Intelligent Classification of Wear Particles Based on Deep Convolutional Neural Network

被引：6

作者：

Jia, Fengguang ^{[1
,2
]}

Yu, Fulin ^{[1
]}

Song, Lei ^{[1
]}

Zhang, Shaojun ^{[1
]}

Sun, Hongyuan ^{[1
]}

机构：

[1] Shandong Jiaotong Univ, Coll Naval Architecture & Marine Engn, Weihai 264209, Peoples R China

[2] Shanghai Maritime Univ, Coll Merchant Marine, Shanghai 201306, Peoples R China

来源：

4TH INTERNATIONAL CONFERENCE ON MECHANICAL, AERONAUTICAL AND AUTOMOTIVE ENGINEERING (ICMAA 2020) | 2020年 / 1519卷

关键词：

DEBRIS ANALYSIS; TEXTURE; SYSTEM;

D O I：

10.1088/1742-6596/1519/1/012012

中图分类号：

V [航空、航天];

学科分类号：

08 ; 0825 ;

摘要：

The intelligent classification of wear particles has remained a high priority research area for ferrography technology and industrial tribology. In this study, five deep convolutional neural network (DCNN) models are used for identification of seven kinds of wear particles. Instead of manually designing and selecting the features, the proposed DCNN realizes an endto-end processing. Wear particles' dataset is built by various kinds of tribology experiments. The experimental results show that the accuracy of DenseNet121 on the test set is 88.39%. A conclusion can be drawn that DCNN is suited for wear particles' classification and can be put into practical use in condition monitoring systems in the future.

引用

页数：13

共 28 条

[1]

Abadi M., 2016, 12 USENIX

[2]

[Ahmed Akhter IFPRI) and DATA team IFPRI) and DATA team], 2008, [No title captured]

[3]

[Anonymous], 2014, WEAR, DOI DOI 10.1016/J.WEAR.2014.04.014

[4]

[Anonymous], 2013, IND LUBR TRIBOL, DOI DOI 10.1108/00368791311292756

[5]

[Anonymous], 2005, TRIBOLOGY LETT, DOI DOI 10.1007/S11249-005-8303-1

[6]

[Anonymous], 1998, TRIBOLOGY LETT

[7]

[Anonymous], 2007, CORROS REV

[8]

[Anonymous], 2002, WEAR

[9]

[Anonymous], 2008, TRIBOLOGY INT, DOI DOI 10.1016/J.TRIBOINT.2007.04.004

[10]

[Anonymous], 2006, TRIBOLOGY INT, DOI DOI 10.1016/J.TRIBOINT.2006.01.019

← 1 2 3 →