基于高阶累积量的盲源分离的理论研究及其在旋转机械中的应用

发布时间：2018-05-13 20:58

本文选题：机械故障诊断 + 盲源分离　；参考：《武汉科技大学》2011年硕士论文

【摘要】：机械故障诊断中最重要的一步是提取故障特征,盲源分离在这一领域的应用具有光明的发展情景,盲源分离是一种多通道的信号处理方法,可以在不知道传输通道的情况下分离出源信号。对于机械故障诊断,最重的是提取出能够反映出机械运行信息的原始特征信号,应用盲源分离来进行故障诊断,可以提取出机械在运行中的各种独立的源信号,本论文探讨将如何将盲源分离应用到机械故障诊断中去,本文主要完成了如下这些工作。讲述了机械故障诊断的意义、盲源分离的简介、盲源分离理论的发展以及在故障诊断应用中的发展。系统的介绍了盲源分离理论所涉及到的各种基础理论知识以及各种盲源分离算法。讲述了机械振动信号在机械设备中的传输机理以及齿轮和轴承的各种常见故障以及频域表现特征。源数估计是盲源分离的一个很重要的前期工作,在讲述了一种各种信号估计方法后,提出了一种新的信号处理方法。基于二阶累积量和高阶累积量联合对角化的盲源分离算法是一种很稳定的盲源分离算法,本文讲述了这两种算法的原理以及分离性能,设计实验方案,再根据盲源分离模型模型,采用低通滤波器,对信号进行处理,再使用这两中算法分离真实的采集信号。
[Abstract]:The most important step in mechanical fault diagnosis is to extract fault features. The application of blind source separation in this field has a bright development scenario. Blind source separation is a multi-channel signal processing method. The source signal can be separated without knowing the transmission channel. For mechanical fault diagnosis, the most important thing is to extract the original characteristic signal which can reflect the information of mechanical operation, and to use blind source separation for fault diagnosis, which can extract all kinds of independent source signals of machinery in operation. This paper discusses how to apply blind source separation to mechanical fault diagnosis. This paper describes the significance of mechanical fault diagnosis, the introduction of blind source separation, the development of blind source separation theory and its application in fault diagnosis. The basic theories and algorithms of blind source separation are introduced systematically. The transmission mechanism of mechanical vibration signal in mechanical equipment, the common faults of gear and bearing and the characteristic of frequency domain are described. Source number estimation is an important work in blind source separation. After describing a variety of signal estimation methods, a new signal processing method is proposed. The blind source separation algorithm based on the diagonalization of second-order cumulant and high-order cumulant is a very stable blind source separation algorithm. According to the blind source separation model, the low pass filter is used to process the signal, and the two algorithms are used to separate the real collected signal.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TN911.7;TH165.3

【参考文献】