论文部分内容阅读
根据冶炼工艺和现场数据 ,对转炉终点磷含量的预报方法进行了研究。采用自组织神经网络模式识别方法对 30 3炉现场数据进行了分类 ,分析了转炉冶炼各变量对终点磷含量的影响 ,确定了终点磷含量的控制变量 ,建立了基于自适应模糊神经网络系统的转炉终点磷含量的预报控制模型。研究表明 ,本模型能够对终点磷含量进行很好的预报和控制。模型计算值与实际值的相关性达到 0 .5 86 7;磷含量 (质量分数 ,% )控制在± 0 .0 0 3范围内的命中率达到 79.2 1%。该模型以低于目标值 0 .0 0 4 %的磷含量来对冶炼过程进行控制 ,冶炼合格率超过 91%。
According to the smelting process and field data, the method of forecasting phosphorus content at the end of the converter was studied. The field data of 30 3 furnaces were classified by using self-organizing neural network pattern recognition method. The influence of converter smelting variables on end point phosphorus content was analyzed, and the control variable of end point phosphorus content was determined. Based on the adaptive fuzzy neural network system Predictive Control Model for Converter End Phosphorus Content. Studies have shown that this model can predict and control the end-point phosphorus content well. The correlation between the calculated value of the model and the actual value reached 0 .5 86 7; the hit rate of phosphorus content (mass fraction,%) in the range of ± 0 .0 0 reached 79.2 1%. The model controls the smelting process at a phosphorus content lower than the target value of 0.040%, with a passing rate of over 91%.