SPRAY FORMING QUALITY PREDICTIONS VIA NEURAL NETWORKS

To produce consistently high-quality spray-formed parts, correlations must be made between the input process parameters and the final part quality. The Spray Forming Technology Group at the Naval Surface Warfare Center decided to ''model'' this correlation through the use of artificial neural networks. In this study, neural networks accurately predicted trends in spray forming process outputs based on variations in process inputs. The graphs generated by the neural network prediction help to define the optimal operating region for the spray forming process and indicate the effect of changing input process parameters on final part quality.