论文部分内容阅读
本构关系是联系材料塑性变形行为与工艺参数的桥梁，也是用有限元法模拟金属成形过程的前提条件．开发了一个基于神经网络的Ｔｉ－５Ａｌ－２Ｓｎ－２Ｚｒ－４Ｍｏ－４Ｃｒ合金的本构关系模型．首先利用Ｔｈｅｒｍｅｃｍａｓｔｏｒ－Ｚ型热模拟机等温压缩Ｔｉ－５Ａｌ－２Ｓｎ－２Ｚｒ－４Ｍｏ－４Ｃｒ合金，研究在不同变形温度、变形程度和应变速率等工艺参数条件下流动应力的变化情况．然后用实验所得的热变形工艺参数与性能间的数据来训练具有ＢＰ算法的人工神经网络．训练结束后的神经网络即成为一个知识基的本构关系模型．预测结果和实验结果对比表明神经网络模型具有较高的精度和推广能力．
The constitutive relation is a bridge between the plastic deformation behavior and the technological parameters of the material, and is also a prerequisite for the simulation of the metal forming process by the finite element method. A constitutive model of Ti-5Al-2Sn-2Zr-4Mo-4Cr alloy based on neural network has been developed. First, the isothermal compression of Ti-5Al-2Sn-2Zr-4Mo-4Cr alloy by Thermecmastor-Z thermal simulator was used to study the variation of flow stress under different deformation temperature, deformation degree and strain rate. Then, the artificial neural network with BP algorithm is trained by using the experimental data of thermal deformation process parameters and performance. After training, the neural network becomes a constitutive model of knowledge base. The comparison between the predicted result and the experimental result shows that the neural network model has higher precision and promotion ability.