MODELING THE T6 HEAT TREATMENT OF Al-Mg-Si ALLOY BY ARTIFICIAL NEURAL NETWORK

Artificial Neural Networks are mathematical modeling tools which has recently been used in the field of prediction and forecasting in engineering applications. In this study the feed-forward neural network with the back-propagation (BP) learning algorithm had been applied. Quench and artificial aging as well as solution treatment temperature and time have been defined as the input parameters of ANN. The output layer of the ANN model consists of hardness. Investigates showed better results when network had a hidden layer with 10 neuron compared to 5 neurons. This model can predict the hardness within an average error of 1% from the experimental values. This simulated ANN model seems to possess an edge over existing constitutive model, like hyperbolic sine equation, and has a great potential to be employed in industries