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

Under extreme service conditions, adiabatic shear banding critically limits the performance of titanium alloys in warhead applications, creating an urgent demand for strategies in strength-ductility synergy. In this work, a knowledge-enabled data-driven multi-objective optimization framework is proposed to investigate the composition of near-α titanium alloys under high strain rates. By integrating domain knowledge with twelve machine learning models, key property parameters (KPPs) governing strength are identified based on feature engineering, including strain rate, Fermi energy, and phase formation parameter, while ductility is controlled by the KPPs of strain rate, B/G ratio, and mixing enthalpy. Using a Gradient Boosting Regression Tree model for strength prediction (test R² = 0.91) and a Random Forest model for ductility prediction (test R² = 0.82), the Nondominated Sorting Genetic Algorithm-II (NSGA-II) is integrated to identify 14 Pareto-optimal alloys from a pool of 200,000 near-α titanium alloy Ti-Al-V-Mo-Zr-Sn candidate compositions. A breakthrough combination of 1,600 MPa dynamic compressive strength and 26% ductility at the strain rate of 3,000 s^-1 is achieved, which is better than that of TA15 alloy by constitutive equation model confirming. This framework successfully designed novel near α-titanium alloys with strength-ductility synergy by knowledge-driven feature engineering and multi-objective optimization algorithms, establishing a new paradigm for the intelligent design of titanium alloys under extreme conditions.