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

Abstract

Neural networks and generic algorithms are two branches of artificialintelligence that can provide many benefits in engineering applications. Theartificial neural networks (ANN) technologies provide on-line capability toanalyze many inputs and provide information to multiple outputs, and also,have the capability to learn or adapt to changing conditions. No doubt that thedetermination of either of the carbon content or the grain size of carbon steel isa time consuming process; which involves a quite tedious work. This paperexamines the feasibility of using an integration system between some measuredultrasound parameters; from nondestructive test (NDT), and a pre - learnedANN to facilitate the determination of grain size and carbon content for thetested samples. The results showed that grain size and carbon content of carbonsteels can be well predicted using a trained neural networks, with anacceptable degree of errors and great reliability.