Modeliranje procesa frezanja več-slojnih kovinskih materialov

magistrsko delo

Antonio Radić (Author), Uroš Župerl (Mentor), Aleš Hace (Mentor), Franc Čuš (Co-mentor)

Abstract

Laser Engineered Net Shaping (LENS) je sodobna dodajalna tehnologija nanašanja kovinskih materialov, pri kateri nastajajo več-slojni materiali. Pri obdelavi več-slojnih kovinskih materialov s frezanjem nastopajo težko določljive rezalne sile zaradi variabilne trdote materiala. Metoda nevronskih mrež je uporabljena za modeliranje povprečne rezalne sile, ki nastopa pri obdelavi več-slojnega kovinskega materiala. Z uporabo enakega orodja je modeliran časovni potek sil. Uporabljene nevronske mreže napovejo rezalne sile z napako manjšo od 0,1%. V drugem delu naloge so opisane meritve debeline navarjenega sloja več-slojnega kovinskega materiala. Za to nalogo je izdelan uporabniški vmesnik, ki avtomatsko meri debelino navarjenega sloja z izbiro mikroskopske metalografske fotografije vzorca več-slojnega kovinskega materiala iz vgrajene baze fotografij.

Keywords

frezanje;več-slojni kovinski materiali;modeliranje rezalne sile;nevronske mreže;merjenje;debelina sloja;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2015
Typology:	2.09 - Master's Thesis
Organization:	UM FS - Faculty of Mechanical Engineering
Publisher:	[A. Radić]
UDC:	004.89:621.914-047.58(043.2)
COBISS:	18824214
Views:	1244
Downloads:	79
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Multi-layered metal materials milling process modelling
Secondary abstract:	Laser Engineered Net Shaping (LENS) is modern additive technology of material deposing. Multi-layered metal materials are made by this technology. Multi-layered metal materials machining creates hardly determinable cutting forces which are affected by material variable hardness. Average cutting force during multi-layered metal material machining is modelled by neural networks. Force time-series is modelled by the same tool. Neural networks that are used are giving very good results with prediction error less than 0,1 %. Second part deals with multi-layered metal material cladded layer measurements. For this task was made graphic user interface which measures cladded layer thickness automatically from embedded metallographic microscopic picture database
Secondary keywords:	milling;mult-layered material;cutting force modelling;neural networks;measurement;cladded layer thickness;
URN:	URN:SI:UM:
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za strojništvo, Mehatronika
Pages:	VIII, 56 f.
ID:	8752547