BOF End-Point Prediction Model Based on Nonlinear Homotopic BI Algorithm

摘要：

　　Aiming at the defects of the neural network itself,such as low speed converging and liable to be trapped in local minimum,a homotopic BI neural network model is developed by combining the homotopy theory and the BI neural network model.While the end-point carbon content and temperature of molten steel is predicted by the new model,the precision of new model is improved accordingly.The hit rates are increased by about 5% and 10%,and the forecasting residuals have decreased 16.31% and 8.67% than the conventional ones respectively.Also,the calculation time of the new model is10% shorter than BI model.

关键词： BOF model end-point prediction neural network homotopy algorithm

作者: Feng Mingxia Chen Ren Li Qiang Zou Zongshu

作者单位: Metallurgy Engineering College, Liaoning Institute of Science and Technology, 117004 Benxi, China School of Materials and Metallurgy, Northeastern University, Shenyang 110819, Liaoning, China

会议类型: 国际会议

会议名称: The 4th Australia-China-Japan Joint Symposium on Iron and Steelmaking(第四届中日澳钢铁冶金学术会议)

会议地点: 沈阳

会议语种:英文

页码: 295-301

在线出版日期: 2012-11-03（万方平台首次上网日期，不代表论文的发表时间）