Application of back-propagation neural network technique to high-energy planetary ball milling process for synthesizing nanocomposite WC–MgO powders

A series of artificial-neural-network (ANN) models is developed for the analysis and prediction of correlations between processing (high-energy planetary ball milling) parameters and the morphological characteristics of nanocomposite WC–18at.%MgO powders by applying the back-propagation (BP) neural network technique. The input parameters of the BP network are milling speed, milling ball diameter and ball-to-powder weight ratio. The properties of the as-milled powders (specifically crystallite size, specific surface area and median particle size) are the output for three individual BP network models. These models are based on the mathematic statistical approach and seem suitable for the complicated ball milling process which is difficult to be accurately described by physical models. Well acceptable performances of the neural networks are achieved. The model can be used for the prediction of properties of composite WC–MgO powders at various milling parameters. It can also be used for the optimization of processing and ball milling parameters.