基于BP神经网络的SP700钛合金本构关系  

Constitutive relationship of SP700 titanium alloy based on BP neural network

作　　者：邱仟 王克鲁[1] 李鑫[1] 王俊[1] 张开铭 高鑫 QIU Qian;WANG Ke-lu;LI Xin;WANG Jun;ZHANG Kai-ming;GAO Xin(School of Aeronautical Manufacturing Engineering,Nanchang Hangkong University,Nanchang 330063,China)

机构地区：[1]南昌航空大学航空制造工程学院,江西南昌330063

出　　处：《塑性工程学报》2021年第11期167-172,共6页Journal of Plasticity Engineering

基　　金：江西省研究生创新专项资金资助项目(YC2020-S539)。

摘　　要：采用Gleeble-3800型热模拟试验机对SP700钛合金进行等温恒应变速率压缩实验,研究其在变形温度700~950℃、应变速率0.001~1 s^(-1)条件下的高温流动行为,同时根据实验数据构建出BP神经网络本构关系模型。结果表明:高温变形过程中的变形温度和应变速率对SP700钛合金的流动应力影响较为显著,流动应力基本上随变形温度的增加而减小,随应变速率的增加而增大,且在变形温度更高时,稳态流动特征更为明显。基于BP神经网络本构关系模型得到的SP700钛合金的流动应力预测值与实验值较为一致,其相关系数和平均相对误差分别为0.998和5.07%,该本构模型对合金流动应力的预测精度较高。The isothermal and constant strain rate compression experiments of SP700 titanium alloy were conducted on Gleeble-3800 thermal simulation testing machine.The high temperature flow behavior was studied under the conditions of deformation temperature of 700-950℃ and strain rate of 0.001-1 s^(-1).Simultaneously,BP neural network constitutive relationship model was constructed based on experimental data.The results show that the deformation temperature and strain rate during high temperature deformation have a significant effect on the flow stress of SP700 titanium alloy.The flow stress basically decreases with the increase of deformation temperature and increases with the increase of strain rate,the steady-state flow characteristics are more obvious when the deformation temperature is higher.The flow stress predicted by the BP neural network constitutive relationship model of SP700 titanium alloy is relatively consistent with the experimental value.The correlation coefficient and average relative error are 0.998 and 5.07%,respectively.The constitutive model can predict the flow stress of alloy with high accuracy.

关 键 词：SP700钛合金 高温流动行为 本构模型 BP神经网络 

分 类 号：TG146.2[一般工业技术—材料科学与工程]