改进序列前向选择法（ISFS）和极限学习机(ELM)相结合的SPC控制图模式识别方法

张宇波, 蔺小楠

（郑州大学 电气工程学院, 河南 郑州 450001）

摘要： 为了提高SPC（Statistical Process Control）控制图的识别效果，提出了一种采用改进序列前向选择法（ISFS）和极限学习机(ELM)相结合的方法来进行控制图模式识别。首先，对控制图进行特征提取；然后，采用改进的序列前向选择法对特征进行选择，减少了特征间的相关性和冗余性；最后，利用极限学习机来进行模式识别。仿真结果显示，改进方法的识别率可达到987%，从而为控制图提供了一种有效的识别方法。

关键词： 控制图； 模式识别； 序列前向选择法； 极限学习机

中图分类号： TH 165           文献标志码： A

Recognition Method of SPC Control Chart Pattern  Based on ISFS and ELM

ZHANG Yubo, LIN Xiaonan

(College of Electrical Engineering, Zhengzhou University, Zhengzhou 450001,China)

Abstract: In order to improve the effect of the SPC control chart pattern recognition, a new method based on improved sequential forward selection and extreme learning machine was presented in the paper. Firstly, time domain features were extracted from control chart; secondly, a modified sequential forward selection method was used to select the features to reduce the relevance and redundancy between features; finally， extreme learning machine was adopted to identify control chart. Experimental results show that the improved method can achieve a significant classification performance with accuracy of 987%, providing a new method for the control chart recognition.

Key words: control chart; pattern recognition; sequential forward selection; extreme learning machine

 收稿日期： 20140221

基金项目： 高等学校博士学科点专项科研基金项目（20124101120001）.

作者简介： 张宇波（1965—），女，副教授.

 文章编号：16726987（2015）03032205； DOI： 10.16351/j.16726987.2015.03.018