Page 236 - Read Online

Research Article | Open Access

Journal of Materials Informatics

Liu et al. J. Mater. Inf. 2026, 6, 18 DOI:10.20517/jmi.2025.88

Knowledge-enabled data-driven smart design of
ultra-strong ductile near-α titanium alloys under
extreme conditions

Shuo Liu , Xiaoqian Fan , Hongjian Ye , Makhambet Ibragim , Haifeng Song , Xingyu Gao , Gulmira
1
4
1
2,3
4
1
Yar-Mukhamedova , Daniel Zellele , Peixuan Li , William Yi Wang 1,2,* , Jinshan Li 1,2,*
1,*
3
3
Keywords:
Machine learning, near-a
titanium alloy, compress
strength, ductility,
multi-objective optimization
Citation: Liu, S.; Fan, X.;
Ye, H.; Ibragim, M.; Song, H.;
Gao, X.;
Yar-Mukhamedova, G.;
Zellele, D.; Li, P.; Wang, W.
Y.; Li, J. Knowledge-enabled
data-driven smart design of
ultra-strong ductile near-a
titanium alloys under
extreme conditions. J. Mater.
Inf. 2026, 6, 18.
https://dx.doi.org/10.20517
/jmi.2025.88
Abstract
Received: 22 Oct 2025 Under extreme service conditions, adiabatic shear banding critically limits the performance
First Decision: 1 Dec 2025 of titanium alloys in warhead applications, creating an urgent demand for strategies to
Revised: 21 Dec 2025 achieve strength-ductility synergy. In this work, a knowledge-enabled data-driven
Accepted: 7 Jan 2026 multi-objective optimization framework is proposed to investigate the composition of
Published: 7 Apr 2026
near-α titanium alloys under high strain rates. By integrating domain knowledge with
Academic Editor: twelve machine learning models, key performance parameters (KPPs) governing strength
Ming Hu are identified through feature engineering, including strain rate, Fermi energy, and phase
Copy Editor:
Pei-Yun Wang formation parameters, while ductility is controlled by the KPPs of strain rate, bulk/shear
Production Editor: modulus (B/G) ratio, and mixing enthalpy. Using a gradient boosting regression tree model
Pei-Yun Wang for strength prediction [test the coefficient of determination (R ) = 0.91] and a random
2
forest model for ductility prediction (test R = 0.82), the nondominated sorting genetic
2
algorithm II (NSGA-II) is integrated to identify 14 Pareto-optimal alloys from a pool of

1 State Key Laboratory of Solidification Processing, Northwestern Polytechnical University, Xi’an 710072, Shaanxi, China.
2 China-Kazakhstan Belt and Road Joint Laboratory on Materials Genome Engineering and Intelligent Science, Northwestern Polytechnical
University, Xi’an 710072, Shaanxi, China.
3 Department of Solid State Physics and New Materials Technology, Al-Farabi Kazakh National University, Almaty 050040, Republic of
Kazakhstan.
4 Institute of Applied Physics and Computational Mathematics, Beijing 100094, China.

* Correspondence to: Prof. William Yi Wang, Dr. Peixuan Li, Prof. Jinshan Li, State Key Laboratory of Solidification Processing,
Northwestern Polytechnical University, Xi’an 710072, Shaanxi, China. E-mail: wywang@nwpu.edu.cn; li_peixuan@nwpu.edu.cn;
ljsh@nwpu.edu.cn

www.oaepublish.com Submit a Manuscript: https://ucenter.oaepublish.com