Bearing fault diagnosis for cement vertical mill based on entropy feature

Vertical mill is an important equipment existing in cement industry. In the normal operation process, defect may easily occurs at some components due to the severe working environment. Taking the bearing component as an example, this paper proposed a novel fault diagnosis method based on entropy feature to discriminate the ball fault, outer race fault and inner race fault. This algorithm consists of empirical mode decomposition and information entropy, and the entropy value calculated based on first ten intrinsic mode functions constitutes the feature vector. Meanwhile, several fault diagnosis space spanned by different feature vectors are exhibited. Finally a numerical simulation compared with classical EMD and improved EMD is carried out to verify the practical feasibility of present feature method, and further research direction is summarized.