Dynamic Modelling and Vibration Analysis for Gear Tooth Crack Detection

Sammanfattning: The aim of applying vibration-based condition monitoring in gear systems is to detect the initiation and development of degradation before the occurrence of failure. Eventually, degradation will result in a malfunction of the gearing system, which will affect the availability of the whole system. Early detection allows proper scheduled shutdown to prevent catastrophic failure and consequently results in a safer operation and higher cost reduction. The current thesis deals with crack modelling from a condition monitoring perspective and focuses on the early detection of cracks propagating in gear teeth using vibration signal analysis. The research approach is based on crack modelling and gear mesh stiffness calculation, dynamic modelling and simulation, and methods for fault detection by means of dynamic response and vibration analysis. The time-varying gear mesh stiffness has been studied for cracked gears, and different crack progression scenarios have been applied. Different gear dynamic models have been used for simulating the dynamic response of the studied gear system. The gyroscopic effect of the gear disc has been studied by introducing a 12 DOF gear dynamic model. Signal processing techniques have been used in vibration signal analysis for detecting any impact of the crack. Moreover, the changes in the system dynamic response with varying mesh stiffness have been investigated. The possibility of detecting cracks has then been studied using the changes in the dynamic response of the system due to a stiffness reduction of the cracked tooth.