ASP热轧过程X70管线钢的组织性能预测模拟  

Prediction of microstructure and properties of X70 pipeline steel in process of ASP hot rolling

作　　者：赵彦峰[1] 许云波[1] 孙卫华[2] 白彦[2] 陈庆军[2] 张殿华[1] 

机构地区：[1]东北大学轧制技术及连轧自动化国家重点实验室,辽宁沈阳110004 [2]济南钢铁股份有限公司技术中心,山东济南250101

出　　处：《材料热处理学报》2011年第1期144-149,共6页Transactions of Materials and Heat Treatment

基　　金：国家"十一五"科技支撑计划(2007BAE51B07);国家重大基础研究发展规划项目(2006CB605208)

摘　　要：采用物理冶金模型结合二维温度场对ASP(Angang Strip Production)热轧X70管线钢再结晶、相变等物理冶金过程进行了模拟,并结合BP神经网络对最终的力学性能进行了预测。研究表明,实验钢在层流冷却前的奥氏体晶粒尺寸为10～25μm,板带横断面奥氏体晶粒尺寸分布不均匀,心部的奥氏体晶粒尺寸比角部大15μm左右;在给定冷却速率的情况下采用前段冷却方式得到的铁素体分数比后段冷却方式大2%～5%;采用BP神经网络可以把伸长率预测结果相对误差标准差提高1.8%;Si含量0.2%～0.3%成为其对力学性能影响的转折点。Based on physical metallurgy model and two-dimensional temperature field,recrystallization and phase transformation of X70 pipeline steel were simulated during ASP（Angang Strip Production） hot rolling.Mechanical properties were predicted by BP neural network.The results show that austenite grain size of the experimental steel is refined to 10-25 μm before laminar cooling section,but it is unevenly distributed along cross-section of strip.Austenite grain size at the core is about 15 μm larger than that at the corner.For a given cooling rate,fraction of ferrite obtained by preceding cooling method is 2%-5% greater than that by back-cooling method.The standard deviation of predicted elongation error can be raised by 1.8% by BP neural network.It is a critical turning point for the effect on mechanical properties of the steel when its Si content is 0.2% ～ 0.3%.

关 键 词：再结晶 相变 BP神经网络 模拟 预测 

分 类 号：TG111.7[金属学及工艺—物理冶金] TG142.13[金属学及工艺—金属学]