Prediction Micro-Hardness of Al-based Composites by Using Artificial Neural Network in Mechanical Alloying

Aluminum composites are one of the most important alloys with a wide range of properties and applications. In this paper, we predict the micro-hardness of aluminum-based alloys by artificial neural method (ANN). First, the effective parameters in mechanical alloying include weight percentage and micro hardness of reinforcement materials, milling time, the ball to powder weight ratio, vial speed, the pressure of presses, sintering time and temperature, selected for inputs and micro-hardness of Al composite considered as the output. A feed-forward back propagation artificial neural network designed with 16 and 10 neurons in the first and second hidden layers, respectively. The created network with the mean percentage error of 5.6% was able to predict micro hardness of the Al composites. Finally, the effect of each parameter was determined by sensitivity analysis which volume fraction of alloying elements, milling speed and sintering time had the highest impact on the micro hardness of Al-based composites.

Prediction Micro-Hardness of Al-based Composites by Using Artificial Neural Network in Mechanical Alloying Keywords:

Aluminum Alloys , Prediction Micro Hardness , Artificial Neural Network , mechanical alloying