摘要：这次的毕业设计的目的是利用，轴承故障振动信号的非平稳特性，结合小波变换的好处提出将压电式振动加速度传感器的轴承的信号采集dblo小波滤波器构筑，过滤构筑信号小波由包络信号变成，最后包络谱细分化为振动幅度谱图，细分化后的振幅谱图判别轴承故障的小波分析方法一包络模拟和实验分析轴承的故障，验证理论分析的正确程度。81075

    小波变换可以局部分析信号。谁着频率变化。它能够对时间（空间）频率本地化的分析，可以达到高频出时间细分，低频出频率细分。本文的第一部分是重点介绍小波变换的概念和方法，在引用MATLAB编程实验后，经过实例分析，小波变换可以极大的减小或去除所提取得不同特征之间的相关性。

信号分析和处理的现有方法，对于非固定信号和奇异信号的高带宽分段是不理想的。 本文的第二块主要介绍，用一种全新的方法——复小波分析和处理故障信号。具有严格的盒形光谱特征和简单分析表达式的小波谱 相关的编程实验，

毕业论文关键词： 近似熵；小波；dblo小波；滚动轴承信号的分析；

Vibration signal analysis of rolling bearing fault in heavy haul freight train

Abstract: the purpose of this graduation design is the use of non-stationary of bearing fault vibration signals with wavelet transform, put forward to build good signal acquisition dblo wavelet filter bearing piezoelectric vibration acceleration sensor, signal into wavelet filter constructed by envelope signal, finally the envelope spectrum differentiation for vibration amplitude spectrum。 Fault analysis method of envelope simulation and experimental analysis of judging bearing wavelet spectrum of bearing amplitude differentiation after the correct degree to verify the theoretical analysis。

     Wavelet transform is a Fourier analysis of idea of localization, inheritance and development of new methods, and overcome the different frequency change window shortcomings。 Wavelet transform of time frequency localization analysis (space), can achieve the high frequency time segments, the frequency of low frequency segments。 The first part of this paper is focused on the concept and method of wavelet transform in reference, MATLAB programming experiment, through case analysis, wavelet transform can greatly reduce or remove the correlation between the features extracted。

    The existing methods of signal analysis and processing, for high bandwidth and non fixed signal singularity signal segment is not ideal。 In order to fix this problem, we must study the new method of signal analysis and processing, fault signal analysis。 This paper mainly introduces the second blocks, with a whole new method, wavelet analysis and processing fault signal。 Including the introduction of harmonic wavelet and complex wavelet dblo principle, through the strict mathematical formulas are introduced, the harmonic wavelet algorithm, the final reference example and practical application, with box shaped spectral characteristics of strict and simple analytical expressions of small spectrum related programming experiment, DBL O wavelet has the same strong ability to extract fault signals

    Keywords: approximate entropy; wavelet transform; dblo wavelet; analysis of rolling bearing signals;

目    录

摘要  1

ABSTRACT  2

绪论    4

1小波分析的理论   6

1。1 小波变换的兴起   6

1。2 小波的定义   6

1。3小波包的分析       14

1。4Mallat算法       14

1。5包络分析      21

1。5。1常见的包络分析法  49