Wind Turbine Gearbox Fault Diagnosis Based on Improved HHT

Abstract

The randomness of wind power and the nonstationarity of gearbox vibration signals greatly increased the difficulty of wind turbine gearbox faults feature diagnosis. Studies proved that Hilbert-Huang transform (HHT) has a satisfying performance when it be used to analyze nonlinear and non-stationary signals, but two troublesome problems in application of HHT are end effect problem and false intrinsic mode functions (IMFs). An improved solution method based on adaptive waveform matching endpoint extension is proposed to solve end effect problem. Meanwhile, an improved energy conservation method is proposed to eliminate false IMFs. Tests prove that improved HHT can greatly improve the accuracy of wind turbine gearbox faults diagnosis and with a balance efficiency.