Wind turbine gearbox operation monitoring using high-resolution distributed fiber-optic sensing
Linqing Luo (Lawrence Berkeley National Laboratory)
Unai Gutierrez Gutierrez Santiago (TU Delft - Team Jan-Willem van Wingerden, Siemens Gamesa Renewable Energy)
Yuxin Wu (Lawrence Berkeley National Laboratory)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Efficient gearbox monitoring is vital for improving fault detection, enhancing design, and reducing operation and maintenance (O&M) costs, particularly for offshore wind turbines. This paper introduces an innovative approach using high-resolution distributed fiber-optic sensing (DFOS) based on optical frequency domain reflectometry (OFDR) to measure gearbox strain in real time. By bonding a single optical fiber around the full circumference of the outer surface of a 2.152 m diameter ring gear of the first planetary stage in a 3.75 MW wind turbine gearbox, we measured circumferential strain from planetary gear passage every 2.6 mm around the ring gear under different input torque levels. Our results show accurate identification of planet gear locations in real time and rotation speed (10.42 revolutions per minute), with a strong linear correlation (R2= 0.9997) between applied torque and measured strain across all 2500 measured locations. Strain variations of approximately 200 microstrains were observed on the ring gear exterior from each (planet passage or planet-ring gear mesh) event, providing granular insights into mechanical behavior and load distribution. Additionally, DFOS also showed its potential for detecting temperature variations during operation, indicating its potential for concurrently monitoring thermal and mechanical anomalies. This study represents the first application of continuous DFOS to a full-scale wind turbine gearbox. The approach offers a scalable and practical solution for early fault detection and the support of design validation and, when combined with modeling, can lead to a more durable and optimized design.