Application of Singular Value Decomposition in Fault Diagnosis of Rolling Bearings
To address the difficulty in determining effective reconstruction components when singular value decomposition(SVD)is used for fault diagnosis of bearings,a new method was proposed based on the kurtosis criterion,so as to realize the fault diagnosis of rolling bearings.The method was applied for fault signals from the inner race,outer race,and rolling elements of the measured rolling bearings,and the envelope spectrum analysis of the reconstructed signals was performed.The experimental results show that the reconstructed signals can not only effectively extract the fault characteristic frequencies but also remove a large number of noise frequency components,providing a reference for fault diagnosis of rolling bearings.
万方数据
维普
