Wear Debris Identification with Neural Networks

Joichi Sugimura, Akihiko Umeda, Yuji Yamamoto

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

A feedforward neural network is applied to identification of wear debris generated under different sliding conditions. In order to describe characteristics of debris of various shapes and sizes, four representative parameters for groups of randomly sampled wear debris, i. e. 50% volumetric diameter, average elongation, average roundness and average reflectivity, are used as inputs to the network. Debris sampled in five steel sliding experiments are chosen as examples. It is shown that identification results depend on the ranges of these parameters, and that the ranges are determined by the sample size used for averaging. In the present case, a sample size of fifty pieces of debris provides a satisfactory identification result with less than ten percent error. It is also demonstrated that use of data for larger particles leads to better results. We discuss how the network determines difference in debris features, and how this approach can be applied to diagnosis of sliding surfaces.

Original languageEnglish
Pages (from-to)4055-4060
Number of pages6
JournalTransactions of the Japan Society of Mechanical Engineers Series C
Volume61
Issue number590
DOIs
Publication statusPublished - Jan 1 1995

Fingerprint

Debris
Wear of materials
Neural networks
Feedforward neural networks
Elongation
Steel
Experiments

All Science Journal Classification (ASJC) codes

  • Mechanics of Materials
  • Mechanical Engineering
  • Industrial and Manufacturing Engineering

Cite this

Wear Debris Identification with Neural Networks. / Sugimura, Joichi; Umeda, Akihiko; Yamamoto, Yuji.

In: Transactions of the Japan Society of Mechanical Engineers Series C, Vol. 61, No. 590, 01.01.1995, p. 4055-4060.

Research output: Contribution to journalArticle

@article{7849231fccae433da6e4692f2ad66cd8,
title = "Wear Debris Identification with Neural Networks",
abstract = "A feedforward neural network is applied to identification of wear debris generated under different sliding conditions. In order to describe characteristics of debris of various shapes and sizes, four representative parameters for groups of randomly sampled wear debris, i. e. 50{\%} volumetric diameter, average elongation, average roundness and average reflectivity, are used as inputs to the network. Debris sampled in five steel sliding experiments are chosen as examples. It is shown that identification results depend on the ranges of these parameters, and that the ranges are determined by the sample size used for averaging. In the present case, a sample size of fifty pieces of debris provides a satisfactory identification result with less than ten percent error. It is also demonstrated that use of data for larger particles leads to better results. We discuss how the network determines difference in debris features, and how this approach can be applied to diagnosis of sliding surfaces.",
author = "Joichi Sugimura and Akihiko Umeda and Yuji Yamamoto",
year = "1995",
month = "1",
day = "1",
doi = "10.1299/kikaic.61.4055",
language = "English",
volume = "61",
pages = "4055--4060",
journal = "Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C",
issn = "0387-5024",
publisher = "The Japan Society of Mechanical Engineers",
number = "590",

}

TY - JOUR

T1 - Wear Debris Identification with Neural Networks

AU - Sugimura, Joichi

AU - Umeda, Akihiko

AU - Yamamoto, Yuji

PY - 1995/1/1

Y1 - 1995/1/1

N2 - A feedforward neural network is applied to identification of wear debris generated under different sliding conditions. In order to describe characteristics of debris of various shapes and sizes, four representative parameters for groups of randomly sampled wear debris, i. e. 50% volumetric diameter, average elongation, average roundness and average reflectivity, are used as inputs to the network. Debris sampled in five steel sliding experiments are chosen as examples. It is shown that identification results depend on the ranges of these parameters, and that the ranges are determined by the sample size used for averaging. In the present case, a sample size of fifty pieces of debris provides a satisfactory identification result with less than ten percent error. It is also demonstrated that use of data for larger particles leads to better results. We discuss how the network determines difference in debris features, and how this approach can be applied to diagnosis of sliding surfaces.

AB - A feedforward neural network is applied to identification of wear debris generated under different sliding conditions. In order to describe characteristics of debris of various shapes and sizes, four representative parameters for groups of randomly sampled wear debris, i. e. 50% volumetric diameter, average elongation, average roundness and average reflectivity, are used as inputs to the network. Debris sampled in five steel sliding experiments are chosen as examples. It is shown that identification results depend on the ranges of these parameters, and that the ranges are determined by the sample size used for averaging. In the present case, a sample size of fifty pieces of debris provides a satisfactory identification result with less than ten percent error. It is also demonstrated that use of data for larger particles leads to better results. We discuss how the network determines difference in debris features, and how this approach can be applied to diagnosis of sliding surfaces.

UR - http://www.scopus.com/inward/record.url?scp=0029390426&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029390426&partnerID=8YFLogxK

U2 - 10.1299/kikaic.61.4055

DO - 10.1299/kikaic.61.4055

M3 - Article

AN - SCOPUS:0029390426

VL - 61

SP - 4055

EP - 4060

JO - Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C

JF - Nippon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C

SN - 0387-5024

IS - 590

ER -