Research On Fault Diagnosis System Of Gear In Gearbox

Gears are common and vital components in gearboxes.They are widely used in modern industry and manufacturing industries.Related studies have shown that gearbox failures caused by gear failures have a great proportion.Therefore,it is significant to troubleshoot gears.The key to fault diagnosis of gears lies in the feature extraction of the fault signals.Through the extracted fault features,the fault type of the gear and the damage location can be discriminated.In this paper,the fault feature extraction algorithm of LMD mathematical morphology is deeply studied.In the process of gear fault feature extraction,mathematic morphology filtering,local mean value decomposition,sample entropy,and BP neural network are used to effectively extract faults from gears.It also effectively extract the characteristic information of gear fault status and realize the fault diagnosis of the gear.The main research work of this paper are as follows:(1)According to the characteristics of gear fault,a gear fault diagnosis method based on local mean decomposition and BP neural network is proposed,and the fault data of gear is collected through the fault simulation test platform.(2)It thoroughly study of mathematical morphology filtering,local mean decomposition,BP neural network and sample entropy,using morphological filter to reduce signal noise.The simulation signal and experimental data are used to verify the method.The results prove the effectiveness and application value of the proposed method.(3)Through the professional virtual instrument software LABVIEW,the intelligent diagnosis system of gear fault based on LMD is developed,and the intelligence of gear fault diagnosis is realized.