基于多层概率集的随机系统预测控制  

Predictive control for stochastic systems based on multi-layer probabilistic sets

作　　者：梁华清[1] 张冬雯[2] 邢少光 张沙沙[2] 

机构地区：[1]河北科技大学电气工程学院,河北石家庄050018 [2]河北科技大学信息科学与工程学院,河北石家庄050018

出　　处：《河北科技大学学报》2016年第2期205-212,共8页Journal of Hebei University of Science and Technology

基　　金：河北省自然科学基金(F2014208169)

摘　　要：针对具有Markov跳变特点的一类离散随机系统,研究了输入量概率约束下的状态反馈预测控制问题。采用多层概率集的概念和方法,给出了具有多个不同概率软约束下的预测控制器设计算法,在多步反馈律的控制下,系统状态以指定概率进入不同的椭圆内,保证了系统的稳定性,而且扩大了控制问题的可行范围,改善了系统性能。最后仿真实例证明了所提方法的有效性。Aiming at a class of discrete-time stochastic systems with Markov jump features,the state-feedback predictive control problem under probabilistic constraints of input variables is researched.On the basis of the concept and method of the multi-layer probabilistic sets,the predictive controller design algorithm with the soft constraints of different probabilities is presented.Under the control of the multi-step feedback laws,the system state moves to different ellipses with specified probabilities.The stability of the system is guaranteed,the feasible region of the control problem is enlarged,and the system performance is improved.Finally,a simulation example is given to prove the effectiveness of the proposed method.

关 键 词：随机过程 预测控制 MARKOV跳变 概率约束 多层概率集 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]