基于BP神经网络对脱氧合金冶炼中元素收得率的预测  

Prediction of element yield in deoxidation alloy smelting based on BP neural network

作　　者：朱家明[1] 燕惹弟 张航 卢敏欣 ZHU Jiaming;YAN Redi;ZHANG Hang;LU Minxin(School of Big Data,Anhui University of Finance and Economics,Bengbu 233030,China;School of Finance,Anhui University of Finance and Economics,Bengbu 233030,China)

机构地区：[1]安徽财经大学大数据学院,安徽蚌埠233030 [2]安徽财经大学金融学院,安徽蚌埠233030

出　　处：《青海大学学报》2020年第4期70-77,共8页Journal of Qinghai University

基　　金：国家自然科学基金项目(71934001);教育部人文社会科学研究项目(19YJCZH069);安徽省教学研究项目(理学)(2018jyxm1305);安徽财经大学教学研究项目(acxkjsjy201803zd,acjyyb2018006)。

摘　　要：为了准确地预测炼钢生产中脱氧合金化过程中元素收得率,本研究基于BP与RBF神经网络模型对脱氧合金冶炼中元素收得率预测精度进行对比分析。结果表明:(1)根据预测结果精度确定脱氧合金冶炼中元素收得率预测使用BP神经网络模型;(2)基于BP神经网络模型对脱氧合金冶炼中C、Mn两种元素收得率进行预测分析得出:C元素收得率预测区间为[0.8949,0.9012]、Mn元素收得率预测区间为[0.9045,0.9195]。基于BP神经网络模型能够较为准确地预测脱氧合金冶炼中元素收得率区间,从而控制合金用量,达到降低炼钢成本的目的。In order to accurately predict the element yield in the deoxidation alloying process in steelmaking production,this study based on BP and RBF neural network models to compare and analyze the prediction accuracy of the element yield in deoxidation alloy smelting.The results show that:(1)the BP neural network model is used to predict the yield of elements in deoxidized alloy smelting according to the accuracy of the prediction results;(2)the yield of C and Mn elements in deoxidized alloy smelting is predicted based on the BP neural network model.The analysis shows that the prediction interval of the yield rate of element C is[0.8949,0.9012],and the prediction interval of the yield rate of element Mn is[0.9045,0.9195].Based on the BP neural network model,it is possible to more accurately predict the interval of element yield in deoxidized alloy smelting,so as to control the amount of alloy to reduce the cost of steelmaking.

关 键 词：脱氧合金化 BP神经网络 RBF神经网络 元素收得率 预测 

分 类 号：TF746[冶金工程—钢铁冶金]