ANN modeling of strain-induced martensite and its applications in metastable austenitic stainless steels

Low mechanical stability of metastable austenitic stainless steels has a profound effect on their mechanical and magnetic properties and their microstructure before and after martensite to austenite reversion. In this paper, the effects of strain and deformation temperature on the volume fraction of strain-induced martensite in the cold-worked AISI 304 stainless steel were modeled by means of artificial neural network (ANN). The appropriate grain refining and safe cold deformation zones were determined from this model. The developed model can be used as a guide for (i) grain refining treatment which needs high amount of strain-induced martensite or (ii) simple cold deformation in which non-magnetic properties are desirable.