K-nearest neighbors based methods for identification of different gear crack levels under different motor speeds and loads: Revisited
详细信息 查看全文
- 作者：Dong Wang dongwang4-c@my.cityu.edu.hk" class="auth_mail" title="E-mail the corresponding author
- 关键词：Gear crack ; Fault diagnosis ; K-nearest neighbors ; Redundant statistical features ; Dimensionality reduction
- 刊名：Mechanical Systems and Signal Processing
- 出版年：2016
- 出版时间：March 2016
- 年：2016
- 卷：70-71
- 期：Complete
- 页码：201-208
- 全文大小：2543 K
文摘
- listitem" id="list_li0005">
- •
620 redundant statistical features are extracted.
- •
New significant features are generated from 620 redundant features.
- •
The developed method is better than other 4 K-nearest neighbors based methods.
- •
Selection of the number of new significant features is investigated.
- •
Parameter selection of K-nearest neighbors is investigated.