基于智能算法的C-Mn钢轧制工艺优化与性能预测

Rolling process optimization and performance prediction of C-Mn steel based on intelligent algorithm

为了研究C-Mn钢的化学成分与生产工艺对其力学性能的影响,以某公司生产的实际生产数据为基础,采用BP神经网络和支持向量机对C-Mn钢的力学性能进行了预测。选取C含量、Si含量、Mn含量、粗轧出口温度、精轧出口温度、卷曲温度以及终轧厚度作为模型的输入,选取屈服强度、抗拉强度、伸长率作为模型的输出。对建立的模型引入相对误差分析并对预测精度进行比较。结果表明:BP神经网络模型的屈服强度预测精度为92.6%,相对误差为5.5%;抗拉强度的预测精度为93.9%,相对误差为3.3%;伸长率的预测精度为67.8%,相对误差为10.1%。支持向量机的屈服强度预测精度能达到95.4%,相对误差为4.2%;抗拉强度预测精度为97.9%,相对误差为2.8%;伸长率的预测精度为88.7%,相关误差为6.8%。因此支持向量机模型的预测精度要高于BP神经网络模型。

To study the influence of chemical composition and production process on mechanical properties of C-Mn steel,based on the actual production data of a company,the mechanical properties of C-Mn steel were predicted by BP neural network and support vector machine.C content,Si content,Mn content,roughing exit temperature,finishing exit temperature,crimping temperature and final rolling thickness were selected as the input of the model,and the yield strength,tensile strength and elongation were selected as the output of the model.The relative error analysis was introduced to the established model and the prediction accuracy was compared.The results show that the prediction accuracy of yield strength of BP neural network model is 92.6%and the relative error is 5.5%;the prediction accuracy of tensile strength is 93.9%and the relative error is 3.3%;the prediction accuracy of elongation is 67.8%and the relative error is 10.1%.The yield strength prediction accuracy of support vector machine can reach 95.4%and the relative error is 4.2%;the prediction accuracy of tensile strength is 97.9%and the relative error is 2.8%;the prediction accuracy of elongation is 88.7%and the correlation error is 6.8%.Therefore,the prediction accuracy of support vector machine model is higher than that of BP neural network model.