Research On Gearbox Fault Diagnosis Based On Wavelet Transform And Blind Source Separation

Gearbox is a basic part of mechanical equipment and its operational status is critical to the safe operation of the whole equipment.Due to the complicated structure of the gear box and the harsh working environment,it is prone to failure during operation,especially the failure of gear and rolling bearing,which will affect the operation of the whole equipment.Therefore,it is necessary to use the appropriate signal processing technology to accurately extract the useful fault signal from the vibration signal of the gear box and diagnose the fault of the gear box.Aiming at the compound fault of gear and rolling bearing in gear box,this paper used improved wavelet threshold noise reduction,CEEMD and CICA to extract the fault signal of gear and rolling bearing from the vibration signal of simulation and actual measurement of gear box.The main work of this paper was as follows:Firstly,the vibration models of gear and rolling bearing and their common failure modes were introduced,and the common signal processing methods for gear box fault diagnosis were summarized.At the same time,DDS test bench were used to simulate the compound fault of gear and rolling bearing in gear box,and the vibration signal of single channel gear box were collected.In addition,the basic principles of ICA and CICA were mainly studied,and the simulation results have showed that CICA algorithm is better than Fast ICA algorithm.Secondly,the improved wavelet threshold de-noising method were studied,wavelet transform and the principle of wavelet threshold de-noising were introduced.On the basis of the traditional hard and soft threshold function,an improved threshold function were used.This threshold function not only has the advantages of the traditional soft-hard threshold function,but also avoids the defect of the soft-hard threshold function to some extent.In addition,the feasibility of the improved wavelet threshold function were verified by simulation and measured signals.Finally,the CEEMD method was used as the single-channel signal preprocessing method,this paper proposed the IWT-CEEMD-CICA algorithm for the single-channel gearbox composite fault signal problem.The selection criteria of effective IMF components were studied.The superiority of the proposed algorithm was verified by the simulated signal and the measured vibration signal,and the fault signal can be effectively extracted from the single-channel composite fault signal.