Abstract:

There has become a new method for analyzing nonlinear and nonstable signals in recent years. This new method could be generally called Hilbert-Huang Transform. Those nonlinear and nonstable signals could be decomposed by Hilbert-Huang Transform. Unfortunately, there would have got a problem with an endpoint effect in the process of decomposing those signals. The nearest simple extreme mirror extension method was proposed to solve this problem. The method was being established to analyze a nonlinear simulation signal. The results show that the method can suppress the endpoint effect very effectively and improve the decomposition accuracy of the nonlinear simulation signal. In addition, it is a humble, adaptable and generic method.

摘要：

近年来出现了一种能有效分析非线性、非稳定振动信号的新方法，即希尔伯特-黄变换法。该方法通过三次样条插值分解信号时会在其边界处出现端点效应问题。在深入研究端点效应问题基础上提出一种能有效解决该问题的新方法，即最近一次极值镜像延拓法。将该方法应用到非线性信号分析中，结果表明该方法不仅能有效抑制端点效应，而且还能提高非线性信号分解的准确性，同时具有简单易行、适应性强、通用性好等特点，对水泵机组运行时泵站厂房结构的振动监测具有一定的实用价值。

关键词: 振动与波, 信号分解, 包络曲线, 端点效应, 希尔伯特-黄变换, 经验模式分解, 固有模态函数