Nalazite se na CroRIS probnoj okolini. Ovdje evidentirani podaci neće biti pohranjeni u Informacijskom sustavu znanosti RH. Ako je ovo greška, CroRIS produkcijskoj okolini moguće je pristupi putem poveznice www.croris.hr
izvor podataka: crosbi

Parametric and nonparametric modelling of earing and hardness of deep drawn cups (CROSBI ID 137821)

Prilog u časopisu | izvorni znanstveni rad | međunarodna recenzija

Lela, Branimir ; Duplančić, Igor ; Bajić, Dražen Parametric and nonparametric modelling of earing and hardness of deep drawn cups // Materials science and technology, 25 (2009), 8; 969-975

Podaci o odgovornosti

Lela, Branimir ; Duplančić, Igor ; Bajić, Dražen

engleski

Parametric and nonparametric modelling of earing and hardness of deep drawn cups

This study compares different approaches in modelling the earing phenomenon and hardness of cups in deep drawing process. The blank holder force (BHF), annealing temperature (AT) and annealing time (ANT) of blanks prior to deep drawing process have been chosen as the three influential parameters on the earing and hardness. In order to obtain mathematical models for the earing and hardness of the deep drawn cups, the methodology of artificial neural networks (ANNs) have been used. Bayesian network (BN), radial basis function network (RBFN), Gaussian processes (GP) and multilayer perceptron (MLP) are four different ANN approaches that have been used for the modelling. The research has been conducted on a cold-rolled Al-Fe-Si (AA8011A) aluminium sheet. After obtaining the mathematical models describing the influence of BHF and annealing on hardness and earing, a comparison of the proposed models has been performed. A search for the optimal parameters of deep drawing process has been carried out too.

Deep drawing; Neural networks; AA8011A; Earing; Hardness

nije evidentirano

nije evidentirano

nije evidentirano

nije evidentirano

nije evidentirano

nije evidentirano

Podaci o izdanju

25 (8)

2009.

969-975

objavljeno

0267-0836

Povezanost rada

Strojarstvo

Indeksiranost