| In the field of automotive transmission technology,gear transmission equipment plays a crucial role.Gear failure is the main cause of failure in transmission equipment.If the early failure of the automobile gear is detected and repaired in time,some major losses caused by gear failure can be avoided.At the same time,during the operation of the car,the gears are running in a non-stable state.Therefore,this paper conducts research and analysis on automotive gear fault detection methods for automotive gears in non-smooth running conditions.Firstly,the meshing vibration mechanism of automobile gears is analyzed,and the spectral correlation analysis of fault signals is carried out by establishing the mechanical model of automobile gears.Then,aiming at the problem that the automobile gear failure signal is difficult to extract,in this paper,the improved wavelet threshold method is used to reduce the noise of the signal and suppress the amount of interference existing in the fault signal.Variational Mode Decomposition(VMD)is used to decompose the gear failure signal after noise reduction into a series of Intrinsic Mode Function.Furthermore,in view of the shortcomings that the number of modes and penalty factors in VMD are difficult to determine,the correlation coefficient analysis results between the original signal and the component signal were used to determine the number of decomposed IMFs and the optimization of the penalty factor using the sample entropy minimum principle.The obtained IMF is FFT transformed,and the FFT spectrum is used to extract gear features.Finally,the automotive gear signal studied in this paper has the characteristics of non-stationary and nonlinear.The Genetic Algorithm is used to optimize the Support Vector machine to achieve gear fault diagnosis.In order to verify the effectiveness of the proposed automobile gear fault detection method,in this paper,an experimental system for automotive gear fault detection is designed,and the proposed method is verified.The experimental results show that the accuracy of the fault classification and diagnosis of automobile gears in this paper reaches 95.41%,and the fault diagnosis of automotive gears with high precision and high efficiency is realized. |