基于奇异值分解的连续小波消噪方法.pdfVIP

第 38 卷 　第 9 期 西　安　交　通　大　学　学　报 Vol. 38 　№9 2004 年 9 月 　 Sep. 2004 JOURNAL OF XI′AN J IAOTONG UNIVERSITY 基于奇异值分解的连续小波消噪方法 梁　霖 , 徐光华 , 侯成刚 (西安交通大学机械工程学院 , 710049 , 西安) 摘要 : 针对小波软阈值消噪的缺点 ,提出了一种基于奇异值分解的连续小波消噪方法. 通过对小波变换的系 数矩阵进行奇异值分解 ,将其中的信号特征成分和噪声分解到不同的正交子空间中 ,在子空间中选取集成信 号特征成分的奇异值矢量进行重构 ,从而提取出淹没在噪声中的信号成分. 通过仿真数据的对比分析和工程 测试信号的应用 ,表明该方法适用于冲击成分信号的提取 ,与软阈值消噪法相比 ,它提取出的信号特征成分 更完整 ,信噪比更高. 关键词 : 连续小波变换 ;奇异值分解 ;消噪 ( ) 中图分类号: TH13313 　文献标识码 : 　文章编号 : 0253 - 987X 2004 09 - 0904 - 05 Continuous Wavelet Transform Denoising Method Based on Singular Value Decomposition Liang Lin , Xu Guanghua , Hou Chenggang ( School of Mechanical Engineering , Xi′an Jiaotong University , Xi ′an 710049 , China) Abstract : Aimed at the deficiency of soft2thresholding denoising , a new continuous wavelet transform (CWT) denoising method based on singular value decomposition ( SVD) was proposed. The matrix of CWT , which contains signal and noise , was obtained via morlet wavelet , and then , instead of the soft2thresholding method , the SVD was applied to de2 compose the signal features and noise into different orthogonal sub2spaces. With the reconstruction of the singular vectors in sub2space , the signal features were extracted effectively. Experimental results and industrial measurements on fault di2 agnosis show that the approach , compared with the soft2thresholding denoising , is an efficient tool to extract impact com2 ponent from mechanical signals. Keywords : continuous wavelet tran