石化行业旋转机械故障机理及诊断方法研究

发布时间：2018-05-19 08:53

本文选题：小波分析 + Mallat算法　；参考：《大连交通大学》2015年硕士论文

【摘要】：随着现代社会的发展与进步,人们越来越多地认识到机械的安全运行对于工业生产的重要性。正因如此,故障诊断方面的技术应用得到了大力推广,而传统的诊断技术则越来越难满足人们对于机械不断发展的安全标准。在过去的诊断方法中,人们最常使用快速傅里叶变换技术来对故障信号进行提取。这种方法为诊断行为提供了极大的便利,如果信号的特性是平稳的,则用该技术进行诊断时并不会发生异常。一旦信号具有时变特性,或是原信号夹杂着瞬时变化,那么这一类基于FFT技术的方法就会显现出弊端。在多分辨分析领域,小波变换可谓是一种新兴技术,它的核心是振荡的、衰减的基函数,其可以分析任意小的频率特征,因而被人们比作“数学显微镜”。本文首先从课题的背景出发,论述了故障诊断技术在国内外发展的现状。在对比经典的故障诊断方法和小波变化方法优劣之前,本文先对旋转机械常见的故障及其机理做了比较全面的总结,从而建立了机械故障与信号特征提取之间的联系。最后探讨了在诊断旋转机械的过程中利用小波分解技术的Mallat算法的问题,并在此基础上进行FFT变换,从而提高频谱特征的提取能力。仿真实验的结果说明,这种方法确实改善了故障诊断的精确性,得到了我们所期待的效果。
[Abstract]:With the development and progress of modern society, more and more people realize the importance of safe operation of machinery to industrial production. Because of this, the application of fault diagnosis technology has been popularized greatly, but the traditional diagnosis technology is more and more difficult to meet the safety standards of the continuous development of machinery. In the past, fast Fourier transform (FFT) is used to extract fault signals. This method provides great convenience for diagnosis behavior. If the signal characteristics are stable, there will be no anomalies when using this technique. Once the signal has the characteristic of time-varying or the original signal is mixed with the instantaneous variation, then this kind of method based on FFT technology will show some disadvantages. In the field of Multiresolution analysis, wavelet transform is a new technology. Its core is oscillating and decaying basis function, which can analyze any small frequency characteristic, so it is compared to "mathematical microscope". At first, this paper discusses the development of fault diagnosis technology at home and abroad from the background of the subject. Before comparing the advantages and disadvantages of the classical fault diagnosis method and the wavelet change method, the common faults and their mechanism of rotating machinery are summarized in this paper, and the relationship between the fault of machinery and the feature extraction of signal is established. Finally, the problem of Mallat algorithm based on wavelet decomposition in the diagnosis of rotating machinery is discussed, and on the basis of this, FFT transform is carried out to improve the ability of spectrum feature extraction. The simulation results show that this method can improve the accuracy of fault diagnosis and get the desired results.
【学位授予单位】：大连交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE65;TQ050.7

【相似文献】