基于模态准小波的密频模态参数识别-太赫兹科学与电子信息学报.pdfVIP

第 9 卷 第 2 期 信 息 与 电 子 工 程 Vo1．9，No．2 2011 年 4 月 INFORMATION AND ELECTRONIC ENGINEERING Apr.，2011 文章编号： 1672-2892(2011)02-0195-07 基于模态准小波的密频模态参数识别 叶庆卫，赵挺凯，周 宇，王晓东 (宁波大学 信息科学与工程学院，浙江 宁波 315211) 摘 要 ：基于 Morlet 小波连续变换的模态参数识别算法在故障识别和多模态参数提取中起到 了重要作用，然而在密频情况下，Morlet 小波连续变换尺度图上模态极值线相互混叠严重，难以有 效分离模态。本文提出一种新型的模态准小波函数，运用连续小波变换理论，推导和证明了模态 准小波的连续变换可以有效分离密频模态，并给出了固有频率和阻尼系数的识别公式与整体识别 算法。文章针对新型的模态准小波进行了大量的仿真实验，证实了该小波在密频模态分离与参数 识别方面的优越性，为密频模态参数提取提供新的思路与手段。 关键词 ：模态准小波；模态参数识别；密频模态 中图分类号 ：TN911.7 ；TP391.4 文献标识码 ：A Parameter identification for concentrated modal based on modal quasi-wavelet YE Qing-wei，ZHAO Ting-kai，ZHOU Yu，WANG Xiao-dong (Information Science and Engineering College，Ningbo University，Ningbo Zhejiang 315211，China) Abstract： Modal parameters identification algorithm based on continuous wavelet transformation of the Morlet wavelet is an important algorithm in fault diagnosis and multi-modal parameters identification area. However, on the time-scale image of wavelet continuous transformation of the Morlet wavelet, there will appear aliasing phenomenon between two modals while vibration modals are concentrated, and it is hard to separate these concentrated modals. A new wavelet function called modal quasi-wavelet is proposed in this paper. The continuous transformation of modal quasi-wavelet can split two concentrated modals effectively. And the new algorithm of modal parameters identification is put forward. Numerous experiments of the new algorithm indicate that the modal quasi-wavelet shows better performance than M