工业系统中广泛存在一类由多个相互关联的子系统组成的大系统.尽管分布式控制结构的性能没有集中式控制好,但由于其具有较高的灵活性和容错性,相对于集中控制更加适合控制上述系统.在保持容错性的情况下如何提高系统的整体性能是分布式控制的一个难点问题.本文提出了一种分布式预测控制(Distributed model predictive control,DMPCI方法,该方法通过在各子系统预测控制器的性能指标中加入输入变量对其下游子系统的影响的二次函数,来扩大分布式预测控制的协调度,进而在不增加网络连通度,不改变系统容错性的前提下,提高系统的性能.另外,本文给出了基于该协调策略的带输入约束的分布式预测控制器的设计方法,在初始可行的前提下,该方法相继可行并可保证系统渐近稳定.
For the control of a class of large-scale systems, which consists of many individual subsystems, the distributed contort framework is more suitable because of its good flexibility and error tolerant, despite the fact that its performance is not as good as that of centralized control. How to improve the performance of the entire system and, at the same time, maintain the flexibility and error tolerance characterises under the distributed framework is an important problem. In this paper, a novel distributed model predictive control (DMPC) is proposed, which enlarges the coordinate degree of DMPC through adding a quasi-function of the impact of control input to its downstream subsystems to the performance index of each subsystem-based model predictive control (MPC). Then, the performance of the entire system is improved with unchanged network connectivity. In addition, a design method with this coordination strategy and inputs constraints is provided, which guarantees the recursive feasiblity of the designed DMPC and asymptotic stablility of the closed-loop system.