|
|
摘 要
随着科学技术的不断发展，机械设备向着高性能、高自动化、高效率和高可靠性的方向发展。齿轮箱因为具有传动比固定、传动转矩大、结构紧凑等优点，因此齿轮箱是用于改变转速和传递动力的最常用的传动部件，是机械设备的一个重要组成部分，也是最容易发生故障的一个部件。而在机械设备中，齿轮的使用频率很高，因此齿轮的故障诊断技术对机器的使用质量和使用寿命都起了非常重要的作用。本文从时域、频域，时频域和经验模式分解进行了齿轮故障诊断的方法研究。时域分析主要应用时域特征参数分析方法进行故障特征参数的提取，频域分析主要通过快速傅里叶变化，从频谱图上进行齿轮正常状态和故障状态振动信号的对比分析。时频域分析主要是通过一维三层离散小波变换，把原始信号细化为三层，每层又分为高频信号和低频信号。经验模式分解主要是在齿轮故障振动信号中的实际应用，对采集到齿轮四种状态下的振动信号通过EMD分解，提取了故障信号的特征信息，为识别故障类型提供了有效的分析手段。故障信息特征提取是齿轮故障诊断中最关键、最重要的问题之一，它直接关系到齿轮故障诊断的准确性和早期故障预报的可靠性。
关键词：齿轮；故障诊断；小波变换；经验模式分解
ABSTRACT
With the continuous development of science and technology,machinery and equipment are toward the direction of the development of the high-performance, high automation, high efficiency and high reliability . Gear box has the advantages of the transmission ratio fixed, the drive torque, compact gear box is the most commonly used to change the speed and transmission of power transmission components. It is an important component of the machinery and equipment, but is also a part of the most prone to failure. And of the machinery and equipment, the gear is a high frequency of use , so the gear fault diagnosis technology have played a very important role for the use of the quality and service life of the machine. The article is from time domain, frequency domain, time-frequency domain, the empirical mode decomposition to study the gear fault diagnosis. Time-domain analysis of the main application domain characteristic parameters of analytical methods for fault feature parameter extraction, frequency domain analysis of fast Fourier transform, the gear normal state and fault state comparative analysis of the vibration signal from the spectrogram. Time-frequency analysis is one-dimensional three-layer discrete wavelet transform, the original signal is subdivided into three layers, each layer is divided into high-frequency signals and low-frequency signals. Empirical mode decomposition is the practical application of fault vibration signals of gearbox fault signal feature information extracted by EMD, the collected vibration signals in gear four states provide an effective analytical tools to identify the type of fault. Fault information feature extraction is the most critical in gearbox fault diagnosis, one of the most important issues, is directly related to gearbox fault diagnosis accuracy and reliability of early fault prediction.
Keywords：Gear，Fault Diagnosis，Wavelet Transform，Empirical Mode Decomposition
目 录
第4.1节 基于EMD（经验模式分解）的振动信号特征提取31
参考文献…………………………………………………………………….45
