陶瓷的增材制造，特别是通过vat-光聚合，由于其高精度和生产复杂几何形状的能力，提供了巨大的潜力。这项研究解决了vat-光聚合的主要挑战：开发粘合剂配方，优化粘度和机械性能，同时适应高陶瓷负载。本研究引入了监督机器学习（ML）算法，作为预测粘合剂配方粘度和拉伸强度的新方法。使用全因子实验设计生成了一个全面的数据集，其中三个因素在三个水平上。对各种ML算法在回归应用中的有效性进行了评估。由于数据集规模较小，采用留一交叉验证（LOOCV）方法来评估这些机器学习算法的性能。具有5个隐藏节点的ANN模型取得了出色的结果，粘度的平均绝对误差为2.504 mPa·s，抗拉强度的平均绝对偏差为1.163 mPa，优于其他ML模型。ANN模型特别擅长捕捉复杂的非线性关系。包含Okoruwa最大饱和电位（OMSP）区域和峰值位置作为输入特征，显著提高了粘度和机械性能的预测精度。这项研究证明了ML算法在彻底改变VPP粘合剂树脂配方过程方面的巨大潜力。

This study introduces Ti6Al4V2Fe, a novel dual-phase titanium alloy, manufactured via Cold Metal Transfer Directed Energy Deposition (CMT-DED) for aerospace applications. This alloy meets stringent performance demands and enables efficient additive manufacturing of large-scale components. We investigated its microstructural evolution and mechanical properties under static and dynamic conditions. Room temperature tensile tests and dynamic impact assessments (1500/s to 3000/s) revealed a refined microstructure with short columnar and fine equiaxed prior-β grains, influenced by undercooling and thermal gradients. The addition of Fe enhanced β phase nucleation, resulting in a basket-weave α + β structure. Dynamic tests highlighted increasing strength with strain rate, with horizontal samples displaying superior strength. A constitutive model confirmed the experimental stress-strain profiles. At 2500/s, adiabatic shear bands indicated heightened sensitivity in horizontal samples, while vertical samples absorbed more energy, linked to dynamic recrystallization. The Ti6Al4V2Fe alloy, fabricated by CMT-DED, demonstrates remarkable strength, enhanced toughness, and reduced anisotropy, marking it as a prime candidate for aerospace applications.