登录

  • 登录
  • 忘记密码?点击找回

注册

  • 获取手机验证码 60
  • 注册

找回密码

  • 获取手机验证码60
  • 找回
毕业论文网 > 毕业论文 > 理工学类 > 轮机工程 > 正文

压缩机气阀故障特征优选及同步优化识别方法毕业论文

 2021-11-07 20:54:01  

摘 要

压缩机作为船舶重要的辅助机械设备,为其相关设备的工作提供高压气源,所以必须确保压缩机的正常运行。其中,气阀作为压缩机的核心部件,在日常运行过程中极易发生磨损故障。因此,研究压缩机气阀磨损故障的故障特征,探讨最适合针对压缩机气阀进行故障分析的诊断方法具有重要的理论意义和工程应用价值。
本文以某船用压缩机为研究对象,对其气阀磨损故障特征分析及识别方法开展了一系列研究工作。全文主要研究工作和结论如下:
(1)压缩机气阀故障模拟试验。人为地破坏气阀结构用于模拟故障,并对气阀正常与故障状态下的压缩机缸盖表面振动进行了试验测试,进一步研究了振动信号时域及频域特征,为下一步故障特征的确定提供了原始数据。
(2)压缩机气阀故障特征提取。为了准确的提取气阀故障的判据,利用Wigner-Hough算法对压缩机缸盖表面的正常气阀以及故障气阀振动信号进行了时频分析,揭示了气阀的故障特征,并且提取了无量纲特征参数作为气阀故障诊断的依据。
(3)压缩机气阀故障识别。采用基于模式识别的KNN算法实现故障的识别,最终借助MATLAB平台结合KNN算法对故障模式进行识别分析。结果表明,利用Wigner-Hough算法进行故障特征提取和利用基于模式识别的KNN算法进行故障模式识别的故障诊断方法能够很好地识别压缩机气阀由于磨损发生的故障。通过研究为压缩机气阀故障的早期预警研究提供了依据。

关键词:压缩机;气阀;故障诊断;Wigner-Hough算法;KNN算法

Abstract

Compressors are important auxiliary equipment for ships, providing a high pressure source of air for the work of their associated equipment, so it is important to ensure the proper operation of the compressor. The valve, as the core component of the compressor, is prone to wear and tear during daily operation. Therefore, the study of the compressor valve wear fault characteristics, to explore the most suitable way for the compressor valve fault analysis of the diagnosis method has important theoretical significance and engineering application value.

In this paper, a series of research work is carried out on the characterization and identification of valve wear faults of a marine compressor as a research object. The main research work and conclusions of the full text are as follows.

(1) Compressor valve failure simulation test. The artificial destruction of the valve structure was used to simulate the fault, and the surface vibration of the compressor cylinder head under the normal and fault state of the valve was tested, and the time domain and frequency domain characteristics of the vibration signal were further studied to provide the raw data for the next step in the determination of fault characteristics.

(2) Compressor valve failure characteristics extraction. In order to accurately extract the evidence of valve failure, the Wigner-Hough algorithm was used to analyze the normal valve and the vibration signal of the faulty valve on the surface of the compressor cylinder head in time and frequency.

(3) Compressor valve fault identification. The KNN algorithm based on pattern recognition is used to realize fault identification, and the MATLAB platform combined with the KNN algorithm is used to identify and analyze fault patterns. The results show that the fault characterization method using the Wigner-Hough algorithm for fault characterization and the KNN algorithm for fault pattern recognition based on pattern recognition can well identify faults in compressor gas valves due to wear. The study provides a basis for early warning studies of compressor valve failures.

Key words: Air Compressor;;Valves;;Fault Diagonsis;Wigner-Hough;KNN

目录

第1章 绪论 1

1.1 课题研究目的和意义 1

1.2 压缩机气阀故障诊断技术 1

1.3 压缩机气阀故障诊断国内外研究现状 2

1.4 主要研究内容及技术路线 3

1.4.1 研究内容 3

1.4.2 技术路线 4

1.5 本章小结 4

第2章 振动信号分析与处理方法 5

2.1 线性时频分析方法 5

2.1.1 短时傅里叶变换 6

2.1.2 小波变换 7

2.2 二次型线性表示方法 8

2.3 本章小结 9

第3章 压缩机气阀故障特征提取 10

3.1 试验对象及平台 10

3.2 振动信号的特征参数提取 11

3.2.1 实测信号的时频分析 11

3.2.2 测试信号特征参数提取 17

3.3 本章小结 18

第4章 基于KNN算法的气阀故障诊断 20

4.1 KNN算法及其优缺点分析 20

4.1.1 KNN算法 20

4.1.2 KNN算法的优缺点 21

4.2 结合MATLAB平台和KNN算法的故障识别 22

4.3 本章总结 23

第5章 结论与展望 24

5.1 结论 24

5.2 展望 24

参考文献 26

致谢 28

绪论

课题研究目的和意义

压缩机作为重要的通用机器,普遍应用在船舶、家用电器、煤油化工等各生产范畴。气阀是往复式空气压缩机核心且极易发生故障的预制构件之一,在往复式压缩机的机械故障中,由于气阀故障引发的占到60%之多,并且由于气阀发生故障而导致的压缩机发生意外停机的占到37% [1]。由于往复式空气压缩机的内部结构极其复杂、并且引起激励的源头众多,这使得压缩机的振动信号分析难度较大,导致压缩机故障诊断的难度远远高于旋转机械故障诊断[2][3],因此压缩机故障诊断技术目前尚未完全成熟,仍然处于实验室研究阶段。而压缩机气阀故障研究主要由诊断信息获取、故障特征信息提取和模式识别这几个步骤组成,在这之中故障特征提取和状态识别是最主要的两个步骤[4],因此,在研究往复压缩机结构及故障机理的基础上,如何选取有效的信号提取方法和故障分析工具对往复压缩机进行故障诊断,对于压缩机的使用与维护具有非比寻常的意义[5]

压缩机气阀故障诊断技术

压缩机不同于旋转机械,压缩机的运转系统为非线性的不规则运转,无法通过常规的方法去针对压缩机气阀进行故障诊断;再加上压缩接内部结构非常复杂,所包含的闭环系统复杂,一处故障就有可能带来诸多异常。总而言之,对压缩机内部结构的复杂性分析如下:

您需要先支付 80元 才能查看全部内容!立即支付

企业微信

Copyright © 2010-2022 毕业论文网 站点地图