论文部分内容阅读
分析了原燃料条件较差、炉况稳定程度较低和检测仪表不多的湘潭钢铁公司２号高炉铁水含硅量数据的统计学特征。按照现代控制论的观点，将高炉视作多输入－单输出系统。当系统处于相对稳定的状态时，可以用若干个操作变量和模型预报偏差的多项式预报下一铁次的含硅量；根据赤池法则确定模型中包括的自变量及其阶数。应用该模型对９１５炉铁次的含硅量作离线预报研究，结果证明了模型的适用性。
The statistical characteristics of the silicon content of hot metal in No.2 blast furnace of Xiangtan Iron and Steel Co. Ltd., which had poor raw fuel conditions, low stability of furnace condition and few instruments, were analyzed. According to modern cybernetics, the blast furnace is considered as a multi-input-single-output system. When the system is in a relatively stable state, the silicon content of the next iron can be predicted by using a number of operational variables and the polynomial of the model prediction bias. The independent variables included in the model and its orders are determined according to the Chi-Chi rule. Applying the model to the offline prediction of silicon content in 915 furnaces, the results prove the applicability of the model.