Study On Gearbox Fault Diagnosis Based On VMD Algorithm

Gearbox plays an important role in mechanical transmission equipment and is widely used in all walks of life.Often a gear failure will cause the entire equipment to stop.In order to ensure safe and reliable operation of gearbox,it is of great significance to carry out condition monitoring and fault diagnosis.In this paper,the variational mode decomposition(VMD)is introduced into the fault diagnosis of gearbox,the detection and fault feature extraction of non-stationary vibration signal of gearbox based on VMD are studied in detail,and the method of combining VMD and constrained independent component analysis(CICA)are proposed to successfully separate the fault of bearing and gear,and find the fault location.The variational mode decomposition,particle swarm optimization(PSO)and support vector machine(SVM)method are used to identify the weak fault of high speed rail wheelset bearings.Firstly,the paper introduces the vibration mechanism of rolling bearing and gear,the common failure forms of rolling bearing and gear,and summarizes the common treatment methods of gearbox fault diagnosis.The VMD algorithm is mainly discussed,and the parameter selection of VMD is optimized by using particle swarm optimization algorithm.Secondly,a gearbox hybrid fault diagnosis algorithm based on VMD-CICA is studied.By particle swarm optimization VMD decomposition to convert single channel signal for multiple channel signals,based on kurtosis and correlation coefficient of standard to choose the intrinsic mode function(IMF)component reconstruction signal effectively,with the CICA algorithm to reconstruct signal separation,from single channel effectively isolated from the gearbox fault signal mixed bearing failure and gear failure,and to determine the position of failure.Finally,the VMD-PSO-SVM algorithm is applied to the pattern recognition of weak faults of high-speed railway wheelsets.The proposed method is compared with others algorithm,which proves that the method has higher recognition accuracy and can effectively diagnose the weak faults of high-speed rail wheelset bearings.To sum up,the VMD decomposition and CICA algorithm are combined in this paper to make up for the single signal analysis method in gearbox hybrid fault diagnosis,which has good practicability.The VMD-PSO-SVM algorithm is applied to the pattern recognition of weak bearing fault.