Application of Artificial Neural Networks to Predict the Carbon Content and the Grain Size for Carbon Steels

Neural networks and generic algorithms are two branches of artificial
intelligence that can provide many benefits in engineering applications. The
artificial neural networks (ANN) technologies provide on-line capability to
analyze many inputs and provide information to multiple outputs, and also,
have the capability to learn or adapt to changing conditions. No doubt that the
determination of either of the carbon content or the grain size of carbon steel is
a time consuming process; which involves a quite tedious work. This paper
examines the feasibility of using an integration system between some measured
ultrasound parameters; from nondestructive test (NDT), and a pre - learned
ANN to facilitate the determination of grain size and carbon content for the
tested samples. The results showed that grain size and carbon content of carbon
steels can be well predicted using a trained neural networks, with an
acceptable degree of errors and great reliability.