Research On Fault Diagnosis Of Reducer Based On Vibration Analysis

Reducer has been widely used in modern industrial society.It can make the machine working in the best condition.It can reduce the production cost and save energy sources for this society.So how to real-time monitor the health of the reducer especially in-depth study based on the analysis of vibration signal which is the most commonly used algorithm to judge reducer fault or not is of great significance.The main research content of this paper is as follows:First of all,introduce the types and basic structure of reducer,the vibration of reducer,the common failure types of reducer and the causes of the failure,the construction of the reducer experiment platform.Finally introduce the vibration signal analysis method of the reducer.Second,study the causes of HHT endpoint effect.Introduce two algorithms for inhibiting effect of the HHT endpoint: the SVR and BLCC.Finally,study and put forward a kind of HHT method based on SVR and BLCC.Contrast the EMD of mirror continuation method,combining time-frequency analysis and the results of the Hilbert instantaneous parameter indicate that decomposition results of the new algorithm is closer to real data results than the mirror continuation method.It proves that the new algorithm can effectively reduce the HHT endpoint effect phenomenon.Then,on the basic principle of blind source separation,one algorithm of blind source separation is studied-joint approximate diagonalisation of Eigen-matrices.Then this paper proposes an EMD-JADE algorithm.Finally,the simulation and the measured signal analysis results show that the EMD-JADE algorithm can solve the problem of JADE algorithm using the virtual noise to match real noise to noise reduction.Finally,when the reducer with abnormal noise but don't know where the specific fault is,combining with the EMD decomposition which endpoint effect is eliminated.Study a set of suitable system to match the worm reducer of RV40 type online fault detection,then displaying the detection system diagnosis interface.