Epicyclic Gear Fault Diagnosis Based on Fault Characteristic Energy Extracting From Stator Current in Closed-Loop Control System

Fault diagnosis of epicyclic gears is critical for mitigating safety risks and minimizing economic losses in industrial applications. Among various diagnostic techniques, Motor Current Signature Analysis (MCSA) stands out as a non-invasive, torque-based method for detecting gear faults through electromagnetic signals. However, there is a lack of research on the impact of motor control on fault diagnosis, resulting in relatively low accuracy in fault diagnosis and localization. This study proposes a computational method to extract gear fault characteristic energy from stator currents. By exploiting the properties of closed-loop control systems, the gear fault characteristic energy is calculated using the gear fault characteristic frequency and the properties of a window function. Fault boundaries and localization indicators are subsequently derived based on the fault mechanisms. The proposed method enhances the accuracy of fault diagnosis and localization, without the need for parameter adjustment. Experimental validation is conducted using a test bench comprising a permanent magnet synchronous motor (PMSM), a swinging arm load, and a rotating vector (RV) reducer.