摘要
考虑到钢渣化学成分的波动性,为解决钢渣重构在配料时所涉及的复杂运算问题,以钢渣、生石灰、粉煤灰、氟化钙(CaF2)的化学成分作为输入变量,活性指数作为输出变量,构建广义回归神经网络(GRNN)模型,并采用模拟退火算法(SA)进行优化计算,建立基于GRNN-SA的重构钢渣最佳配方优化模型。通过实证分析得出,该模型能够实现重构配料过程的智能化计算,而且具有普适性,对不同来源的钢渣均有指导意义,对钢渣重构的试验结果也有一定的预测效果,实际活性指数值与预测值相对误差在5%以下,模拟准确性较高。
In order to solve the complex operation involved in steel slag reconstruction caused by the fluctuation of chemical compositions of steel slag,the general regression neural network(GRNN)model was constructed,taking the chemical compositions of steel slag,quicklime,fly ash and calcium fluoride(CaF2)as input variables and activity index as output variables,respectively.And the formula optimization model of reconstructed steel slag based on GRNN-SA was established by using simulated annealing algorithm(SA)to optimize the calculation.Through empirical analysis,it is concluded that the model can realize the intelligent calculation of reconstruction batching process.This model is universal and can guide the intelligent calculation for steel slag from different sources.And it can predict the experimental results of steel slag reconstruction.The relative error between the actual activity index value and the predicted value is less than 5%,having a high simulation accuracy.
作者
许莹
杨姗姗
王巧玲
Xu Ying;Yang Shanshan;Wang Qiaoling(College of Materials Science and Engineering,North China University of Science and Technology,Tangshan 063210,Hebei,China;College of Metallurgy and Energy,North China University of Science and Technology,Tangshan 063210,Hebei,China)
出处
《钢铁钒钛》
CAS
北大核心
2020年第1期75-81,94,共8页
Iron Steel Vanadium Titanium
基金
国家自然科学基金资助项目(51574109)。
关键词
重构钢渣
配料
广义回归神经网络
模拟退火算法
reconstructed steel slag
formula optimization
general regression neural network
simulated annealing algorithm
