首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于SVD的柴油机振动信号盲分离预处理
引用本文:高洪滨,欧阳光耀,朱石坚,邵利民.基于SVD的柴油机振动信号盲分离预处理[J].武汉理工大学学报(交通科学与工程版),2008,32(4).
作者姓名:高洪滨  欧阳光耀  朱石坚  邵利民
作者单位:海军工程大学船舶与动力学院,武汉,430033
基金项目:海军工程大学校科研和教改项目
摘    要:在利用独立分量分析对缸盖振动信号进行盲分离时,原始采样信号不满足独立分量分析的基本假设,而且缸盖振动信号信噪比较低,影响分离的速度和精度.采用小波阈值降噪的方法进行了降噪处理,提高了信噪比.采用奇异值分解的方法估计了缸盖振动的独立源数,并对缸盖振动信号进行了白化处理,为独立分量分析的应用建立了基础.

关 键 词:柴油机  盲分离  独立分量分析  奇异值分解

Preprocessing Vibration Signals on Cylinder Head of Diesel Engine Before Blind Separation Based on SVD
Gao Hongbin,Ouyang Guangyao,Zhu Shijian,Shao Limin.Preprocessing Vibration Signals on Cylinder Head of Diesel Engine Before Blind Separation Based on SVD[J].journal of wuhan university of technology(transportation science&engineering),2008,32(4).
Authors:Gao Hongbin  Ouyang Guangyao  Zhu Shijian  Shao Limin
Institution:Naval Architecture and Power Eng.College;Naval Univ.of Engineering;Wuhan 430033
Abstract:When vibration signals on cylinder head of diesel engine are separated blindly based on ICA(independent component analysis),sampled signals don't satisfy the assumptions of ICA.SNR(signal to noise ratio) of sampled signal is low,which causes low precision and slow convergence of ICA.Wavelet threshold denoising method is applied and SNR is enhanced.Then singular value decomposition is applied to estimate the independent sources of the signals and whiten the sampled signals,which is the base of ICA.
Keywords:diesel engine  blind separation  independent component analysis  singular value decomposition  
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号