Research on the Application of the Multi-sensor Data Fusion Technology in Fault Diagnosis of Rolling Bearings
On the basis of analyzing common faults of rolling bearings and vibrations mechanism, PCI Bus instrument was adopted to build a rolling bearing fault diagnosis system based on LabVIEW. Acceleration transducer and sound transducer were used to acquire multi-source vibration signals. The acquired signals are decomposed by wavelet packet transform (WPT) in order to extract features that represent the fault in some way. The signal energy acquired by calculating spectrographs certain band frequencys energy was adopted as features and formed the training and testing datasets. BP neutral network method was designed for feature-level fusion, which realizes intelligent fault diagnosis of rolling bearings and enhanced the diagnosis efficiency and accuracy.
data fusion wavelet packet transform fault diagnosis BP neutral network LabVIEW
CHEN Xia HUANG Zhichu Li WEixuan
School of Mechanical and Electronic Engineering, Wuhan University of Technology , Wuhan 430070, China
国际会议
北京
英文
2007-08-05（万方平台首次上网日期，不代表论文的发表时间）