一种新时频研究方法局部特征尺度分解.docVIP

一种新时频研究方法局部特征尺度分解摘 要：在定义瞬时频率具有物理意义的内禀尺度分量（Intrinsic scale component，ISC）的基础上，提出了一种新的自适应时频分析方法――局部特征尺度分解(Local characteristic?scale decomposition，LCD)，该方法可以自适应地将一个复杂信号分解为若干个ISC分量之和.对LCD方法的基本理论进行了阐述，分别采用LCD方法和经验模态分解（Empirical mode decomposition，EMD）方法对仿真信号进行了分析，分析结果表明：两种方法都可以有效地对信号进行分解，但LCD方法在计算效率和抑制端点效应等方面要优于EMD 方法.此外，还将LCD方法应用于滚动轴承故障诊断，实验信号的分析结果进一步表明了该方法的有效性. ? 关键词：故障诊断；局部特征尺度分解；内禀尺度分量；滚动轴承? 中图分类号：TH115 文献标识码：A A New Time?frequency Analysis Method the ?Local Characteristic?scale Decomposition ?? YANG Yu??, ZENG Ming,CHENG Jun?sheng? (State key Laboratory of Advanced Design and Manufacture for Vehicle Body, Hunan Univ, Changsha, Hunan 410082, China) Abstract:Based on the definition of intrinsic scale component (ISC), a new self?adaptive time?frequency analysis method, local characteristic?scale decomposition (LCD), is proposed in this paper. By using LCD, a complicated signal can be decomposed into a number of ISC whose instantaneous frequencies have physical meaning. After expatiation on the basic theory of LCD, this paper analyzes the processed simulation signal by LCD and empirical mode decomposition (EMD). The analysis results have demonstrated that the validity of both the two decomposition methods. Moreover, LCD is superior to EMD in computational efficiency and restriction of end effects. In addition, LCD is also applied to fault diagnosis for roller bearing and the analysis results from the actual fault vibration signal have further proved the effectiveness of LCD. ? Key words:fault etection；local characteristic?scale decomposition; intrinsic scale component; roller bearing ?? 对非平稳信号的分析一直是相关学者关注的热点，由于时频分析方法能同时提供非平稳信号在时域和频域的局部化信息而在非平稳信号的分析中得到了广泛的应用.典型的时频分析方法有窗口傅里叶变换、Wigner分布、小波变换等，这些方法有一个共同的缺陷，那就是缺乏自适应性??［1-3］?.? 近年来最具代表性的自适应时频分析方法是经验模态分解??［4-5］?，该方法在定义瞬时频率具有物理意义的内禀模态函数分量的基础上，将复杂的多分量信号自适应地分解为若干个?IMF?分量之和.局部均值分解??［6］?是另外一种新的自适应时频分析方法，该方法将一个单分量的调幅调频信号看成是其本身的包络信号和一个纯调频信号的乘积，即?PF?（Product function）分量.该方法首先采用极值点获得局部均值函数和包络估