[1]曾小军,黄宜坚.利用AR模型和支持向量机的调速阀故障识别[J].华侨大学学报(自然科学版),2011,32(1):13-17.[doi:10.11830/ISSN.1000-5013.2011.01.0013]
　ZENG Xiao-jun,HUANG Yi-jian.Fault Recognition of Speed Control Valve Based on AR Model and Support Vector Machine[J].Journal of Huaqiao University(Natural Science),2011,32(1):13-17.[doi:10.11830/ISSN.1000-5013.2011.01.0013]
点击复制

利用AR模型和支持向量机的调速阀故障识别()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第32卷
期数:
2011年第1期
页码:
13-17
栏目:
出版日期:
2011-01-20

文章信息/Info

Title:
Fault Recognition of Speed Control Valve Based on AR Model and Support Vector Machine
文章编号:
1000-5013(2011)01-0013-05
作者:
曾小军; 黄宜坚
华侨大学机电及自动化学院
Author(s):
ZENG Xiao-jun;  HUANG Yi-jian
College of Mechanical Engineering and Automation, Huaqiao University, Quanzhou 362021, China
关键词:
调速阀;  故障识别;  自回归模型;  支持向量机
Keywords:
speed control valve;  fault recognition;  autoregressive model;  support vector machine
分类号:
TH137.5
DOI:
10.11830/ISSN.1000-5013.2011.01.0013
文献标志码:
A
摘要:
提出一种基于时间序列的自回归(AR)模型和支持向量机故障识别方法.以液压调速阀的故障识别为例,利用采集到的调速阀体的振动信号建立AR模型; 然后,将AR模型自回归系数和残差方差组成的特征向量输入到支持向量机.最后,通过支持向量机完成对调速阀的正常和各种故障工况的分类识别.实验结果和分析表明,识别率不仅与核函数的选取有关系,而且与支持向量机参数的选取也有关系,以径向基RBF为核函数的识别率明显优于以多项式形式为核函数的识别率.
Abstract:
A fault recognition method based on time series autoregressive(AR) model and support vector machine has been put forward for fault recognition of hydraulic speed control value.Firstly,the AR model of vibration signal from speed control valve body is established; then the AR coefficients and error variance are regarded as the feature vectors which are used as an input of support vector machine; lastly,the normal state and all kinds of faults are classified by support vector machine.The result and analysis of the experiment indicate that the recognition rate is affected not only by the selection of the kernel function but also by the selection of parameters of support vector machine,the recognition rate is obviously more better when the kernel function is radial basis function(RBF)than that when the kernel function is polynomial kernel function.

参考文献/References:

[1] 蔡奇志, 黄宜坚. 三谱切片在调速阀故障诊断中的应用 [A]. 华侨大学学报（自然科学版）, 2009, (1):16-21.
[2] 蔡奇志, 黄宜坚. 三谱切片在调速阀故障诊断中的应用 [J]. 华侨大学学报(自然科学版), 2009(1):16-21.
[3] 李良敏, 屈梁生. 基于遗传编程和支持向量机的故障诊断模型 [A]. 西安交通大学学报, 2004, (3):239-242.
[4] 李良敏, 屈良生. 基于遗传编程和支持向量机的故障诊断模型 [J]. 西安交通大学学报, 2004(3):239-242.doi:10.3321/j.issn:0253-987X.2004.03.005.
[5] 童进, 吴昭同, 严拱标. 大型旋转机械升降速过程故障诊断HMM-AR方法研究 [A]. 振动与冲击, 1999, (2):79-80.
[6] 童进, 吴昭同, 严拱标. 大型旋转机械升降过程故障诊断HMM-AR方法研究 [J]. 振动与冲击, 1999(2):79-80.doi:10.3969/j.issn.1000-3835.1999.02.018.
[7] 刘天雄, 郑明刚, 陈兆能, 朱继梅. AR模型和分形几何在设备状态监测中的应用研究 [A]. 机械强度, 2001, (1):61-65.
[8] 刘天雄, 郑明刚, 陈兆能. AR模型和分形几何在设备状态监测中的应用研究 [J]. 机械强度, 2001(1):61-65.doi:10.3321/j.issn:1001-9669.2001.01.017.
[9] 杨叔子, 吴雅. 时间序列分析的工程应用 [M]. 武汉:华中理工大学出版社, 1991.
[10] 杨叔子, 吴雅. 时间序列分析的工程应用 [M]. 武汉:华中理工大学出版社, 1991.
[11] VAPNIK V. The natural statistical theory [M]. New York:springer-verlag, 1995.
[12] VAPNIK V. The natural statistical theory [M]. New York:springer-verlag, 1995.
[13] NELLO C, JOHN S T. An introduction to support vector machines and other kernel based learning methods [M]. Cambridge:Cambridge University Press, 2000.
[14] NELLO C, JOHN S T. An introduction to support vector machines and other kernel based learning methods [M]. Cambridge:Cambridge University Press, 2000.
[15] HSU C W, LIN C J. A comparison of methods for multiclass support vector machines [J]. IEEE Transactions on Neural Networks, 2002(3):415-425.
[16] HSU C W, LIN C J. A comparison of methods for multiclass support vector machines [J]. IEEE Transactions on Neural Networks, 2002, (3):415-425.

相似文献/References:

[1]蒋雨燕,黄宜坚.调速阀故障诊断的AR双谱定阶方法比较[J].华侨大学学报(自然科学版),2009,30(2):123.[doi:10.11830/ISSN.1000-5013.2009.02.0123]
　JIANG Yu-yan,HUANG Yi-jian.Comparison of Different Methods of Determining the Order of Autoregressive Bi-Spectrum[J].Journal of Huaqiao University(Natural Science),2009,30(1):123.[doi:10.11830/ISSN.1000-5013.2009.02.0123]

备注/Memo

备注/Memo:
福建省科技计划项目(2005H035)
更新日期/Last Update: 2014-03-23