基于L-M算法BP神经网络的转炉炼钢终点磷含量预报  

Prediction of End-Point Phosphorus Content for BOF Based on LM BP Neural Network

作　　者：李长荣[1,2] 赵浩文[1] 谢祥[3] 尹青[1] 

机构地区：[1]贵州大学材料与冶金学院,贵州贵阳550003 [2]贵州省材料结构与强度重点实验室,贵州贵阳550003 [3]水城钢铁集团炼钢厂,贵州六盘水553028

出　　处：《钢铁》2011年第4期23-25,30,共4页Iron and Steel

基　　金：贵州省科技厅工业攻关项目(黔科合GY字(2008)3062)

摘　　要：转炉炼钢过程是一个非常复杂的物理化学变化过程,人工控制很难一次达到终点目标值,通常需要经过多次补吹才能出钢。通过研究影响转炉冶炼终点磷含量的主要因素,确定了影响转炉终点磷含量的参数,建立了基于Levenberg-Marquardt(LM)算法BP神经网络转炉终点磷含量的预报模型。结果表明:在预报误差目标精度为±0.002%内,命中率达到了90%。BOF steelmaking is a very complex physical chemistry process;it is hard to achieve the target value of end-point by manual control.Multiple reblowing operations were usually necessary to taping off.Based on analyzing the influence major factors of phosphorus end-point in converter,the dominative factors of prediction model of end-point for Conrerter smelting were fixed.A prediction model of end-point phosphorus content for BOF process is established based on Levenberg-Marquardt（LM） algorithm of BP neural network.The results show that the phosphorus content of end-point hitting rates could be reached 90% if the accuracy of target error were ±0.002%.

关 键 词：BP神经网络 终点磷含量 LEVENBERG-MARQUARDT算法 预报模型 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TF703.8[自动化与计算机技术—控制科学与工程]