TC21合金超塑性本构关系的BP人工神经网络模型  

Constitutive relationship model of TC21 alloy during superplastic deformation based on BP artificial neural network

作　　者：赵文娟[1] 丁桦[1] 周舸[1] 曹富荣[1] 侯红亮[2] 

机构地区：[1]东北大学材料与冶金学院,辽宁沈阳110004 [2]北京航空制造工程研究所,北京100024

出　　处：《锻压技术》2009年第4期138-142,共5页Forging & Stamping Technology

摘　　要：TC21合金是一种高强、高韧、高损伤容限型两相钛合金,具有极佳超塑成形性能。建立合理的超塑性本构关系,对了解该合金的超塑性变形特征以及超塑性成形工艺优化有着重要的指导作用。本文对TC21合金在Gleeble1500热模拟试验机上进行了超塑性等温压缩变形试验。结果表明,随着温度的升高或应变速率的降低,材料的流变应力显著降低,动态再结晶是其主要的软化机制。根据所获得的实验数据,应用BP人工神经网络建立了TC21合金的超塑性本构关系模型,较好地反映了TC21合金的超塑变形过程中流动应力的变化规律。TC21 alloy is a two-phase alloy with high strength, high toughness and high damage-tolerant, and expresses excellent superplasticity. So it is necessary to build an appropriate constitutive relationship for profound study on dynamic response between flow stress and thermal parameters during superplastic deformation, which has important guiding effect on optimization of superplastic processing parameters. In this study, superplastic compression tests were carried out for TC21 alloy by Gleeble1500 stress-strain simulator. Deformation behaviors were analyzed during superplastic compression. The results show that flow stress decreased with the increasing of temperature or decreasing of strain rate during superplastic compression deformation, and that dynamic recrystallization was the main softening mechanism. Then constitutive relationship of superplastic deformation was obtained on the basis of experimental data by BP artificial neural network method, which reflects the relationship between flow stress and its influencing factors satisfactorily.

关 键 词：TC21合金 超塑性 本构关系 BP神经网络 

分 类 号：TG146.23[一般工业技术—材料科学与工程]