基于动态优化模型集的多模型自适应控制

针对多模型自适应控制的传统算法中固定模型集无法准确覆盖对象的不确定域的问题，提出了一种新型的能够在线动态优化模型集的方法。该方法能够在线自动添加、删除以及修改模型集中的模型，有效地控制模型数量以及优化模型集，使模型集对不确定域进行精确覆盖，并且在系统运行的稳定阶段可以停止优化模型集的动作，从而大幅减少计算量。该方法在优化控制性能的同时不会过分增加系统运算负担，仿真实验表明了此算法的优越性。

Multiple Model Adaptive Control Based on Dynamically Optimizing Model Bank

A new multiple model adaptive control scheme based on dynamically optimizing model bank is introduced. Focus on the problem that fixed model bank can not cover the plant uncertainty exactly in traditional multiple model adaptive control strategy. This scheme can add, remove and update models online automatically in model bank. So it can control the number of models effectively, and optimize model bank to cover the plant uncertainty exactly. This scheme can stop optimizing model bank when system is running steadily which reduces computational load. Therefore, this scheme improves control performance and reduces the computational load at the same time. Simulation results demonstrate the effectiveness of this method.