Diagnosis of Vibration Variation of Bearing Based on Fuzzy Set Theory (Part I: Theory)

Abstract

Based on the fuzzy set theory, a method for diagnosing the variation of bearing vibration performance is proposed. Firstly, the original signal data of the bearing vibration acceleration are grouped to obtain the samples, and then the standard deviation, the average membership, the distance and the closeness degree of each sample are calculated as the sample indicators values. The sample indicators values are processed in the order of magnitude and form fuzzy similarity relation matrix by the linear mapping formula, then the fuzzy similarity relation matrix is transformed into fuzzy equivalence matrix by using the transfer closure method, and the equivalence coefficient in the fuzzy equivalence relation matrix is compared with the set threshold. Finally, the variation of bearing vibration performance is diagnosed.