Oil-failure predictor for wind-turbine gearbox based on extreme gradient boosting and SCADA

Qitan Sun; Nana Lu; Gaojuan Li; Yusheng Liu

doi:10.1117/12.3015626

6 February 2024 Oil-failure predictor for wind-turbine gearbox based on extreme gradient boosting and SCADA

Qitan Sun, Nana Lu, Gaojuan Li, Yusheng Liu

Proceedings Volume 12979, Ninth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023); 129794Q (2024) https://doi.org/10.1117/12.3015626
Event: 9th International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023), 2023, Guilin, China

Abstract

The wide application of the Supervisory-Control and Data-Acquisition (SCADA) system in Wind Turbines (WT) enables users to assess their performance from multi-perspectives. However, data redundancy and dirty information are also an inevitable issue following along with such a system, making it hard to well distinguish and utilize useful data. In this paper, a gearbox-oil monitoring structure based on the SCADA system is proposed, which can automatically handle data-preprocessing, data analysis and performance evaluation. In the first stage, an Extreme Gradient Boosting (XGBoost) model is established to identify oil-element-relevant failures with well-preprocessed data collected from the proper- and mal-functioning period of WTs’ gearboxes. Later professional expertise is incorporated into the model to construct the threshold that an abnormal state should exceed in order to evoke alarms. Finally, the whole setup scores related oil element according to the previous analysis for the maintenance department to determine fixing priority. The effectiveness of the overall architecture is justified by being tested on a series of WTs in a specific wind farm.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Qitan Sun, Nana Lu, Gaojuan Li, and Yusheng Liu "Oil-failure predictor for wind-turbine gearbox based on extreme gradient boosting and SCADA", Proc. SPIE 12979, Ninth International Conference on Energy Materials and Electrical Engineering (ICEMEE 2023), 129794Q (6 February 2024); https://doi.org/10.1117/12.3015626

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available