Ti60合金的流变应力行为及人工神经网络本构模型  

Flow Stress Behavior and Constitutive Model Based on Artificial Neural Network of Ti60Alloy

作　　者：王天祥 鲁世强[1] 王克鲁[1] 欧阳德来[2] 姚泉 Wang Tianxiang;Lu Shiqiang;Wang Kelu;Ouyang Delai;Yao Quan(School of Aeronautical Manufacturing Engineering,Nanchang Hangkong University;School of Materials Science and Engineering,Nanchang Hangkong University)

机构地区：[1]南昌航空大学航空制造工程学院 [2]南昌航空大学材料科学与工程学院

出　　处：《特种铸造及有色合金》2020年第9期1019-1023,共5页Special Casting & Nonferrous Alloys

基　　金：国家自然科学基金资助项目(51964034,51864035,51761029);江西省研究生创新基金资助项目(YC2018-S365)。

摘　　要：采用Thermecmaster-Z型热加工模拟试验机对Ti60合金试样进行等温恒应变速率压缩试验,研究该合金在变形温度为600~950℃,应变速率为0.001~10s-1条件下的流变应力行为,并构建人工神经网络(ANN)本构模型。结果表明,Ti60合金的流变应力随变形温度的升高和应变速率的下降而减小。在不同的变形条件下,Ti60合金的流变应力-应变曲线分别呈现流动稳态型和流动软化型两种特征。基于ANN所建立的本构模型预测精度高,流变应力试验值与预测值之间的相对误差在±5%以内,平均相对误差为1.82%,且相关系数趋近于1,能准确地描述并预测Ti60合金的流变应力行为。The isothermal constant strain rate compression experiment of Ti60alloy samples was carried out by Thermecmaster-Zthermal processing simulation machine.The flow stress behavior of the alloy in at 600~950℃and 0.001~10s-1 was investigated,and the artificial neural network(ANN)constitutive model was constructed.The results reveal that the flow stress of Ti60alloy is decreased with the increase of deformation temperature and the decrease of strain rate.The flow stress-strain curves of Ti60alloy is characterized by steady flow type and soft flow type under different deformation conditions.The constitutive model established based on ANN has the satisfied prediction accuracy.The relative error between the experimental value of flow stress and the predicted value is basically ranged within±5%,and the average relative error is 1.82%,meanwhile,the correlation coefficient tends to 1,which can accurately describe and predict the flow stress behavior of Ti60alloy.

关 键 词：Ti60合金 流变应力行为 人工神经网络 本构模型 

分 类 号：TG146.23[一般工业技术—材料科学与工程] TP311[金属学及工艺—金属材料]