Hybrid Feature Selection in Fault Diagnosis

摘要：

To reduce complexity in design of fault diagnosis system for large scale equipments, a hybrid feature selection algorithm is put forth. By introduction of Markov Blanket, reluctant features can be effectively eliminated to decrease the feature space for input parameters of diagnosis system design. An improved ChISquare method with introduction of frequency, distribution and concentration is adopted to improve the relevance evaluation performance of the Markov Blanket. The hybrid feature selection algorithm showed high performance in design and implementation of an aeroengine automatic fault diagnosis system based on both neural network and decision tree.

关键词： feature selection fault dianosis Markov blanket

作者: Jian-Feng Yan

作者单位: School of Computer Science & Technology Soochow University Suzhou, China

会议类型: 国际会议

会议名称: 2010 2nd IEEE International Conference on Information Management and Engineering(2010年IEEE第二届信息管理与工程国际会议 IEEE ICIME 2010)

会议地点: 成都

会议语种:英文

页码: 1-5

在线出版日期: 2010-04-16（万方平台首次上网日期，不代表论文的发表时间）