Bearing Fault Diagnosis Based on Improved Wavelet Denoising and EMD Method

By combining improved wavelet threshold denoising with EMD decomposition,a new method is applied to bearing fault diagnosis.The original signal is denoised by the improved method of wavelet threshold denoising,then the denoised signal is decomposed into several IMFs by EMD adaptively.The IMFs are selected through energy-correlation coefficient method to reflect the fault characteristics,and fault frequency is gotten from envelope spectrum analysis.The experiment results show that the method can extract fault characteristic frequency effectively.