在分析焦炉立火道温度特性的基础上,提出了一种基于线性回归(LR)和监督式分布神经网络(SDNN)的立火道温度智能集成软测量方法。通过特性分析提出了典型蓄热室的选取原则,并利用级数验证了选取的有效性。为反映蓄顶温度与立火道温度的关系,首先分别建立了一元、二元和十二元LR模型,并通过智能集成将3个模型的输出进行有机融合;然后在对样本监督式聚类后,利用SDNN获得各个子网的综合输出;最后由专家协调器协调LR和SDNN的输出,得到立火道温度的软测量值。实际运行结果验证了该方法的有效性。
An integrated model combining linear regress (LR) and supervised distributed neural networks (SDNN) based on the features of coke oven flue temperature is proposed. Progression is used to analyze the properties of flue temperatures, and rules of selecting typical regenerators are proposed. LR models with one variable, two variables and twelve variables are built and rationally integrated to map the linear relationship between flue temperature and top of regenerators' temperature. At the same time, after supervised clustering the samples, SDNN models are employed to synthesize the outputs of every sub-network. The flue temperature is obtainted through the expert coordinator which is used to coordinate the outputs of LR and SDNN. The running results of the models validate the method.