摘要
针对非线性CSTR(continuously stirred tank reactor)过程,提出一种新的预测控制的设计与仿真实现。在对一类特殊非线性过程分析的基础上,从系统的输入输出数据出发,基于子空间辨识算法建立双线性系统模型来近似描述被控系统;设计新的预测控制算法实现对CSTR过程的跟踪控制;为补偿模型失配以消除控制中的稳态误差,将积分作用包含在预测控制器的设计中,实现对控制输出的良好跟踪性能;最后通过一个仿真实例验证算法的有效性。
A data-driven model predictive control （MPC） to nonlinear continuously stirred tank reactor （CSTR） system is presented. Based on the analysis of a class of nonlinear systems, the controlled model is identified through the bilinear subspace identification. Applying the identified bilinear predictor model, the bilinear MPC is implemented through exploiting the structural properties of the bilinear subspace predictor model. The integral action is included in the controller design to eliminate the state error. These improvements in the system modeling and optimization make the hilinear MPC approach good tracking performance. The simulation result shows the effectiveness of the proposed control approach.
出处
《控制工程》
CSCD
北大核心
2009年第2期145-147,共3页
Control Engineering of China
基金
国家科技支撑计划基金资助项目(2006BAF01A02)
国家自然科学基金资助项目(60774015)
