QCr0.5的高温变形行为及基于BP神经网络的本构关系预测  

Hot Deformation Behavior of QCr0.5 and Prediction of Constitutive Relationship Based on BP Artificial Neural Network

作　　者：王青[1] 宋克兴[1] 郜建新[1] 国秀花[1] 

机构地区：[1]河南科技大学材料科学与工程学院,河南洛阳471003

出　　处：《热加工工艺》2008年第14期1-3,8,共4页Hot Working Technology

基　　金：河南科技大学校基金资助项目(2002QN01);河南省高校杰出科研人才创新工程项目(2007KYCX008);河南省教育厅科技攻关项目(2007430004);河南科技大学重大科技前期预研专项(2005ZD003);河南科技大学人才科研基金资助项目(of023)

摘　　要：在热模拟试验机(Gleeble1500)上对QCr0.5进行高温压缩实验,获得不同温度、应变和应变速率下材料的流动应力,并建立基于人工神经网络的QCr0.5高温流动应力模型。结果表明,采用BP建立的网络模型能够较精确地反映温度、应变速率、应变和流动应力关系,为优化QCr0.5热变形过程中的工艺参数提供依据。The hot deformation behavior of QCr0.5 was investigated by Gleeble 1500 thermo-simulator, the flow stress at different temperature, different strain and different strain rate were observed, the model of the flow stress during hot deformation was set up by the modified algorithm-Levenberg-Marquardt （L-M） BP neural network. The results indicate that the BP neural network model can accurately describe the influences of the temperature, strain rate and true strain on the flow stress, and also provide basis for optimizing hot deforming process parameters.

关 键 词：QCr0.5 流动应力 BP神经网络 

分 类 号：TG144[一般工业技术—材料科学与工程] TP39[金属学及工艺—金属材料]