Root Crack Identification of Sun Gear in Planetary Gear System Combining Fault Dynamics with VMD Algorithm

Abstract

Planetary gearbox is widely used in various low-speed machines due to its large transmission ratio. However, the gears in a planetary gearbox are prone to the mechanical faults due to the complex dynamic heavy load. Vibration frequencies caused by an early tooth root crack of sun gear are usually difficult to accurately extract, so its fault diagnosis is one of the main challenges of planetary gearbox reliability. In this paper, a simplified tooth root crack model of sun gear is proposed, and then a rigid-flexible coupled dynamics model of the whole planetary gear system is constructed. By the numerical simulation, the fault frequencies caused by a tooth root crack of sun gear are obtained. A Variational Mode Decomposition (VMD) algorithm for the vibration frequency extraction is proposed. The measured vibration signals are decomposed into the sparse Intrinsic Mode Functions (IMFs) by the VMD, and then the IMFs are further analyzed by the spectral method to accurately extract the crack-induced frequency components. The experimental results show that the proposed dynamics model and VMD method are feasible; an error between the characteristic frequencies from the tested signal analysis and the theoretical calculation is less than 1%.