Fault Diagnosis And Application Of Permanent Magnet Synchronous Motor Bearing Based On Sparse Representation Algorithm

Compared with traditional motors,Permanent Magnet Synchronous Motor(PMSM)mainly has higher efficiency,higher power factor,larger starting torque,better performance index,lower temperature rise and smaller armature response,Excellent performance such as strong anti-overload ability is more and more widely used in industrial production and civil products such as elevators and automobiles.Motor faults can generally be divided into mechanical system faults and electromagnetic system faults.For the mechanical system of the motor,due to the increasing complexity of the working environment and the increasingly refined work requirements,the sensitivity to faults has gradually increased,especially the high failure rate of the motor bearings,which makes the failure cost of the motor correspondingly increased.The fault diagnosis of the bearing is the focus of the mechanical fault diagnosis of the motor.Aiming at the shortcomings that traditional signal analysis methods are mainly based on existing mathematical models,have poor adaptive capabilities,and may not be able to accurately determine in complex and diverse working conditions,this paper takes the permanent magnet synchronous motor mechanical system as the research object,and analyzes the signal denoising effect and cycle Starting from the extraction of sexual pulse signals,the sparse representation algorithm is applied to the field of motor bearing fault diagnosis.Finally,the diagnosis effect of this method is verified through the analysis of simulation data and actual data.The thesis mainly includes the following work:(1)A fault diagnosis method for GMC motor bearings based on adaptive penalty parameters is proposed.This method uses GMC as the constraint item,which can avoid the inherent amplitude underestimation of the L1 constraint,and its penalty parameters can be automatically adjusted according to the error and sparsity,which can better Filter out the noise in the signal.Compared with the traditional signal-based fault diagnosis method and the basic dictionary learning K-SVD algorithm,the GMC algorithm with adaptive penalty parameters is introduced,and the effectiveness of the algorithm is verified through simulation and actual signals.(2)A fault diagnosis method of motor bearing combined with GMC and fast spectral correlation algorithm is proposed.In order to more clearly and distinctly separate the low-oscillation signal from the continuous-oscillation signal and extract the fault characteristics more effectively,a method combining GMC algorithm and fast spectral correlation algorithm is introduced.(3)The motor fault diagnosis platform is established based on the sparse representation algorithm.Based on the fault diagnosis of permanent magnet synchronous motor bearings,a motor fault diagnosis platform system with complete functions,simple analysis and reliable results was established based on Qt 5.10.1,and the application was verified on the actual motor test bench.