基于信息熵熵距的涡旋压缩机故障诊断

发布时间：2018-05-28 20:14

本文选题：涡旋压缩机 + 阈值　；参考：《兰州理工大学》2016年硕士论文

【摘要】：近年来随着涡旋压缩机应用领域的不断扩展,不同的工作环境,引发涡旋压缩机故障的因素变得愈加复杂,涡旋压缩机的故障诊断也变得更有必要。涡旋压缩机的振动测试实验平台的搭建和对其振动噪声的分析显得尤为重要。但由于涡旋压缩机工作环境较为复杂,测试系统较难搭建等因素的影响,难以满足实时监测、诊断的要求。此外涡旋压缩机的应用时间较短,故障数据库不健全,且振动源较多,壳体信号为非平稳性和非线性,因此故障特征难以准确地区分且故障诊断过程较为复杂。本文以涡旋压缩机振动测试实验平台为基础,将信息熵和熵距相结合实现对非平稳信号进行故障判别。具体实施过程如下:(1)从振动信号分析的思路出发,结合信息论中熵和欧氏距离理论,提出一种基于信息熵和熵距的故障诊断方法。(2)搭建改进已有实验平台,安装传感器,更换硬件,调试软件进行试验。采集信号,建立机体正常运行的特征标准和四种典型故障的特征标准。(3)使用实验平台分别模拟四种故障,利用Matlab信号处理工具箱的强大功能,对采集到的故障信号进行基于阈值的小波包去噪,分解和重构,提取信息熵作为故障特征。(4)将提取的故障特征,与典型故障对比,使用熵距计算方法得到熵距,综合考虑信息熵和熵距对测试信号进行故障诊断。实验结果表明:该方法对转子不平衡和轴承故障的故障诊断具有很高的准确度,对其余故障也能很好的区分。信息熵能反映故障类型及故障严重程度,而熵距曲线图则进一步提高诊断的准确性。同时其也能够表示复合故障的部分特征,因而给复合故障诊断提供一种新的思路,也为涡旋压缩机的结构设计、制造加工和安装检测提供一定的帮助。
[Abstract]:In recent years, with the continuous expansion of the application field of scroll compressor, the factors causing the scroll compressor fault become more and more complex in different working environment, and the fault diagnosis of scroll compressor becomes more necessary. It is very important to build the experimental platform and analyze the vibration and noise of scroll compressor. However, due to the complex working environment of the scroll compressor and the difficulty of setting up the test system, it is difficult to meet the requirements of real-time monitoring and diagnosis. In addition, the application time of scroll compressor is short, the fault database is not perfect, and there are many vibration sources, the shell signal is non-stationary and nonlinear, so it is difficult to distinguish the fault features accurately and the fault diagnosis process is more complicated. In this paper, based on the vibration test platform of scroll compressor, the information entropy and entropy distance are combined to realize the fault identification of non-stationary signals. The concrete implementation process is as follows: (1) starting from the idea of vibration signal analysis and combining the theory of entropy and Euclidean distance in information theory, a fault diagnosis method based on information entropy and entropy distance. Replace hardware and debug software for test. Collect the signal, establish the characteristic standard of the normal operation of the body and the characteristic standard of four kinds of typical faults. Use the experiment platform to simulate the four kinds of faults, and utilize the powerful function of the Matlab signal processing toolbox. The acquired fault signal is de-noised, decomposed and reconstructed based on threshold wavelet packet, and the information entropy is extracted as the fault feature. The extracted fault feature is compared with the typical fault, and the entropy distance is obtained by using entropy distance calculation method. The information entropy and entropy distance are considered comprehensively to diagnose the fault of test signal. The experimental results show that the method has a high accuracy for the fault diagnosis of rotor unbalance and bearing fault, and can distinguish the other faults well. Information entropy can reflect fault type and fault severity, while entropy distance curve can further improve the accuracy of diagnosis. At the same time, it can also express some characteristics of complex faults, which provides a new idea for complex fault diagnosis, and also provides some help for the structure design, manufacturing and processing of scroll compressor and installation and detection.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TH45

【参考文献】