Unveiling the quantitative relationship between microstructural features and quasi-static tensile properties in dual-phase titanium alloys based on data-driven neural networks

The quasi-static mechanical properties of alpha+(3 dual-phase titanium alloys are susceptible to their microstructural features, presenting a complex, high-dimensional nonlinear relationship, which hinders the rapid development of high-performance materials. In this work, 4065 micro-representative models were virtually constructed with varying volume fractions of alpha and (3 phases and characteristic dimensions via high-throughput finite element simulation, incorporating a cohesive zone model to simulate the interfaces between the two phases. Especially, the established representative models were experimentally verified by two groups of real material microstructures, and the results showed that the relative errors were not more than 9.5% in microstructural characteristics and quasi-static mechanical properties. Afterward, a neural network model was developed to correlate the quasistatic tensile properties with the microstructural features of the dual-phase TC6 titanium alloys, achieving an 88.2 % accuracy in predicting overall mechanical performance. Utilizing the Shaply Additive Explanation method, it was found that the primary alpha phase's volume fraction and the secondary alpha phase's width were the most significant microstructural features affecting quasi-static strength. Specifically, the volume fraction of the primary alpha phase and the width of the secondary alpha phase negatively affected strength, while the width of the secondary alpha phase positively influenced plasticity. Notably, the primary alpha phase's volume fraction had a quadratic curve pattern of influence on plasticity. The intrinsic mechanisms behind these laws were further revealed based on local stress-strain responses and crack propagation analysis. Ultimately, the optimal microstructural features with strength-plasticity balance were identified through the lower threshold method: a secondary alpha phase width of about 1 mu m and a primary alpha phase volume fraction ranging from 0.1 to 0.2, effectively facilitating microstructure design.