engleski

An application of Artificial Neural Network to the Numerical Modeling of Experimental Data and Simulations of the Extrusion Process

The Artificial Neural Network is employed to find complex dependencies between parameters and output values of the experiment. The procedure is compared with the more conservative approach known as regression analysis. The proposed approach is applied to the experimental analysis of the extrusion of the aluminum hollow sections in order develop ability of numerical simulation of the process. The accuracy and reliability of the method is discussed as well as training efficacy.

numerical simulation; extrusion; artificial neural network

nije evidentirano