Modeling the influence of a process control agent on the properties of metal matrix composite powders using artificial neural networks

Abstract

An artificial neural network (ANN) model was developed for modeling the effect of the amount of a process control agent (PCA) on the apparent density, particle size and microhardness of Al–10wt.%Al2O3 composite powders. Methanol was used as the process control agent in varying amounts of 1, 2, and 3wt.% to investigate the effect of the PCA on the properties of the composite powders. Composite powders were fabricated using high energy planetary ball milling, and the training data for the ANN were collected from experimental results. After the training process, the experimental data were used to verify the accuracy of the system. The properties of the composite powders were analyzed using a hall flowmeter for the apparent density, a laser particle size analyzer for the particle size, a microhardness tester for the powder microhardness and a scanning electron microscopy (SEM) for the powder morphology. The two input parameters in the proposed ANN were the amount of methanol and duration of the milling process. Apparent density, particle size and microhardness of the composite powders were the outputs obtained from the proposed ANN. As a result of this study the ANN was found to be successful for predicting the apparent density, particle size and microhardness of Al–Al2O3 composite powders. The mean absolute percentage error (MAPE) for the predicted values didn't exceed 4.93%. This model can be used for predicting Al–Al2O3 composite powder properties produced with different amounts of methanol and using different milling times.