Rolling Bearing Fault Diagnosis Method Based on EMD and Continuous Hidden Markov Model

In view of the difficulty in classifying and identifying the vibration signals of small damages in rolling bearings, a fault diagnosis method for rolling bearings based on EMD and continuous hidden Markov model is proposed. This method perform empirical mode decomposition on the framed preprocessed signal, selects the first five IMF components with high correlation with the original signal by calculating correlation coefficients, and then extract energy, kurtosis and margin from these filtered IMF components to form feature vectors and finally the hidden Markov model is trained to identify the test signal failure. Through experimental verification, after pre-setting the fault type of the bearing fault, the method proposed in this paper can effectively and accurately identify the fault type of the bearing to be tested.