计算机应用 ›› 2012, Vol. 32 ›› Issue (09): 2520-2522.DOI: 10.3724/SP.J.1087.2012.02520

• 人工智能 • 上一篇    下一篇

基于核主元分析的神经网络控制图模式识别

胡胜1*,李太福2,魏正元1,颜克胜1   

  1. 1.重庆理工大学 数学与统计学院,重庆 400054;
    2.重庆科技学院 电气与信息工程学院,重庆401331
  • 收稿日期:2012-03-14 修回日期:2012-05-10 发布日期:2012-09-01 出版日期:2012-09-01
  • 通讯作者: 胡胜
  • 作者简介:胡胜(1988-),男,湖北黄冈人,硕士研究生,主要研究方向:应用统计、质量管理; 李太福(1971-),男,四川资阳人,教授,博士,主要研究方向:智能控制; 魏正元(1975-),男,湖北襄阳人,副教授,博士,主要研究方向:应用概率、随机过程统计; 颜克胜(1988-),男,湖北荆州人,硕士研究生,主要研究方向:应用统计、特征选择。
  • 基金资助:

    国家自然科学基金资助项目(51075418);重庆市自然科学基金资助项目(CSTC2010BB2285)

Neural network for control chart pattern recognition based on kernel principle component analysis

HU Sheng1*,LI Tai-fu2,WEI Zheng-yuan1,YAN Ke-sheng1   

  1. 1.School of Mathematics and Statistics,Chongqing University of Technology,Chongqing 400054,China;
    2.School of Electrical and Information Engineering,Chongqing University of Science and Technology,Chongqing 401331,China
  • Received:2012-03-14 Revised:2012-05-10 Online:2012-09-01 Published:2012-09-01

摘要: 针对异常特征之间存在较大的相似性而带来的网络结构复杂和识别精度不高的问题,提出一种基于核主元分析的神经网络控制图模式识别方法。先通过核方法将低维空间中的非线性特征转化为高维空间中的线性特征,再将其进行线性组合并向低维空间投影,然后用BP神经网络分类器对控制图模式进行识别。通过仿真进行验证,结果显示该方法对控制图各个模式能够有效聚类,并且识别精度得到提高。

关键词: 控制图, 模式识别, 核主元分析, 神经网络, 特征提取

Abstract: Considering the problem that the abnormal features have great similarity so that simple structure and high precision modeling cannot be achieved, a control chart pattern recognition method based on Kernel Principal Component Analysis (KPCA) and neural network was proposed. Firstly, the kernel method was used to translate the nonlinear feature into a higher dimensional linear feature space. Secondly this feature was projected to lower dimensional feature space. Finally the BP neural network classifier was introduced to identify the control chart pattern. This method was verified through stochastic simulation. The result demonstrates that the model can cluster each control chart pattern effectively and improve recognition accuracy.

Key words: control chart, pattern recognition, Kernel Principal Component Analysis (KPCA), neural network, feature extraction

中图分类号: