改进经验模态分解及其在齿轮故障诊断中的应用研究

发布时间：2018-01-30 01:59

本文关键词： 经验模态分解灰色理论 GM(1 1) 端点效应齿轮故障诊断希尔伯特一黄变换　出处：《太原理工大学》2011年硕士论文　论文类型：学位论文

【摘要】：随着信号处理技术的迅速发展,信号的时频分析方法已经成为分析处理非线性、非平稳信号的重要方法之一。它从信号的时域和频域两个不同角度来综合研究信号的特征,能够同时了解信号在时域和频域的特征信息,是信号处理领域的一个重大突破。其中经验模态分解(Empirical Mode Decomposition,简称EMD)是近几年发展起来的一种新的时频分析方法,由美籍华人N. E.Huang等人于1998年首次提出,它已经成为信号时频分析的重要途径之一。本文简要介绍了几种现代时频分析方法以及它们的优缺点,研究了经验模态分解理论的计算原理及其存在的不足之处,重点分析了经验模态分解在处理非线性、非平稳信号的“筛分”过程中产生端点效应的主要原因,在此基础上,提出一种新的抑制经验模态分解端点效应的方法,即将离散序列预测灰色理论GM(1,1)模型应用于在经验模态分解过程中,使得被分解的离散信号序列的端点值向外进行适当延拓,延拓后产生的新序列很好的反映了原信号的内部信息和发展趋势,使得在利用三次样条插值形成的上、下包络线时,端点效应被抑制而不污染到原始离散序列内部,从而可以保证分解出真实而有效的本征模函数。此外,齿轮是机械传动中重要传动部件之一,齿轮故障诊断方法对现代化工业发展有着举足轻重的推动作用。当齿轮在运转中出现有断齿、磨损和点蚀等故障时,就会引起齿轮强烈的啮合冲击振动,该振动信号中包含有周期性故障冲击振动成分,利用改进后的经验模态分解方法从齿轮啮合振动信号中提取齿轮故障特征,为齿轮故障诊断早期预测和诊断提供了一种更加可靠的方法。实验是获取数据和验证理论是否正确的重要途径。本文数据全部来自于齿轮故障的物理模拟实验,该实验分别采集了齿轮在正常状态和故障状态下的齿轮啮合振动信号,然后借助于软件MATLAB对实验所得数据进行编程处理。在试验数据基础上,分别使用改进和未改进的经验模态分解来处理同样的实验数据,由得到的本征模函数对比可以看出前者能有效地抑制经验模态分解的端点效应。然后再将故障齿轮振动信号和正常齿轮振动信号分别利用改进经验模态分解进行处理,得到他们各自的本征模函数,并对其进行希尔伯变换,进一步获得齿轮在故障状态和正常状态下振动信号的希尔伯特谱及边际谱,从对比中可以明显发现齿轮故障的存在,同时也进一步证明了灰色GM(1,1)模型能够有效地抑制经验模态分解的端点效应。
[Abstract]:With the rapid development of signal processing technology, signal time-frequency analysis method has become nonlinear. Non-stationary signal is one of the most important methods. It synthesizes the characteristics of signal from two different angles of time domain and frequency domain, and can understand the characteristic information of signal in time domain and frequency domain at the same time. It is an important breakthrough in the field of signal processing, in which empirical mode decomposition (EMD) is an empirical Mode Decomposition. EMD, a new time-frequency analysis method developed in recent years, was first proposed by N.E.Huang et al., a Chinese American in 1998. It has become one of the important ways of signal time-frequency analysis. In this paper, several modern time-frequency analysis methods and their advantages and disadvantages are briefly introduced, and the computational principles and shortcomings of empirical mode decomposition theory are studied. The main causes of endpoint effect in the process of dealing with nonlinear and non-stationary signals by empirical mode decomposition (EMD) are analyzed. A new method to suppress the end-point effect of empirical mode decomposition (EMD) is proposed, in which the discrete sequence prediction grey theory (GM-1) model is applied in the process of EMD. The end value of the decomposed discrete signal sequence is extended outwardly, and the new sequence after extension reflects the internal information and the development trend of the original signal well, which makes it possible to make use of cubic spline interpolation to form the new sequence. In the lower envelope, the endpoint effect is suppressed without contamination into the original discrete sequence, thus the real and effective eigenmode function can be decomposed. In addition, the gear is one of the important transmission components in mechanical transmission. The method of gear fault diagnosis plays an important role in the development of modern industry. When the gear has broken teeth, wear and pitting in operation, it will cause the gear strong meshing shock vibration. The vibration signal contains periodic fault shock vibration component. The improved empirical mode decomposition method is used to extract gear fault characteristics from gear meshing vibration signal. It provides a more reliable method for early prediction and diagnosis of gear fault diagnosis. The experiment is an important way to obtain data and verify whether the theory is correct. All the data in this paper come from the physical simulation experiment of gear fault. In this experiment, the gear meshing vibration signals in normal state and fault state were collected, and the experimental data were processed by software MATLAB. Based on the experimental data, the improved and unimproved empirical mode decomposition are used to deal with the same experimental data. The comparison of eigenmode function shows that the former can effectively suppress the endpoint effect of empirical mode decomposition. Then, the vibration signal of fault gear and the vibration signal of normal gear can be decomposed by improved empirical mode decomposition, respectively. Handle it. Their Eigenmode functions are obtained, and Hilbert transform is used to obtain the Hilbert spectrum and marginal spectrum of gear vibration signal in fault state and normal state. From the comparison, it can be found that the gear fault obviously exists, and it is further proved that the grey GM1 / 1) model can effectively suppress the end-point effect of empirical mode decomposition (EMD).
【学位授予单位】：太原理工大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3

【参考文献】