基于复数据的EMD在水导轴承故障诊断中的应用

发布时间：2018-05-30 14:22

  本文选题：复数据经验模态分解 + 模态混叠　； 参考：《水力发电学报》2017年02期

【摘要】：水轮机组是一个复杂的非线性动力系统,振动信号往往表现为非平稳性、非线性的特点,经验模态分解是一种新的时域分析方法,具有很好的适应性,较为适合处理非平稳性信号,但存在严重端点效应、模态混叠等问题。改进的集成经验模态分解一定程度上能够抑制模态混叠,但也会带来新的模态混叠、频谱丢失、运算量增大等问题。因此,本文将复数据经验模态分解运用到水轮机水导轴承的故障诊断中,添加白噪声作为虚部,从而构成复信号,通过白噪声在各个方向的投影来影响极值点的选取,同时利用噪声投影的影响再求包络质心的时候被消除的特性,从而抑制模态混叠。并通过水电站的实测信号验证该方法的有效性。
[Abstract]:The hydraulic turbine unit is a complex nonlinear dynamic system. The vibration signal is usually nonstationary and nonlinear. Empirical mode decomposition is a new time domain analysis method with good adaptability. It is more suitable to deal with non-stationary signals, but there are some problems such as serious endpoint effect, modal aliasing and so on. The improved empirical mode decomposition can restrain the mode aliasing to a certain extent, but it will also bring some new problems such as mode aliasing, spectrum loss and the increase of computation. Therefore, in this paper, the empirical mode decomposition of complex data is applied to the fault diagnosis of hydraulic turbine hydraulic bearing, and the white noise is added as the imaginary part to form the complex signal, and the selection of extreme points is influenced by the projection of white noise in all directions. At the same time, the noise projection is used to calculate the characteristic of the centroid of the envelope, which can suppress the mode aliasing. The validity of the method is verified by the measured signal of the hydropower station.
【作者单位】： 西安理工大学水利水电学院;甘肃省电力科学研究院;
【基金】：国家自然科学基金(51279161) 陕西省水利科技计划项目(2015slkj-04) 电网公司科技项目(522722150012)
【分类号】：TV738
，

本文编号：1955612