Abstract: Thermal simulation test for 0. 045Ti-0. 003 C Ti-IF steel with strain 0 ~ 0. 7 and strain rate 1 ~ 70 s^-1 at 700 ~950℃ has been carried out by thermecmastor-Z simulation mat'hine to analyze the effect of temperature, strain and strain rate on deformation resistance. The prediction model of tieformation resistance of the Ti-IF steel at 800 - 875℃ was obtained by BP neural net calculation method. With comparison between model predicted value and actual measured value, it indicated that the relative error of deformation resistance was less than 4. 0% 。

Key words: Ti-IF Steel, Ferrite Rolling, Deformation Resistance, BP Neural Net, Math Model

摘要： 在Thermecmastor-Z热模拟试验机上对0.045%Ti-0.003%C的Ti-IF钢进行700～950℃、应变0～0.7、应变速率1～70s^-1条件下的热模拟实验,以分析温度、应变速率和应变量对变形抗力的影响。用BP(back propagation)网络神经算法给出了Ti-IF钢在800～875℃变形抗力预报模型。通过对模型预报值和实验数据实测值的比较得出,变形抗力相对误差在4.0%以内。

关键词: Ti-IF钢, 铁素体轧制, 变形抗力, BP神经网络, 数学模型