運用希爾伯特-黃轉換及無因次單位瞬時頻率正規化技術之特徵擷取於非固定轉速之軸承故障診斷;Fault Diagnosis of Roller Bearing under Variable Rotation Speed via Hilbert-Huang Transform and Dimensionless Instantaneous Frequency Normalization Techniques

NCUIR > Engineering College > Graduate Institute of Mechanical Engineering > Electronic Thesis & Dissertation > Item 987654321/53798

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/53798

Title:	運用希爾伯特-黃轉換及無因次單位瞬時頻率正規化技術之特徵擷取於非固定轉速之軸承故障診斷;Fault Diagnosis of Roller Bearing under Variable Rotation Speed via Hilbert-Huang Transform and Dimensionless Instantaneous Frequency Normalization Techniques
Authors:	賴家祥;Lai,Chia-hsiang
Contributors:	機械工程研究所
Keywords:	支持向量機;無因次單位頻率正規化;希爾伯特-黃轉換;階次追蹤;故障診斷;Hilbert-Hung Transform;dimensionless frequency normalization;order tracking;Support Vector Machines;fault diagnosis
Date:	2012-07-04
Issue Date:	2012-09-11 18:16:31 (UTC+8)
Publisher:	國立中央大學
Abstract:	本研究主要針對轉動機械的軸承元件於變轉速下，軸承發生內圈損壞、滾子損壞、外圈損壞等情形之故障診斷方法探討。利用一般等時間取樣訊號量測及階次追蹤等角度訊號量測兩種方式擷取軸承振動訊號，前者的分析為提取振動訊號之包絡線，並透過希爾伯特-黃轉換結合無因次單位頻率正規化進行分析；後者的分析為提取振動訊號之包絡線，直接透過希爾伯特-黃轉換進行分析，於希爾伯特時頻譜與邊際譜探討軸承不同的損壞特徵。最後，提取邊際譜上軸承特徵頻率之幅值作為分類的特徵，並以支持向量機分類進行軸承的故障診斷，結果顯示有相當高的準確率。　　The main purpose of this paper is to characterize the faulted features of roller bearings, such as inner race, rolling element, and outer race with defect, under variable rotation speeds. There are two different ways to measure the vibration signals of bearings. One is the general vibration measurement with fixed sampling rate. The other one is based on the order tracking technique with identical angle increment. The envelope signals of the fixed sampling rate data are analyzed through Hilbert-Hung Transform and dimensionless frequency normalization. On the other hand, the envelope signals of the identical angle-increment data are analyzed by Hilbert-Hung Transform approach. The extracted features of the faulted bearings are discussed by observing the Hilbert time-frequency spectra as well as the marginal Hilbert spectra. With the magnitude of characteristic frequencies on the marginal Hilbert spectra as the extracted features, the classification results of supper vector machines demonstrate the high accuracy of bearing fault diagnosis.
Appears in Collections:	[Graduate Institute of Mechanical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	576	View/Open

社群 sharing

Loading...