Study on Coupling Faults Diagnosis of Gear Rotor System

Abstract

Coupling faults diagnose methods of gear rotor system are studied. Moment invariants are extracted from axis orbit of rotor system, which used to identify imbalances, loosening faults. Mode functions with different characteristic time scales are obtained based on the EMD method. Marginal spectrum is obtained by Hilbert transform, which can used to identify the wear faults of gear system. Moment invariant and BP neural network are used to identify the unbalanced fault and loosening faults. Hilbert transform method is used to diagnose gear wear fault. Experiments prove the effective of the methods.