论文部分内容阅读
针对高炉炉温与铁水硅含量呈正相关而非严格的线性关系和机制建模的主观性以及其难以建立各变量之间隐含的数学关系等的不足,在数据挖掘理论的基础上,对海量的样本数据进行预处理和特征提取,然后以高炉铁水温度为研究对象,建立了基于T-S模糊神经网络的高炉铁水温度预测模型。最后,应用某高炉数据进行模型验证,并将该模型与T-S模糊多元回归模型以及BP神经网络模型进行比较研究,仿真结果表明T-S模糊神经网络模型的有效性和优越性。
Aiming at the deficiency of the linear relationship between the blast furnace temperature and the molten silicon content and the subjectivity of the mechanism modeling and the difficulty of establishing the implied mathematical relation among the variables, based on the data mining theory, The pretreatment and feature extraction of the sample data are carried out. Then, the blast furnace hot metal temperature prediction model based on TS fuzzy neural network is established. Finally, a blast furnace data is used to verify the model. The model is compared with the T-S fuzzy multiple regression model and the BP neural network model. The simulation results show the effectiveness and superiority of the T-S fuzzy neural network model.