Bearing Fault Diagnosis Method Based on Genetic Algorithm and SOM Neural Network
Bearing is one of the core component of rotating machinery.The research on fault diagnosis of bearing is helpful to monitor the running state of rotating machinery.In view of the fact that the weak signal of the bearing fault is easily submerged in the vibration signals of other components,the characteristic extraction method is used to extract the time domain and frequency domain statistical characteristic parameters from the vibration signal of the rolling bearing under the four working conditions of normal,inner ring fault,outer ring fault and rolling element fault.And genetic algorithm is introduced to eliminate the coupling and collinearity between statistical characteristic in time domain and frequency domain,and 9 optimal time domain and frequen-cy domain characteristic parameters are extracted as input of SOM neural network.The research shows that under different fault types,the activated SOM neurons do not show obvious difference.According to the statistical rules of neurons activation in this pa-per,it show that SOM neural network has certain fault identification,and the adjusting the rules of neuron activation statistical rules can improve the diagnosis effect of SOM neural network.
NETL
NSTL
万方数据
维普
