基于分级递阶结构的铁矿石烧结过程智能控制

Intelligent control of iron ore sintering process based on a hierarchical configuration

针对铁矿石烧结过程具有强非线性、大滞后、强耦合性的特点,提出一种基于分级递阶结构的烧结过程智能控制方法,建立具有智能控制层和基础自动化层的递阶控制系统。在智能控制层,首先融合灰色理论与BP神经网络方法建立烧结终点预测模型;然后建立烧结终点混杂模糊-预测控制模型;最后提出基于满意度的协调控制方法,实现烧结终点与混合料料槽料位之间的智能协调。基础自动化层实现对操作参数的稳定化跟踪控制。实际运行结果表明,该系统有效地抑制了烧结终点的波动,提高了烧结矿产量质量。

Sintering process features strong nonlinearity,large time-delay and strong coupling.This paper presents an intelligent control method based on a hierarchical configuration and designs a hierarchical control system with an intelligent control layer and a basic automation layer for iron ore sintering process.At intelligent control layer,firstly, a prediction model of burn-through point（BTP） is established using grey theory and back-propagation neural network.Then,a hybrid fuzzy-predictive controller for BTP is devised.Finally,an intelligent coordinating control method based on satisfactory solution principle is devised to implement the coordination of BTP and bunker-level. The basic automation layer provides tracking control of variables of process operations.Results of actual operations show that the system not only sufficiently suppresses the fluctuation of BTP,but also increases the quantity and quality of the sinters.