摘要
以铸态904L超级奥氏体不锈钢热压缩实验数据为样本,采用BP人工神经网络建立了以变形温度、应变速率和应变量为输入层,流变应力为输出层的热变形本构模型。结果表明,BP网络预报的流变应力值与实验值比较吻合,平均绝对误差值为2.06%。该模型能够准确反映铸态904L超级奥氏体不锈钢的高温流变行为,为合理控制热轧工艺参数提供一定参考。
Hot deformation constitutive model was established using BP neural network method based on the data obtained from hot compression tests for as-cast 904L superaustenitic stainless steel,taking deformation temperature(T),strain(ε) and strain rate(σ) as input variable,while the flow stress(σ) as output variable.The results show that the predicted flow stresses obtained from BP network agree well with the experimental values,and the absolute average error is 2.06%.It is confirmed that the proposed model possesses excellent capability to predict the high temperature flow behavior of as-cast 904L superaustenitic stainless steel,which can provide some reference for optimizing hot rolling parameters.
出处
《铸造技术》
CAS
北大核心
2012年第10期1162-1164,共3页
Foundry Technology
基金
国家火炬计划资助项目(2010GH030146)
山西省科技攻关项目(20120321015-01)
关键词
904L奥氏体不锈钢
人工神经网络
流变应力
预测
904L austenitic stainless steel
artificial neural network
flow stress
prediction