基于特征模型的预测函数控制的仿真研究

传统的预测函数控制通常采用一阶的预测模型,该预测模型不能完全表征被控对象,因此使得传统的预测函数控制的鲁棒性受到一定限制。特征模型是一种比动力学模型简单,但能表征被控对象特征的模型。针对预测函数控制算法的缺陷,提出了利用二阶特征模型来构成预测模型。通过与采用一阶预测模型的预测函数控制进行比较,理论分析与仿真实验结果均表明,采用特征模型的预测函数控制具有比一阶预测模型更好的控制效果,该算法在SUPCON-JX300X集散控制系统上实现。

Simulation Study of Predictive Functional Control Based on Characteristic Model

The traditional predictive functional control usually uses first-order predictive model, which can not fully express the plant, and it can not get good robust. The characteristic model is simpler than dynamics model but can fully express the plant. Aiming at the weakness of predictive functional control, two order characteristic model for predictive model was used. The results of theoretical analysis and simulation experiment demonstrate that the new control strategy can acquire better effect comparing with the traditional predictive functional control. The control strategy can be directly applied to SUPCON-JX300X.