The current work aims to develop a classification method devoted to gear defect diagnosis. In this paper, the proposed classification method is based on the Neural Networks, Discrete Wavelet Transform and Principal Component Analysis. A gearbox system with six degrees of freedom (DOF) is simulated in MATLAB and Simulink. Defects are introduced in the model by the meshing stiffness function which is computed by considering in series the bending, shear, axial compressive, fillet foundation and Hertzian stiffness. The signals dataset is collected by changing system or defect parameters. In addition, an experimental data is tested with the proposed method. Signal features are extracted using the Discrete Wavelet Transform with the Principal Component Analysis. This method allows us to classify the extracted features into two classes, healthy and faulty, with a good rate of correct classification. Both simulated and experimental data are tested with the proposed method.

In the current work, the nonlinear dynamic behavior of an eight degrees of freedom gear system is investigated. Tooth crack is introduced into the model. The main sources of excitation are the time varying mesh stiffness (TVMS), time varying mesh damping, backlash and friction inter teeth. By using the potential energy method applied into the cantilevered beam, the TVMS is calculated. The backlash is considered as a dead zone. The effect of backlash, friction and tooth crack on the vibration frequencies is highlighted. The crack detection is based on the time-frequency analysis.

Gear is one of the most omnipresent components in the mechanical and industrial fields. Some defects can occur causing a change in the vibration signature. In the dynamic of gears system, several sources of excitation are considered. The principal objective of the present work is to study the use of the Reassigned Smoothed Pseudo Wigner Ville Distribution (RSPWVD) in the gear vibration monitoring. The used signals are obtained by the simulation of an eight degrees of freedom gearbox system for the healthy and cracked case. The considered model takes into account the presence of friction force inter teeth in contact, backlash and time varying mesh stiffness. Also, the presence of noise is highlighted. A comparison with the Smoothed Pseudo Wigner Ville Distribution (SPWVD) is done.