Abstract: It is widely recognized that model uncertainty is inevitable. A robust design method of predictive controller parameter based on Min-Max ruler is presented that takes into account model uncertainty. Simulation results show that it enables predictive controller to keep better control performance when the plant dynamic characteristics change and re-design of predictive controller parameters is not needed, which improves system robust performance.

Key words: 预测控制, Min-Max参数设计, 遗传算法

摘要： 工业控制中模型的不确定性是不可避免的.提出基于Min-Max的预测控制器鲁棒参数设计方法，充分考虑到模型的不确定性.仿真结果表明，控制器在对象模型一定范围内变化时仍具有较好的控制品质，不需要重新整定控制器参数，提高了系统的鲁棒性能.

关键词: 预测控制, Min-Max参数设计, 遗传算法