可変運転条件における機械設備の異常診断 : 第1報, 遺伝的プログラミングとウェーブレット解析による回転機械の異常診断方法

概要

論文の詳細を見る
In the field of condition diagnosis of machinery, one of popular and practical methods is vibration analysis. Recently, dimensional and undimensional symptom parameters have been used for the condition diagnosis. However, these parameters can not be used to precisely detect failures of machinery in unsteady operating conditions, namely, in this case, the rotating speed and operating load of the machine are always changing. In order to overcome this difficulty, this paper proposes a new methods, by which the optimum symptom parameters can be automatically generated by Genetic Programming and the optimum frequency band can be decided to extract failure signal by Wavelet analysis for the precise diagnosis. The efficiency of the new methods is verified by applying them to the condition diagnosis of rolling bearing.
一般社団法人日本機械学会の論文
1999-12-25