基于信息熵的滚动轴承声发射信号故障诊断

发布时间：2018-01-17 16:34

本文关键词：基于信息熵的滚动轴承声发射信号故障诊断　出处：《兰州理工大学》2012年硕士论文　论文类型：学位论文

【摘要】：滚动轴承是机械设备中最重要的零部件之一，其运行状态的好坏直接关系到整个系统的工作效率和生产安全。因此，对滚动轴承进行状态监测和故障诊断有广泛的经济和社会效益。本文致力于利用滚动轴承运行过程中产生的声发射信号对其进行故障诊断。首先，本文对传统的滚动轴承故障诊断方法做了概述，并对其中典型的振动信号分析法做了阐述，并指出其局限性，即：在对低速滚动轴承进行故障诊断时，振动信号的提取以及去噪方面效果不很理想，因此，提出了用声发射信号进行轴承故障诊断的方法。其次，介绍了信息熵的基本概念，并详细描述了滚动轴承声发射信号在时域的奇异谱熵、频域的功率谱熵、时-频域的小波空间特征谱熵和小波能谱熵这四种信息熵特征。通过对故障轴承四种信息熵的计算，得出针对每一种轴承故障信息熵的熵带，并能初步做出故障的诊断。最后，基于信息融合的思想，引入了融合信息熵距的概念，并以此作为轴承故障诊断的方法和依据，由于熵距比较综合的考虑到了各种征兆域的故障特征，因而，故障判别的效果也比较理想，，并通过仿真实验验证理论的可行性。
[Abstract]:Rolling bearing is one of the most important parts in mechanical equipment. Its running condition is directly related to the working efficiency and production safety of the whole system. The condition monitoring and fault diagnosis of rolling bearings have a wide range of economic and social benefits. In this paper, the acoustic emission signals generated during the operation of rolling bearings are used for fault diagnosis. First of all, this paper summarizes the traditional fault diagnosis methods of rolling bearings, and describes the typical vibration signal analysis method, and points out its limitations, that is, in the fault diagnosis of low-speed rolling bearings. The effect of vibration signal extraction and de-noising is not satisfactory. Therefore, a method of bearing fault diagnosis using acoustic emission signal is proposed. Secondly, the basic concept of information entropy is introduced, and the singular spectrum entropy in time domain and power spectrum entropy in frequency domain of acoustic emission signal of rolling bearing are described in detail. The wavelet spatial characteristic spectrum entropy and wavelet energy spectrum entropy of time-frequency domain are four kinds of information entropy features. Through the calculation of four kinds of information entropy of fault bearing, the entropy band for each kind of bearing fault information entropy is obtained. And can initially make fault diagnosis. Finally, based on the idea of information fusion, the concept of fusion information entropy distance is introduced, which is used as the method and basis of bearing fault diagnosis. Therefore, the effect of fault discrimination is ideal, and the feasibility of the theory is verified by simulation experiments.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH165.3;TH133.33

【参考文献】