engleski

Optimization of aluminum extrusion process by genetic algorithms and neural networks

In this paper new approach for optimization of aluminum extrusion process is proposed. It is based on Artificial Neural Network (ANN) and Genetic Algorithm (GA) which is applied on the analysis of metal flow in extrusion of aluminum tubes by means of bridge die in laboratory conditions. The neural network is trained on the extrusion experiment data to be able to describe complex dependencies between influence parameters and outgoing significant values. A genetic algorithm based optimization of the technological process is proposed, where the extrusion parameters are treated as heritage information and neural network is used as fitness function. The reliability of artificial neural network as an objective function is investigated in comparison with regression analysis.

Artificial Neural Network (ANN); Genetic Algorithm (GA); Extrusion; Hollow sections; Objective function; Optimization

nije evidentirano