(1. 东北大学自动化研究中心，沈阳110006 2. 宝山钢铁公司自动化研究所，上海201900)
摘 要: 转炉炼钢终点温度和成分是转炉炼钢的控制目标，它与吹氧量、铁水加入量等多个变量之间存在着严重的非线性关系，且无法在线连续测量。作者提出了基于RBF神经网络的转炉炼钢终点温度及碳含量预报模型，并结合某钢铁企业一座180 t 转炉的实际数据进行模型验证研究。结果表明，该方法收敛速度快，预报精度高。
关键字: 转炉；炼钢；预报；神经网络
（1. Research Center of Automation, Northeastern University, Shenyang 110006, P.R.China
2. Automatic Institute, Baoshan Iron and Steel Company, Shanghai 201900, P.R.China）
Abstract:The endpoint temperature and carbon content of basic oxygen furnace (BOF) are the control objects of the BOF steelmaking process. There exists serious non-linearity among them and blowing oxygen etc, and the online continue measurement can not be made. The predictive model of endpoint temperature and carbon content of the BOF steelmaking based on RBF neural network was put forward, and the research of verifying the model was made by comparing the predictive value with the practical data of an 180t converter in a factory. The results show that the method has fast convergence speed and accurate prediction.
Key words: BOF; steelmaking; prediction; neural network
