作者：周训强, 王见

作者单位：重庆大学测试中心, 重庆 400044

关键词：轴心轨迹; 图像预处理; 信号处理; 模式识别; 虚拟仪器

摘要：

转子轴心轨迹是判断转子运行状态和故障征兆的重要依据,为了提高转子轴心轨迹分析的效率和智能化程度,重点研究了基于数学形态学的轴心轨迹图像的预处理方法,基于神经网络的轴心轨迹图像识别方法,在此基础上设计了一种基于虚拟仪器的一体化轴心轨迹分析仪。介绍了仪器的构成与工作原理,完成了仪器的设计与开发,并进行了现场试验,该仪器可广泛应用于各种旋转机械转子的状态检测与故障预判。

Design of shaft centerline analyzer based on virtual instrument

ZHOU Xun-qiang, WANG Jian

Test Center, Chongqing University, Chongqing 400044, China

Abstract: The orbit of shaft centerline is an efficient evidence to determine the rotor's running state and default information. In order to improve the efficiency and intelligence of shaft centerline orbit analysis, the shaft centerline image pretreatment method based on mathematical morphology and the image recognition method based on neural network were studied in detail. On this basis, virtual instrument based on the integration of Shaft Centerline Analyzer was designed. The construction and principle of the instrumentation was introduced, and practical tests of the virtual instrument was also carried out. The results suggested the instrument was applicable for the fault diagnosis of rotating machines.

Keywords: orbit of shaft centerline; image preprocessing; signal processing; pattern recognition; virtual instrument

2010, 36(6): 45-48  收稿日期: 2009-10-22;收到修改稿日期: 2010-1-13

基金项目: 国家863高新技术发展计划项目(2008AA04208)

作者简介: 周训强(1983-),男,湖北荆门市人,硕士研究生,专业方向为虚拟仪器、智能测试与故障诊断。

参考文献

[1] 江志农,李艳妮. 旋转机械轴心轨迹特征提取技术研究[J]. 振动、测试与诊断,2007,27(2):98-102.
[2] 郎 锐. 数字图像处理学Visual C++实现[M]. 北京:北京希望电子出版社,2003:306.
[3] 阮秋琦. 数字图像处理学[M]. 北京:电子工业出版社,2001.
[4] 赵林度,盛昭瀚. 离散余弦变换在轴心轨迹自动识别中的应用[J]. 振动、测试与诊断,1999,19(1):35-39.
[5] 陈 豫. 轴心轨迹提纯与自动识别的研究[J]. 武汉理工大学学报:交通科学与工程版,2003,27(6):878-881.
[6] 杨淑莹. 图像模式识别[M]. 北京:清华大学出版社,2005:115-134.
[7] 张宏林. Visual C++数字图像模式识别技术及工程[M]. 北京:人民邮电出版社,2003:16-38.
[8] Wood J. Invariant pattern recognition: a review[J]. Pattern Recognition,1996,29(1):1-17.
[9] Li Y. Reforming the theory of invariant moments for pattern recognition[J]. Pattern Recognition,1992,25(7):723-730.
[10] Jiang X Y, Bunke H. Simple and fast computation of moments [J]. Pattern Recognition,1991,24(8):801-806.