Accelerating the Discovery of New DP Steel Using Machine Learning-Based Multiscale Materials Simulations
- PDF / 2,022,020 Bytes
- 12 Pages / 593.972 x 792 pts Page_size
- 16 Downloads / 216 Views
APPLICATIONS of artificial intelligence and machine learning in the automotive industry are growing rapidly with the advancements in information technologies and computational hardware. In this work, the goal is to leverage machine learning to accelerate and improve the design of materials. Specifically, by
ABDALLAH A. CHEHADE and GEORGES AYOUB are with the Department of Industrial and Manufacturing Systems Engineering, University of Michigan-Dearborn, Dearborn, MI. Contact e-mail: [email protected] TAREK M. BELGASAM is with the Materials Research Division, Honda R&D America, Inc., Raymond, OH and also with the Mechanical Engineering Department, Faculty of Engineering, University of Benghazi, Benghazi, Libya. HUSSEIN M. ZBIB is with the School of Mechanical and Materials Engineering, Washington State University, Pullman, WA. Manuscript submitted January 13, 2020.
METALLURGICAL AND MATERIALS TRANSACTIONS A
modeling the mechanical properties as a function of the microstructure parameters, we attempt to accelerate the discovery of dual-phase (DP) steel suitable for use in crash-relevant automotive components.[1,2] In recent years, the need to manufacture fuel-efficient vehicles has resulted in a significant increase in the use of DP steels. DP steels exhibit high-strength properties compared to conventional steel materials, and therefore thinner structures can be used while maintaining the same safety criteria. Furthermore, its good combination of strength and ductility (formability) makes DP steel a perfect candidate for cold forming processes. For all those reasons, implementing DP steels for crashworthy components is of high interest to the automotive industry. DP steel is an advanced high-strength steel (AHSS) presenting a heterogeneous microstructure consisting of a ferritic (a) matrix containing hard martensitic (a0 ) second-phase inclusions. Generally, DP steels contain a purely ferrite phase as a matrix with about a
3.3 to 47 pct fraction of martensite islands spread as a hard phase over a matrix. The mechanical properties of DP steel are significantly affected by a number of parameters, such as the volume fraction of martensite, morphology, and carbon content. The design of DP steel alloys to meet the requirement of a specific structural application is very challenging due to the complexity of their microstructure.[3–8] This heterogeneous microstructure and the fraction volume of the two phases significantly influence the mechanical behavior of DP steel resulting from deformation mechanisms acting at different length scales. The complex structure–property relationship was studied recently in research efforts combining experimental characterization techniques with advanced multiscale-based models.[9–12] The distribution of stress and strain in the two phases and the overall response of the DP steel was modeled using various micromechanical models, such as empirical rules of mixture,[13,14] an Eshelby-based homogenization approach,[15–18] and finite-element-based simulations using representative volume element.[19–23] To
Data Loading...
