Optimization of Cooling Rate for Bainite Evolution in AHSS using Machine Learning

Abstract

Mechanical properties of Advanced-High-Strength-Steel (AHSS) are linked to its microstructure, influenced by processing techniques during production, specifically by hot rolling thermo-mechanical processing. In this article, a novel adaptive machine-learning (ML) model coupled with controlled-cooling of hot-rolled plates were developed to predict bainite in AHSS. A neural-network model of the time-temperature-transformation diagram was used at each cooling step to predict continuous-cooling-transformation kinetics. To verify bainite fraction, dilatometry experiments were performed with AHSS specimens cooled at rates from 0.1 to 10°C/s. An adaptive-ML model for bainite was trained using inputs from experiments and simulation, offering a predictive tool for optimizing AHSS processing.

Department(s)

Mechanical and Aerospace Engineering

Second Department

Materials Science and Engineering

Keywords and Phrases

Advanced high strength steel; Bainite; Continuous-cooling-transformation; Neural network; Optimization; Time-temperature-transformation

International Standard Book Number (ISBN)

978-093076737-2

International Standard Serial Number (ISSN)

1551-6997

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2025 Association for Iron and Steel Technology, All rights reserved.

Publication Date

01 Jan 2025

Link to Full Text

Share

 
COinS