Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (29): 231-234.

• 工程与应用 • Previous Articles     Next Articles

Feature extraction of diesel engine fault data based on rough set theory

YIN Jie,CHAI Yi,GUO Maoyun   

  1. College of Automation,Chongqing University,Chongqing 400030,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-11 Published:2011-10-11

应用粗糙集提取柴油机故障数据特征

殷 杰,柴 毅,郭茂耘   

  1. 重庆大学 自动化学院,重庆 400030

Abstract: Rough set theory is used to research how to extract features of diesel engine fault data due to its own character.Rough set can deal with discrete data only and most parameters are continuous,so this paper presents a discretization method which uses SOM for rough set;it gives a quick attribute reduction algorithm based on simplified discernibility matrix;it extracts features of 6135D diesel engine fault data,reduces its attributes from 8 to 3 successfully and lays the foundation of follow-up work.

Key words: rough set, Self Organizing Feature Map(SOM), attribute reduction, feature extraction

摘要: 根据柴油机故障数据的特点,采用粗糙集理论对其进行特征提取研究。由于实际测量的参数大多为连续数据,而粗糙集只能处理离散数据,提出了一种适用于粗糙集的SOM网络离散化方法;给出一种基于简化差别矩阵的快速属性约简算法;以6135D型柴油机故障诊断数据为例进行特征提取,成功地将原始8个属性约简为3个,为后续研究工作打下了基础。

关键词: 粗糙集, 自组织特征映射(SOM), 属性约简, 特征提取