Please wait a minute...
J4  2010, Vol. 44 Issue (9): 1805-1810    DOI: 10.3785/j.issn.1008-973X.2010.09.028
能源与机械工程     
基于V-detector算法的滚动轴承故障诊断方法
杨先勇1,2,周晓军1,林勇1,张文斌1,沈路1
1.浙江大学 浙江省先进制造技术重点实验室,浙江 杭州 310027;2.中国舰船研究设计中心,湖北 武汉 430064
Fault diagnosis approach for rolling bearing based on V-detector algorithm
YANG Xian-yong1,2, ZHOU Xiao-jun1, LIN Yong1,ZHANG Wen-bin1, SHEN Lu1
1. Zhejiang Provincial Key Labroatory of Advanced Manufacturing Technology, Zhejiang University, Hangzhou 310027, China;
2. China Ship Development and Design Center, Wuhan 430064, China
 全文: PDF  HTML
摘要:

针对实值阴性选择(RNS)算法的检测器尺寸不能自适应变化的问题,提出基于变检测半径的RNS算法(Vdetector)的轴承故障诊断方法.将计算轴承振动信号局域波分解后各基本模式分量的关联维数作为特征向量,并根据故障模式将其划分为多个自体样本集,采用Vdetector算法训练多个检测器集,用其对轴承故障进行诊断.结果表明:自体半径过小则误诊率高,自体半径过大则检测器灵敏度低,这都将导致准确率减小;覆盖率越高,则准确率越高、计算花费越大,当覆盖率≥95%时,覆盖率对准确率的影响远小于其对计算花费的影响;相对于基于RNS的诊断方法,Vdetector算法具有同样高的准确率,且计算花费显著减小、稳定性更高,可有效地识别轴承故障.

Abstract:

Aimed at the detectors size of realvalued negative selection  (RNS)  algorithm is in capable  of adaptive adjustment, a fault diagnosis method for rolling bearing was proposed based on RNS with variablesized detectors (Vdetector). Firstly, correlation dimensions of intrinsic mode functions decomposed from bearing vibration signal were calculated as feature vectors. Secondly, the feature vectors were divided into different selfsamples sets according to fault modes, and then Vdetector algorithm was used to train and generate corresponding detector sets, which used for fault diagnosis. The experimental results show that the misdiagnosis rate will be high if self radius is too small, while the detectors sensitivity will be low conversely, either of which results in low diagnostic accuracy; higher estimated coverage corresponds to higher accuracy and time cost, and the estimated coverage has far less effects on the accuracy than on the time cost when it is higher than 95%. Compared with the diagnosis method based on RNS, the proposed method has far lower time cost and higher stability with the same high accuracy, and can identify the rolling bearing fault patterns effectively.

出版日期: 2010-09-01
:  TH 17  
通讯作者: 周晓军,男,教授,博导.     E-mail: cmeesky@163.com
作者简介: 杨先勇(1980-),男,湖北孝感人,博士生,从事舰船设计、振动噪声与信号处理、设备故障诊断研究.
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  

引用本文:

杨先勇, 周晓军, 林勇, 张文斌, 沈路. 基于V-detector算法的滚动轴承故障诊断方法[J]. J4, 2010, 44(9): 1805-1810.

YANG Xian-Yong, ZHOU Xiao-Jun, LIN Yong, ZHANG Wen-Bin, CHEN Lu. Fault diagnosis approach for rolling bearing based on V-detector algorithm. J4, 2010, 44(9): 1805-1810.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2010.09.028        http://www.zjujournals.com/eng/CN/Y2010/V44/I9/1805

[1] 张海勇. 一种新的非平稳信号分析方法:局域波分析[J]. 电子与信息学报, 2003, 25(10): 13271333.
ZHANG Haiyong. A new method for analyzing nonstation ary signallocal wave analysis [J]. Journal of Electronics and Information Technology, 2003, 25(10): 13271333.
[2] YANG Junyan, ZHANG Youyun, ZHU Yongsheng. Intelligent fault diagnosis of rolling element bearing based on SVMs and fractal dimension [J]. Mechanical Systems and Signal Processing, 2007, 21(5): 20122024.
[3] 张庆彬,吴惕华,刘波. 克隆选择单变量边缘分布算法[J]. 浙江大学学报:工学版, 2007, 41(10): 17151718.
ZHANG Qingbin, WU Tihua, LIU Bo. Hybrid univariate marginal distribution algorithm based on clonal selection principle [J]. Journal of Zhejiang University:Engineering Science, 2007, 41(10): 17151718.
[4] 董永贵,孙照焱,贾惠波. 时间序列中异常值检测的负向选择算法[J]. 机械工程学报, 2004, 40 (10): 3034.
DONG Yonggui, SUN Zhaoyan, JIA Huibo. Negative selected algorithm for anomaly detection in time series data [J]. Chinese Journal of Mechanical Engineering, 2004, 40 (10): 3034.
[5] FORREST S, PERELSON S, ALLEN L, et al. Selfnonself discrimination in a computer [C]∥Proceedings of the IEEE Symposium on Research in Security and Privacy. Los Alamitos: IEEE Computer Society,1994:202212.
[6] GONZLEZ F, DASGUPTA D, KOZMA R. Combining negative selection and classification techniques for anomaly detection[C]∥Congress on Evolutionary Computation. Honolulu: IEEE, 2002:705710.
[7] ZHOU J, DASGUPTA D. Reavalued negative selection algorithm with variablesized detectors [C]∥Genetic and Evolutionary Computation. Seattle : IEEE, 2004: 287298.
[8] ZHOU J, DASGUPTA D. Estimating the detector coverage in a negative selection algorithm [C]∥Genetic and Evolu tionary Computation. Washington:IEEE,2005:8897.
[9] HUANG N E, SHEN Z, STEVEN R L, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis [J]. Proc R Soc Lond A , 1998, 454: 903995.
[10] GRASSBERGER P, PROCACCIA I. Characterization of strange attractors [J]. Phys Rev Lett,1983, 50(5): 346349.
[11] KIM H S, EYKHOLT R, SALAS J D. Nonlinear dynamics, delay times, and embedding windows[J]. Physica D , 1999, 127(2): 4860.
[12] Case Western Reserve University.Bearing Data Center [EB/OL]. [20081006]. http://www.eecs.cwru.edu/laboratory/bearing

No related articles found!