Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
A machine-learning approach for a sintering process using a neural network
Autore:
Shigaki, I; Narazaki, H;
Indirizzi:
Osaka Inst Technol, Dept Ind Management, Asahi Ku, Osaka 535, Japan Osaka Inst Technol Osaka Japan 535 anagement, Asahi Ku, Osaka 535, Japan Kobe Steel Ltd, Elect & Informat Technol Res Lab, Nishi Ku, Kobe, Hyogo, Japan Kobe Steel Ltd Kobe Hyogo Japan ol Res Lab, Nishi Ku, Kobe, Hyogo, Japan
Titolo Testata:
PRODUCTION PLANNING & CONTROL
fascicolo: 8, volume: 10, anno: 1999,
pagine: 727 - 734
SICI:
0953-7287(199912)10:8<727:AMAFAS>2.0.ZU;2-D
Fonte:
ISI
Lingua:
ENG
Keywords:
neural network; knowledge acquisition; machine learning; manufacturing technology; product innovation;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
Citazioni:
4
Recensione:
Indirizzi per estratti:
Indirizzo: Shigaki, I Osaka Inst Technol, Dept Ind Management, Asahi Ku, Omiya 5-6-1,Osaka 535,Japan Osaka Inst Technol Omiya 5-6-1 Osaka Japan 535 Osaka 535,Japan
Citazione:
I. Shigaki e H. Narazaki, "A machine-learning approach for a sintering process using a neural network", PROD PLAN C, 10(8), 1999, pp. 727-734

Abstract

This paper presents a machine-learning approach using a multi-layered neural network (NN) with application to a sintering process in an iron- and steer-making plant. Our method induces 'operational rules' that determine operational conditions to obtain products that meet a given quality specification. In our application, an operational condition decides the appropriate ranges of chemical composition and heat input to obtain sinter with desirableproperties. Our approach consists of two stages. First, backpropagation (BP) training is performed to obtain a NN which decides whether a given condition is appropriate or not. Secondly, from the trained NN, we extract ruleswhich explain what operational conditions are appropriate. In spite of theeffective learning capability, a major drawback of a NN is 'unreadability'of the learned knowledge, or the lack of an explanatory capability, which is crucial in the second stage. We developed a rule extraction algorithm which contributes to overcoming this 'unreadability'. The extracted rules arefound to agree well with the knowledge in material science.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 11/08/20 alle ore 15:50:12