Study Of Vertical Emulsifier Fault Diagnosis System Based On EMD-Sample Entropy And BP Neural Network

摘要

Aiming at the nonlinear characteristics of the fault vibration signal of the emulsifier, a feature extraction method based on Empirical Mode Decomposition (EMD) and Sample Entropy (SampEn) is proposed. In this method, the original vibration signal is decomposed into a finite number of intrinsic mode functions by EMD, then we select the Sample Entropy (SampEn) of the intrinsic mode function which contains the main failure information as the characteristic parameter, and BP neural network is used to diagnose the faults of the emulsifier in this study. Experiments verified that the BP neural network diagnosis method can get better fault diagnosis effect by using the EMD pretreatment to extract the Sample Entropy as the characteristic parameter.