TY - GEN
T1 - A RP image-based health indicator construction method for wind turbine RUL prediction
AU - Han, Danyang
AU - Yu, Jinsong
AU - Tang, Diyin
AU - Kong, Lingkun
AU - Li, Xin
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - New clear energy is critical to prevent global environment pollution and as one of the major wind energy conductors, wind turbines play important roles. Considering low-sampling rate and high noise from Supervisory Control and Data Acquisition (SCADA) system, a novel image-based method with deep learning model is proposed for remaining useful life (RUL) prediction of wind turbine gearbox in this paper. Firstly, recurrent plot image method is introduced for transforming one-dimension temperature signals to two-dimension images to describe dependency of time series. Then, a residual-based autoencoder model is utilized for deeper feature extraction from images. Finally, a end-to-end model based on health indicator labels is used for RUL prediction. An experiment result based on signals from SCADA validated the effectiveness of proposed method.
AB - New clear energy is critical to prevent global environment pollution and as one of the major wind energy conductors, wind turbines play important roles. Considering low-sampling rate and high noise from Supervisory Control and Data Acquisition (SCADA) system, a novel image-based method with deep learning model is proposed for remaining useful life (RUL) prediction of wind turbine gearbox in this paper. Firstly, recurrent plot image method is introduced for transforming one-dimension temperature signals to two-dimension images to describe dependency of time series. Then, a residual-based autoencoder model is utilized for deeper feature extraction from images. Finally, a end-to-end model based on health indicator labels is used for RUL prediction. An experiment result based on signals from SCADA validated the effectiveness of proposed method.
KW - deep learning
KW - image construction
KW - prognostic and health management
KW - wind turbine gearbox
UR - https://www.scopus.com/pages/publications/85134433731
U2 - 10.1109/I2MTC48687.2022.9806512
DO - 10.1109/I2MTC48687.2022.9806512
M3 - 会议稿件
AN - SCOPUS:85134433731
T3 - Conference Record - IEEE Instrumentation and Measurement Technology Conference
BT - I2MTC 2022 - IEEE International Instrumentation and Measurement Technology Conference
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE International Instrumentation and Measurement Technology Conference, I2MTC 2022
Y2 - 16 May 2022 through 19 May 2022
ER -