ANN modelling for the determination of moulding sand matrix grain size

One of the modern methods of the production optimisation are artificial neural networks. Neural networks are gaining broader and broader application in the foundry industry, among others for controlling melting processes in cupolas and in arc furnaces, for designing castings and supply systems, for controlling moulding sand processing, for predicting properties of cast alloys or selecting parameters of pressure castings. An attempt to apply neural networks for controlling the quality of bentonite moulding sands is presented in this paper. This is the assessment method of sands suitability by means of detecting correlations between their individual parameters. The presented investigations were obtained by using the Statistica 9.0 program. The aim of the investigations was to select the neural network suitable for prediction the moulding sand matrix grain size on the basis of the determined sand properties such as: permeability, compactibility, and compressive strength.