Early Fault Diagnosis for Wind Turbine Gearbox Based on Multi-Source Feature Fusion
Li Wang a, He Jianjun b, Wang Jiawei c, Tang Zhiwei d, Jie Jun e, Luo Guangwu f and Liu yan f
aEngineer, CHN Energy Hunan Power Co., Ltd, ChangSha 410000, China
bProfessor, School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China
cGraduate Student, School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China
dGraduate Student, School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China
eEngineer, Longyuan Jiangyong Wind Power Generation Co., Ltd of CHN Energy Group, Changsha 410000, China
fEngineer, Longyuan Jiangyong Wind Power Generation Co., Ltd of CHN Energy Group, Changsha 410000, China
|
Abstract:
As the primary moving part of a wind turbine, the
gearbox has a high failure rate and is particularly
detrimental to the device. The diagnosis of early
gearbox problem signals is less effective using the
typical vibration detection techniques now in use.
Considering this, Based on the KPCA-VMD approach,
this research offers a wind turbine gearbox early fault
monitoring and multidimensional feature assessment
method for analyzing wind turbine gearbox
inconspicuous early failure signals. Firstly, the preprocessed dataset is subjected to feature extraction, the
gearbox feature data is downscaled and reconstructed
by the KPCA method, the gearbox status is monitored
using two statistics, T2 and SPE, and the monitored
abnormal signals are analysed by VMD. The
experimental data show that the method can
effectively diagnose the gear early failure
characteristic frequency
|