Roller Bearing Fault Diagnosis Based On Envelope Spectrum Of GA And EMD

On the basis of studying methods of fault diagnosis to roller bearing and aiming at end effect of EMD( empirical mode decomposition),a analysis algorithms of GAand EMD envelope spectrum w as proposed. Using the features of optimization and parallel computing of genetic algorithm,determining the sampling signal decision variables and constraints,determining encoding method,obtaining the extreme points sampled signal. Using empirical mode decomposition to decompose the signal component of a number of IMF,solving Hilbert transform of component signal and building envelope spectrum of mode. Analysing the failure of tapered roller bearing,through the experiment proves the validity of the method and increasing the efficiency of the algorithm.