气体分馏装置多变量预测函数控制方法

Multivariable Predictive Functional Control for a Gas Fractionation Plant

针对气体分馏工业过程中存在的多变量、强耦合、大滞后等系统特性,设计了一种预测时域可调的多变量预测函数控制算法。该方法结合单变量预测函数控制及Simith预估器的特点,通过对预测时域进行分段设定,以期达到较好的快速性和稳定性。以气体分馏装置的脱丙烷塔为例,依托现场的实测模型,对提出的算法进行了验证。结果表明,所提出预测时域可调的多变量预测函数控制算法调节时间短,跟踪效果好。

To deal with the problem of multi variable,strong coupling characteristic and long time delay in a gas fractionation plant,multivariable predictive functional control algorithm optimizing time domain is proposed.This algorithm combines the scalar predictive functional control and the Smith predictor.In order to improve the rapidity and stability of the system,the optimizing time domain is divided to many sections with different values.The proposed algorithm is applied to the gas fractionation plant based on the actual measurement model.The simulation results reveal that the proposed algorithm has a short regulation time and can guarantee rapid tracking.