论文部分内容阅读
分析马氏体转变温度的影响因素,基于RBF神经网络建立马氏体开始转变温度预测模型,对其训练至稳定,预测钢的马氏体开始转变温度。与经验公式计算结果对比,基于RBF神经网络的马氏体开始转变温度预测模型具有较高预测精度。对4种钢的合金元素进行定量分析,结果表明:增加C含量能降低马氏体开始转变温度;马氏体开始转变温度与C、Si、Mn、Cr、Ni和Mo含量一般呈非线性关系。
The influencing factors of martensitic transformation temperature were analyzed. Based on RBF neural network, the prediction model of martensite transformation temperature was established. The training temperature was stable and the martensite transformation temperature of steel was predicted. Compared with the results of empirical formula calculation, the predictive model of martensitic transformation temperature based on RBF neural network has higher prediction accuracy. Quantitative analysis of the alloying elements of the four kinds of steels showed that increasing the C content decreased the martensitic transformation temperature, and the martensitic transformation temperature had a nonlinear relationship with the contents of C, Si, Mn, Cr, Ni and Mo .