上海交通大学学报 ›› 2020, Vol. 54 ›› Issue (9): 953-960.doi: 10.16183/j.cnki.jsjtu.2020.154

• 制导、导航与控制专栏 • 上一篇    下一篇

基于机器视觉和盲源分离的机械故障检测

彭聪(), 刘彬, 周乾   

  1. 南京航空航天大学 自动化学院,南京 210016
  • 收稿日期:2020-05-26 出版日期:2020-09-28 发布日期:2020-10-10
  • 作者简介:彭聪(1988-),女, 江苏省泰州市人,教授,主要从事主动振动控制、智能控制和人工智能的计算机视觉研究.电话(Tel.):18551852606; E-mail: pengcong@nuaa.edu.cn.
  • 基金资助:
    国家自然科学基金(61703203);江苏省自然科学基金(BK20170812);中央高校基本科研业务费专项(56XAA19040)

Mechanical Fault Detection Based on Machine Vision and Blind Source Separation

PENG Cong(), LIU Bin, ZHOU Qian   

  1. School of Automation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
  • Received:2020-05-26 Online:2020-09-28 Published:2020-10-10

摘要:

为解决从复杂的信号环境下提取所需的信号,克服传统方式上信号获取与处理方法的不足和多源故障振动信号位置不确定等问题,针对机械转子的多源故障情况进行研究,提出一种基于机器视觉和盲源分离的旋转机械故障检测方法.首先介绍了基于机器视觉和盲源分离问题的数学原理,然后基于盲源信号分离方法和超定视觉盲源分离方法分析获取的高速视频,从而实现多源振动信号的分离与定位.最后,实验结果表明本文中提出的检测方法能够对旋转机械多源故障进行准确定位.该方法将机器视觉测量方法与盲源分离信号处理方法进行结合,实现了对多源故障有效分离识别.

关键词: 多源故障, 机器视觉, 盲源分离, 分离识别

Abstract:

In order to solve the difficulties in extracting the required signal from a complex signal environment, and overcome the shortcomings of traditional methods for signal acquisition and processing, and the uncertain location of multi-source fault vibration signals, the multi-source fault of mechanical rotor is studied, and a fault detection method for rotating machinery based on machine vision and blind source separation is proposed. First, the basic mathematical principles of machine vision and blind source separation problem are introduced. Next, the acquired high-speed video is analyzed based on the blind source signal separation method and the overdetermined visual blind source separation method to achieve the separation and positioning of multi-source vibration signals. The experimental results show that the detection method proposed in this paper can accurately locate the multi-source faults of rotating machinery. This method combines the measurement methods of machine vision with the signal processing method of blind source separation to achieve an effective separation and identification on the multi-source faults.

Key words: multi-source fault, machine vision, blind source seperation, separation and identification

中图分类号: