Microstructure of Ti-46.5Al-2.5V-1.0Cr-0.3Ni Alloy at Elevated Temperatures Predicted by BP Artificial Neural Network Model
摘 要
采用高温热压缩试验了Ti-46.5Al-2.5V-1.0Cr-0.3Ni合金在温度1 000~1 200 ℃、应变速率0.001~1.0 s-1及应变量0%~70%内的热变形显微组织特征,分析了试验参数对变形组织的影响;在此基础上,建立了该合金高温变形过程组织演化的BP人工神经网络预报模型,并对其预测结果的准确性进行了测试.结果表明:试验过程中随着应变量的加大、温度的升高和应变速率的降低,动态再结晶形成的等轴晶粒增多,残余层片团减少;所建立的预报模型能够较为精确地预测该合金在高温变形过程中的组织变化,对于动态再结晶体积分数的预测值和试验值之间的误差仅为2.82%,对于动态再结晶晶粒尺寸,误差值仅为0.22 μm.
Abstract
The microstructure characteristics of Ti-46.5Al-2.5V-1.0Cr-0.3Ni alloy were studied by high temperature compression tests at temperatures from 1 000 ℃ to 1 200 ℃,strain rates from 0.001 s-1 to 1.0 s-1 and strains from 0% to 70%.The influences of test parameters on the deformed microstructure were analyzed.Based on the tests,BP artificial neural network model of the microstructure evolution of the alloy during high temperature deformation was established,and the accuracy of the predicted results was tested.The results show that during high temperature deformation with the increase of strain and temperature and the decrease of strain rate,the equiaxed grains formed throngh dynamic recrystallizalion increased and the residual lamellar colony decreased.The model for deformation microstructure at elevated temperatures could accurately predict the microstructure changs of the alloy during the deformation at elevated temperatures.The average error between predicted values and measured values was 2.82% for the volume fraction,and 0.22 μm for the grain size of dynamic recrystallization.
中图分类号 TG316
所属栏目 物理模拟与数值模拟
基金项目 国家“863”高技术研究发展计划资助项目(2006AA03A204)
收稿日期 2009/5/6
修改稿日期 2009/12/15
网络出版日期
作者单位点击查看
备注司家勇(1978-),男,安徽巢湖人,博士研究生.
引用该论文: SI Jia-yong,HAN Peng-biao,FU Ming-jie,ZHANG Ji. Microstructure of Ti-46.5Al-2.5V-1.0Cr-0.3Ni Alloy at Elevated Temperatures Predicted by BP Artificial Neural Network Model[J]. Materials for mechancial engineering, 2010, 34(6): 92~96
司家勇,韩鹏彪,付明杰,张继. 用BP人工神经网络模型预测Ti-46.5Al-2.5V-1.0Cr-0.3Ni合金的高温变形组织[J]. 机械工程材料, 2010, 34(6): 92~96
共有人对该论文发表了看法,其中:
人认为该论文很差
人认为该论文较差
人认为该论文一般
人认为该论文较好
人认为该论文很好
参考文献
【1】SEMIATIN S L,SEETHARAMAN V,JAIN V K.Microstructure development during conventional and isothermal hot forging of a near-gamma titanium aluminide[J].Metall & Mater Trans A,1994,25(12):2753-2768.
【2】LIU C T,SCHNEIBEL J H,MAZIASZ P J.Tensile properties and fracture toughness of TiAl alloys with controlled microstructures[J].Intermetallics,1996(4):429-440.
【3】KIM Y W.Strength and ductility in TiAl alloys[J].Intermetallics,1998(6):623-628.
【4】李士勇.模糊控制、神经控制和智能控制论[M].哈尔滨:哈尔滨工业大学出版社,1995:72-99.
【5】高永生,张鹏,崔军,等.应用人工神经网络预测50CrV4钢的变形抗力[J].钢铁,1998(4):27-30.
【6】DOBRZANSKI L A,SITEK W.Application of a neural network in modelling of hardenability of constructional steel[J].J Mater Process Technol,1998,78:59-55.
【7】夏善木,李大勇.用于铸铁力学性能快速预测的人工神经网络[J].应用科学学报,2002,20(3):309-312.
【8】飞思科技产品研发中心.神经网络理论与Matlab7实现[M].北京:电子工业出版社,2006:259-297.
【2】LIU C T,SCHNEIBEL J H,MAZIASZ P J.Tensile properties and fracture toughness of TiAl alloys with controlled microstructures[J].Intermetallics,1996(4):429-440.
【3】KIM Y W.Strength and ductility in TiAl alloys[J].Intermetallics,1998(6):623-628.
【4】李士勇.模糊控制、神经控制和智能控制论[M].哈尔滨:哈尔滨工业大学出版社,1995:72-99.
【5】高永生,张鹏,崔军,等.应用人工神经网络预测50CrV4钢的变形抗力[J].钢铁,1998(4):27-30.
【6】DOBRZANSKI L A,SITEK W.Application of a neural network in modelling of hardenability of constructional steel[J].J Mater Process Technol,1998,78:59-55.
【7】夏善木,李大勇.用于铸铁力学性能快速预测的人工神经网络[J].应用科学学报,2002,20(3):309-312.
【8】飞思科技产品研发中心.神经网络理论与Matlab7实现[M].北京:电子工业出版社,2006:259-297.
相关信息