摘要
LF炉是钢铁冶炼中的重要设备,其主要作用是对钢水中合金成分进行调整,然而目前实际生产中大多仍使用人工经验来对合金成分进行调整,而且已有的合金加料模型效果并不令人满意。为了使添加的合金更加准确、成本进一步降低,设计了一种基于天鹰优化器(AO)来优化Elman神经网络(ENN)的合金收得率预报模型。首先根据相关性分析出对合金元素收得率影响较大的因子,然后利用AO优化后的ENN建立合金元素收得率预报模型,最后通过预测的合金元素收得率来计算所需加入的合金量。利用实际生产中的真实数据来进行仿真试验,仿真结果表明,建立的AO-ENN模型相较于BP模型和Elman模型,误差更小,精度更高,对实际生产中的合金加入问题有较好的指导意义。
LF is an important equipment in iron and steel smelting.Its main purpose is to adjust the alloy composition in steel.However,at present,most of actual production still use manual experience to adjust the alloy composition,and the effect of existing alloy charging model is not satisfactory.In order to make the added alloy more accurate and further reduce the cost,an alloy yield prediction model based on Aquila Optimizer(AO)to optimize Elman neural network(ENN)was designed in this paper.Firstly,the factors that have a great impact on the alloy yield were analyzed according to the correlation.Then,the Elman neural network optimized by Aquila Optimizer was used to establish the alloy yield prediction model.Finally,the amount of alloy to be added was calculated through the predicted alloy yield.The simulation experiment was carried out using the real data in the actual production.The simulation results show that the AO-ENN model established in this paper has smaller error and higher precision than BP model and Elman model,which has certain guiding significance for the alloy addition in the actual production.
作者
易振
柴琳
刘惠康
杨磊
YI Zhen;CHAI Lin;LIU Hui-kang;YANG Lei(School of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan 430081,Hubei,China;Engineering Research Center of Metallurgical Automation and Measurement Technology,Ministry of Education,Wuhan University of Science and Technology,Wuhan 430081,Hubei,China)
出处
《中国冶金》
CAS
北大核心
2022年第5期40-48,57,共10页
China Metallurgy
基金
国家自然科学基金资助项目(51774219)。
