Development of hybrid aluminum nanocomposites for automotive applications: An in-depth analysis using experimental approaches and predictive machine learning techniques

The demand for lightweight, high-performance materials has driven advancements in aluminum matrix hybrid nanocomposites (AMHCs), but their wear resistance remains a challenge, particularly for automotive applications. This study integrates experimental characterization and machine learning (ML) to p...

Full description

Saved in:
Bibliographic Details
Main Authors: Chitti Babu Golla, R. Narasimha Rao, Syed Ismail, Mutlu Özcan, P. Syam Prasad
Format: Article
Language:English
Published: Elsevier 2025-05-01
Series:Journal of Materials Research and Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2238785425009354
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The demand for lightweight, high-performance materials has driven advancements in aluminum matrix hybrid nanocomposites (AMHCs), but their wear resistance remains a challenge, particularly for automotive applications. This study integrates experimental characterization and machine learning (ML) to predict wear behavior and optimize composite design. AMHCs were synthesized using ultrasonic-assisted stir casting, incorporating 4 wt% titanium carbide (TiC) while varying graphite (Gr) content (0, 2, 4, and 6 wt%). Microstructural analysis using FE-SEM, XRD, and EDS confirmed uniform dispersion and grain refinement. The Al-4Gr nanocomposite exhibited the best mechanical properties, with 30.5 % higher hardness, a 40.4 % increase in ultimate tensile strength (UTS), and a 36.3 % improvement in yield strength compared to the base alloy. Wear resistance improved significantly, with wear rates reduced by 15 %, 30 %, 25.2 %, and 22.8 %, while the friction coefficient decreased by 5 %, 7.9 %, 14.8 %, and 10 %, respectively, as graphite content increased. Key wear mechanisms—adhesion, delamination, abrasion, oxidation, and plastic deformation—were identified, providing insights into tribological performance. The random forest model achieved high predictive accuracy (R2 = 0.93), demonstrating ML's potential for wear rate forecasting. This study offers a cost-effective, ML-driven approach to developing wear-resistant, lightweight AMHCs, advancing their applications in automotive and aerospace industries.
ISSN:2238-7854