Analysis and Optimization of the Properties of High-Strength Austenitic Steels by Approximation of a Primary Database

Abstract

The study recommends an approach for adequate prediction of the properties of high-strength austenitic steels with variation of the elements C, Si, Mn, Cr and Al by approximation with polynomials and second-order regression models. According to pre-set requirements, analysis and multicriteria optimization are performed, on the basis of which compositions are obtained, outside the database used for approximation. The models obtained in the research can be used to simulate and optimize the alloying components of steels, which improve the quality in terms of properties while considering the cost in terms of the amount of the alloying element. Compositions with simultaneously high (over 68%) values of strength and plastic properties from the maximal ones have been determined. The compositions were obtained through the author's approach to analysis and multicriteria optimization – DEFMOT1. 1 * DEFMOT computational approach for single-criteria and multi-criteria optimization, working on the principle of “determination by motion”.