Multivariate Information Theory for Condition Diagnosis of Rotating Machinery

概要

論文の詳細を見る
In order to raise the accuracy and reliability of condition diagnosis for plant machinery, the paper proposes a new method using multivariate information theory. The principle of condition diagnosis by the theory of the Multivariate Kullback-Leibler Information (MKI) is discussed, and the decision method of uniform criteria for condition diagnosis of plant machinery is established. The failure detection sensitivity of this method is higher than conventional method using only single type of signal. The efficiency of this method proposed in this paper is verified by actual example for detecting failures of a roller bearing using vibration signal, sound signal and AE signal.
一般社団法人日本機械学会の論文