Możliwości zastosowania sztucznych sieci neuronowych do modelowania procesów odlewniczych

The application of the artificial neural networks to solving various problems in foundry industry, is reviewed The fundamentals of the ductile iron metallurgy and its influence on the material's properties are presented. The general usefulness of the neural networks for modeling melting processes is shown. Several versions of the multi-layer networks with back propagation, including additional modification of the input signals, have been constructed. Records of the melt control parameters as well as the testing results of the ductile iron have been collected in one of the Polish foundries. The results of the network training have shown that neural network modeling is a promising tool for the prediction of the ductile iron melting results.