Prediction of the Vickers hardness in austempered ductile irons using neural networks

Abstract

To the design engineer, hardness often means an easily measured quantity which indicates something about the strength and heat treatment of a metal. Austempered ductile iron (ADI) is an alloyed and heat treated ductile cast iron which has a good combination of mechanical properties. This paper describes a neural network model created with a Bayesian framework using published data. The model created is capable of successfully expressing the hardness of austempered ductile irons and can be used as a tool in the processing or design of ADI. The computer programs associated with the work have been made freely available at: http://www.msm.cam.ac.uk/map/mapmain.html

Keywords:

• hardness
• austempering
• bainite
• ductile iron
• retained austenite
• neural networks