摘要
针对一类输入和输出受约束且具有多胞结构的离散LPV系统,提出一种基于多面体不变集的鲁棒模型预测控制(RMPC)算法.选取一系列收敛于原点的离散状态点,计算每个状态的反馈控制率,构建相应的多面体不变集.在每一个采样时刻,确定包含当前状态的最小多面体不变集,通过计算与相邻两个多面体不变集的位置关系,执行连续的状态反馈控制率.仿真结果表明,相比椭圆不变集离线RMPC算法,所提出算法扩大了系统的稳定区域,取得了保守性较小的结果.
Based on the polyhedral invariant sets, a robust model predictive control algorithm is developed for an input- constrained and output-constrained polytopic linear parameter varying discrete system. A sequence of discrete states converging to the origin is chosen to compute the corresponding state feedback control laws, and also construct each polyhedral invariant sets. At each sampling time, the smallest polyhedral invariant sets that the current measured state can be embedded is determined. The continuous state feedback control laws based on the position of the current measured state between the adjacent polyhedral invariant sets are implemented. Simulation results show that, compared to the ellipsoidal off-line RMPC algorithm, the proposed algorithm yields a substantial expansion of the region stabilized and achieves a less conservative result.
出处
《控制与决策》
EI
CSCD
北大核心
2013年第11期1661-1666,共6页
Control and Decision
基金
国家自然科学基金项目(61132008)
关键词
线性变参数
多面体不变集
鲁棒模型预测控制
椭圆不变集
离线
linear parameter varying
polyhedral invariant sets
robust model predictive control
ellipsoidal invariant sets
off-line
