滚动轴承早期故障的特征提取与智能诊断_陈果.pdfVIP

　30 　2 　　　 Vol.30 No.2 　2009 　　2 ACTA AERONA UT ICA ET A ST RONA UT ICA SINICA Feb.　2009 :1000-6893(2009)02-0362-06 　　 ( , 　2 10016) Feature Extraction and Intelligent Diagnosis for Ball Bearing Early Faults Chen Guo (Colle e of Civil Aviation, Nanjin University of Aeronautics and A stronautics, Nanjin 　2100 16, China) 　:, 。 , , 。 , 。 , 。 :;;;; :TB 123　　　:A Abstract:In the study on ball bearin fault dia nosis based on wavelet transform, the parameter selection of wavelet transform and computation of fault features cannot be carried out automatically at present.Aimin at these problems, a new ball bearin fault feature auto-extractin method based on binary discrete w avelet t rans- form is proposed in this article, w hich can select automatically wavelet function parameters and extract the fault features.In addition, an intelli ent dia nosis model based on the neural netw ork with self-adaptive st ruc- ture is established to implement the intelli ent dia nosis of ball bearin faults.Finally, practical ball bearin experiment data is used to verify the new method put forward in this article, and the results fully validate its application. Key words:ball bearin s;binary discrete wavelet transform;neural netw orks;feature extraction;intelli ent dia nosis 　　, , 。, 、 , 。, [2-4], [1] , , , 。, ; , , M allat , , , ,