load estimation value of the wind turbine generator system by inputting the obtained feature parameters into a virtual load sensor;
adjusting a control strategy of the wind turbine generator system based on the obtained load estimation value,
wherein the feature parameters for the load prediction comprise: a configuration parameter of the wind turbine generator system, a control parameter corresponding to the control strategy used, an operation parameter of the wind turbine generator system, and a preset flag; and
the preset flag comprises a preset event flag and/or a preset fault flag,
wherein the load estimation value of the wind turbine generator system comprises a real-time load estimation value at a current moment and a load prediction value after a predetermined duration,
the step of adjusting the control strategy of the wind turbine generator system based on the obtained load estimation value comprises:
controlling the wind turbine generator system to perform a shutdown control strategy or a load reduction control strategy based on the real-time load estimation value and the load prediction value of the wind turbine generator system,
wherein the step of controlling the wind turbine generator system to perform the shutdown control strategy or the load reduction control strategy based on the real-time load estimation value and the load prediction value of the wind turbine generator system comprises:
comparing the real-time load estimation value to a load risk threshold;
controlling the wind turbine generator system to perform the shutdown control strategy and sending an alarm signal under a condition that the real-time load estimation value is not less than the load risk threshold;
comparing the real-time load estimation value to a load warning threshold under a condition that the real-time load estimation value is less than the load risk threshold, wherein the load warning threshold is less than the load risk threshold;
comparing the load prediction value to the load risk threshold under a condition that it is determined that the real-time load estimation value is not less than the load warning threshold;
controlling the wind turbine generator system to perform the load reduction control strategy under a condition that the load prediction value is not less than the load risk threshold.

2. The load control method according to claim 1 , wherein the step of controlling the wind turbine generator system to perform the load reduction control strategy comprises:
determining whether the preset flag includes a flag having a value that indicates an event indicated by the flag occurs;
determining, based on a corresponding control strategy when the event indicated by the flag occurs, an adjustable control parameter under a condition that the preset flag includes the flag having the value that indicates the event indicated by the flag occurs;
determining a predefined control parameter as the adjustable control parameter under a condition that the preset flag does not include the flag having the value that indicates the event indicated by the flag occurs;
re-determining a load estimation value by using the determined adjustable control parameter to replace a corresponding control parameter of the feature parameters for the load prediction.

3. The load control method according to claim 1 , wherein the virtual load sensor is trained by:
obtaining simulation data of the wind turbine generator system under predetermined operating conditions;
building, based on the simulation data, a feature matrix and a target matrix, wherein the feature matrix is composed of the feature parameters for the load prediction extracted from the simulation data, and the target matrix is composed of a real-time load value at a current moment and a load value after a predetermined duration extracted from the simulation data;
training the virtual load sensor by using the feature matrix as an input of the virtual load sensor and the target matrix as an output of the virtual load sensor.

4. The load control method according to claim 3 , further comprising:
correcting the virtual load sensor using a real load measurement value of the wind turbine generator system collected by a physical load sensor.

5. A load control apparatus for a wind turbine generator system, wherein the load control apparatus comprises:
a feature parameter obtaining module configured to obtain feature parameters used by a wind turbine generator system for load prediction;
a load estimation module configured to obtain a load estimation value of the wind turbine generator system by inputting the obtained feature parameters into a virtual load sensor;
a control strategy adjusting module configured to adjust a control strategy of the wind turbine generator system based on the obtained load estimation value,
wherein the feature parameters for the load prediction comprise: a configuration parameter of the wind turbine generator system, a control parameter corresponding to the control strategy used, an operation parameter of the wind turbine generator system, and a preset flag; and
the preset flag comprises a preset event flag and/or a preset fault flag,
wherein the load estimation value of the wind turbine generator system comprises a real-time load estimation value at a current moment and a load prediction value after a predetermined duration,
the control strategy adjusting module is further configured for:
controlling the wind turbine generator system to perform a shutdown control strategy or a load reduction control strategy based on the real-time load estimation value and the load prediction value of the wind turbine generator system,
wherein controlling the wind turbine generator system to perform the shutdown control strategy or the load reduction control strategy based on the real-time load estimation value and the load prediction value of the wind turbine generator system comprises:
comparing the real-time load estimation value to a load risk threshold;
controlling the wind turbine generator system to perform the shutdown control strategy and sending an alarm signal under a condition that the real-time load estimation value is not less than the load risk threshold;
comparing the real-time load estimation value to a load warning threshold under a condition that the real-time load estimation value is less than the load risk threshold, wherein the load warning threshold is less than the load risk threshold;
comparing the load prediction value to the load risk threshold under a condition that it is determined that the real-time load estimation value is not less than the load warning threshold;
controlling the wind turbine generator system to perform the load reduction control strategy under a condition that the load prediction value is not less than the load risk threshold.

6. A controller comprising:
a processor;
memory for storing a computer program which, when executed by the processor, causes the processor to perform operations including:
obtaining feature parameters of the wind turbine generator system for load prediction;
obtaining a load estimation value of the wind turbine generator system by inputting the obtained feature parameters into a virtual load sensor;
adjusting a control strategy of the wind turbine generator system based on the obtained load estimation value,
wherein the feature parameters for the load prediction comprise: a configuration parameter of the wind turbine generator system, a control parameter corresponding to the control strategy used, an operation parameter of the

Assignees

• Xinjiang Goldwind Science & Tech Co Ltd

Inventors

• Kang Weixin
• Zhou Guilin
• Zhang Pengfei
• Huang Xiaofang

Classifications

• H02P9/008

  wherein the generator is controlled by the requirements of the prime mover · CPC title

• F05B2270/331

  Mechanical loads · CPC title

• H02P2101/15

  for wind-driven turbines · CPC title

• F03D7/046

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

• F03D17/006

  Estimation methods · CPC title