Naslov
Modular Structure of Tool Condition Monitoring System

Autori
Mulc, Tihomir ; Udiljak, Toma ; Brezak, Danko

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
Proceeding of the 3rd International Conference on Sustainable Life in Manufacturing SLIM 2012, Istanbul, 2012, pp. 95-102 / Oguz Colak ; Janez Kopač - Istanbul : CAD/CAM Research and Application Center, Suleyman Demirel University, Turkey, 2012, 95-102

ISBN
978-9944-452-61-8

Skup
International Conference on Sustainable Life in Manufacturing SLIM 2012

Mjesto i datum
Istanbul, Turska, 02.10.2012. - 05.10.2012

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
tool condition monitoring; signal selection; artificial neural network; modular concept

Sažetak
A tool condition monitoring (TCM) system has to ensure reliable monitoring of the condition of cutting tool edge. The system should be simple to configure for different cutting conditions and the obtained results should be interpreted in a unambiguous way. Implementation of monitoring system involves finding solutions to a number of sub-tasks such as selection of the type of signal, the position and way sensors are installed, definition of sampling rate, selection of samples, signal filtering, methods for the selection and ranking of features, selection of appropriate type of artificial neural network (ANN), the process of learning and creation of knowledge base, user friendly procredure for operating TCM, and methods of reacting on established tool condition. Based on these considerations, a modular concept of tool condition monitoring system is proposed. The concept has been tested on machining processes with variable cutting depths and material hardness.

Izvorni jezik
Engleski

Znanstvena područja
Strojarstvo

POVEZANOST RADA