US 12,442,360 B2
Method for computer-implemented monitoring of a component of a wind turbine
Niels Lovmand Pedersen, Gedved (DK)
Assigned to SIEMENS GAMESA RENEWABLE ENERGY A/S, Brande (DK)
Appl. No. 17/612,005
Filed by Siemens Gamesa Renewable Energy A/S, Brande (DK)
PCT Filed Jun. 2, 2020, PCT No. PCT/EP2020/065196
§ 371(c)(1), (2) Date Nov. 17, 2021,
PCT Pub. No. WO2020/245108, PCT Pub. Date Dec. 10, 2020.
Claims priority of application No. 19178834 (EP), filed on Jun. 6, 2019.
Prior Publication US 2022/0228569 A1, Jul. 21, 2022
Int. Cl. F03D 17/00 (2016.01); F03D 15/00 (2016.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01)
CPC F03D 17/00 (2016.05) [F03D 15/00 (2016.05); G06N 3/045 (2023.01); G06N 3/08 (2013.01); F05B 2240/50 (2013.01); F05B 2260/84 (2013.01); F05B 2270/334 (2013.01); F05B 2270/709 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A method for computer-implemented monitoring of a component of a wind turbine, where the wind turbine is a first wind turbine and the component is a first component, the method comprising:
i) operating the first wind turbine to produce vibration signals in a predetermined domain that are measured in a vicinity of the first component the operating of the first wind turbine;
ii) mapping the vibration signals to corresponding vibration signals valid for a second component of a second wind turbine based on one or more given kinematic parameters of the first component and one or more given kinematic parameters of the second component; and
iii) feeding, as an input, the corresponding vibration signals valid for the second component into a trained machine learning model, the trained machine learning model being trained for the second component of the second wind turbine, which is a same type as the first component, the second wind turbine being of another type than the first wind turbine, wherein the trained machine learning model is configured to provide an output referring to a predetermined fault occurring at the second component of the second wind turbine by processing vibration signals in a predetermined domain which are measured in a vicinity of the second component during an operation of the second wind turbine; and
iv) generating an output from the trained machine learning model indicative of a predetermined fault of the first component of the first component.