Gear Fault Diagnosis and Evaluation Based on MCSA in Gear Phase-domain

Gears are one of the most important load-bearing structures of mechanical systems. Motor current signature analysis (MCSA) provides a non-invasive and torque-related approach for gear fault diagnosis. However, there are still problems that the diagnosis result on early failure is not effective, and static transmission error characteristics are similar to the fault characteristics. Therefore, a gear fault diagnosis method based on MCSA in gear phase-domain is proposed. This method replaces the frequency domain analysis with direct feature extraction in the phase domain to solve the above problems. Gear fault information is extracted under interference such as installation eccentricity error. The relationship between failure features and load torque are studied by Simulink simulation. The feature of stiffness decline become obvious as the load torque increase while the wear fault does not. Three indicators are proposed to distinguish static transmission error, tooth surface wear and stiffness drop faults. Experiments have been carried out for partial validation.