摘要
钢包底吹精炼过程中,渣眼行为对钢成分和夹杂物数量的调控有着至关重要的作用。由于生产现场工况复杂、温度较高、成本较大等原因,现场测试存在一定难度和危险,测试结果的精度也无法保证。为准确探究钢包底吹孔位置、单孔底吹气体流量和渣层厚度对渣眼面积的影响,采用1∶5的相似比,以150 t实际工业钢包为原型建立水模型,利用BP神经网络算法拟合实验数据生成模型,对精炼过程中渣眼的演化行为进行预测。分析表明,当隐藏层神经元数目n=16,迭代次数Epoch=60000时,模型损失函数Error值达到最小,决定系数R^(2)=93.439%,模型性能优异,预测精度满足工业需求,可有效地指导工业生产。
The behavior of slag eye during the refining process of bottom blown steel ladle plays a crucial role in regulating the steel composition and the quantity of inclusions.Due to the complex working condition,high temperature,and high costs at the production site,on-site testing poses certain difficulties and dangers,and the accuracy of the test results cannot be guaranteed.In order to accurately investigate the effects of the position of the bottom blown hole,the gas flow rate of a single hole,and the thickness of the slag layer on the slag eye area in the steel ladle refining process,using a 1∶5 similarity ratio,a water model was established using a 150 t actual industrial ladle as a prototype.The BP neural network algorithm was used to fit experimental data and generate a model to predict the evolution behavior of slag holes during the refining process.The analysis shows that when the number of neurons in the hidden layer is 16 and the number of iterations Epoch is 60000,the model loss function Error value reaches its minimum,and the determination coefficient R^(2)is 93.439%,the model demonstrates excellent performance,and its prediction accuracy meets industrial requirements,effectively guiding industrial production.
作者
刘晓航
王肸杰
刘畅
贺铸
李光强
王强
LIU Xiaohang;WANG Xijie;LIU Chang;HE Zhu;LI Guangqiang;WANG Qiang(The State Key Laboratory of Refractories and Metallurgy,Wuhan University of Science and Technology,Wuhan 430o81,Hubei,China;Key Laboratory for Ferrous Metallurgy and Resource Utilization of Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081,Hubei,China;School of Resources and Safety Engineering,Wuhan Institute of Technology,Wuhan 430073,Hubei,China)
出处
《钢铁研究学报》
CAS
CSCD
北大核心
2024年第4期469-480,共12页
Journal of Iron and Steel Research
基金
国家自然科学基金重点资助项目(U22A20173)
湖北省重点研发计划资助项目(2022BAA021,2022BAD043)。
