低S/N信号のバイスペクトル解析による転がり軸受の異常診断 : 転がり軸受の新しい異常診断法

概要

論文の詳細を見る
Since vibration excited by a flaw from assembled ball bearing is very weak, the flaw signal is usually buried in background noises. In order to detect such a weak signal, a new analytical method is proposed in this paper. The method makes use of a bispectrum analysis for flaw diagnosis in the assembled ball bearing. As previous works in this field, tests were performed for ball bearings which were not assembled in a machine. In those tests, vibration signals from flaws had relatively high S/N's. However when the flaw diagnosis of a ball bearing installed in a machine is intended, a new signal processing method must be developed to reduce the noise level. In this study, we use ball bearings with artificially dented flaws on the innerrace. The bearings are installed in a rotating machine. Then vibration of rotating machine is measured and flaw signals are extracted from background noises. Flaw signals are analysed by the bispectrum method. The newly developed bispectrum analysis proved to be a very effective method for vibration signals of low S/N obtained both from simulation and from ball bearing tests. This bispectrum analysis seems to be applicable to an on-line flaw diagnostic method for the ball bearing and other mechanical elements installed in a machine.
一般社団法人日本機械学会の論文
1991-07-25