在定义一种瞬时频率具有物理意义的单分量信号—内禀尺度分量(简称ISC)的基础上,提出一种新的信号自适应时频分析方法—局部特征尺度分解方法(简称LCD)。LCD方法可以自适应地将任意一个复杂信号分解为若干个瞬时频率具有物理意义的ISC分量之和。对LCD方法的基本理论进行研究,分别采用LCD方法和经验模态分解(简称EMD)方法对仿真信号进行分析,对比结果表明:LCD方法的有效性及在端点效应、计算时间等方面都优于EMD方法,并且把LCD方法应用于齿轮的实验振动信号分析,LCD方法可以有效地应用于齿轮故障诊断。
Based on the definition of a component signal which has physical significance--Intrinsic scale component (ISC), a new kind of self-adaptive time-frequency analysis method--Local characteristic scale decomposition (LCD) is proposed. By using LCD method, each complicated signal can be decomposed into a number of ISC whose instantaneous frequencies own physical meaning. The theory of LCD is analyzed firstly, then the LCD method is compared with the empirical mode decomposition (EMD) method by analyzing a simulation signal. The analysis results show that the LCD method is effective and is superior to the EMD method in restraining end effect, decomposition time and so on. The LCD method is also used to analyze the gear vibration signal, and the LCD method can be applied to the gear fault diagnosis effectively.