APPLICATION OF THE FULL VECTOR SPECTRUM BASED ON EMD IN FAULT DIAGNOSIS OF BEARINGS

Abstract

Aiming at the deficiency of the analysis of single channel signal EMD and dual channel signal traditional full vector spectrum analysis,this paper presents the full vector spectrum extraction based on EMD.The method of vibration signals collected from the same cross section of the two perpendicular on the sensor,then use EMD on each signal respectively,so that the two signals are decomposed into several IMF components.By the fusion of the IMF component,we can get the full vector spectrum based on obtained EMD.The simulation results show that,the method of obtaining fault characteristics is more comprehensive,accurate.The analysis results of bearing fault vibration signals of TRT generating units in a power rating of3000 KW,further shows the feasibility and effectiveness of the method.

Keywords

• Empirical mode decomposition;Full vector spectrum;TRT generator;Fault diagnosis;Homologous information fusion;Sliding bearing