研究了变形参数对Ti 15 3合金热变形后晶粒轴比的影响 ,并采用人工神经网络方法建立了其晶粒轴比与变形程度、变形速率和变形温度之间的数学模型。将此模型与热力耦合刚粘塑性有限元方法相结合 ,对Ti 15 3合金热反挤成型过程的晶粒轴比场进行数值模拟和相应实验研究 ,结果表明 ,预测值与实测值吻合较好。 The effect of deformation parameters on the grain axial ratio of Ti 15 3 alloy after hot deformation was studied. The mathematical model of the grain axial ratio and deformation degree, deformation rate and deformation temperature was established by artificial neural network. This model was combined with the thermomechanically coupled rigid-viscoplastic finite element method to simulate the grain-axial ratio field of the hot-forging process of Ti 15 3 alloy and the experimental results show that the predicted value is in good agreement with the measured value .