Inter-shaft bearing fault diagnosis based on TIEgram and autocorrelation

Inter-shaft bearing is a crucial supporting component of a dual rotor structure aviation engine. Its structural and operational characteristics make it prone to frequent and highly hazardous faults, which can easily lead to catastrophic accidents such as the failure of the entire rotor system. Therefore, it is significant to realize vibration fault monitoring and warning of inter-shaft bearings. Aero-engine vibration has complex characteristics, with diagnostic methods facing major limitations. Firstly, the compact engine structure and inter-shaft bearing placement cause weak, attenuated fault signals with significant interference, complicating feature extraction. Secondly, in counter-rotating dual-rotor systems, inter-shaft bearing components rotate oppositely, producing higher characteristic frequencies than co-rotating bearings. High-frequency fault signals often overlap or appear harmonic, and their propagation is easily affected by structural complexities, making accurate monitoring challenging. To address these challenges, a method combining Traversal Index Enhanced-gram (TIEgram) and autocorrelation (AC) is proposed for extracting weak fault features in inter-shaft bearings. TIEgram selects the optimal frequency band for resonance demodulation, isolating fault-related signal components. To counter non-stationary signals from aero-engine dynamics, slip-ratio domain order tracking transforms time-domain signals into angular-domain stationary signals. Autocorrelation analysis then yields the squared envelope autocorrelation spectrum, compared with bearing fault characteristic orders for diagnosis. Simulation and experimental results demonstrate the method's effectiveness in extracting weak inter-shaft bearing fault features.