提高分布式多预测控制器协调及稳定性的鲁棒区域控制

大型复杂化工程过程控制中,常规的集中式控制方式不利于实时性、灵活性和容错控制。而采用多预测控制器协调的分布式控制是解决这一问题的有效方法:。针对子系统间的动态耦合行为严重影响多预测控制器协调以及稳定性的问题,提出一种鲁棒区域控制策略。即在每个子系统的目标函数中加入松弛因子增加控制器间协调时的余量来达到分布式预测控制的稳定性。通过以反应器-存储器分馏器组成的过程为事例,仿真结果:表明该方法:的可行性和有效性。

Robust zone control for improving stability performance of distributed multiple model predictive coordination control

Distributed control is an more efficient approach to improve the performances including real-time performance,operation flexibility and fault-tolerant than centralized control strategies for control large scale complex process control.However,coupled dynamic behaviors between different subsystems are the main factors that effected the optimal cooperation of total subsystems,reliability and steady running for total process control system.In this study,Theoretical analysis combining experiment,simulation were used to study optimization coordination of different sub-controller via implement robust zone control strategy in distributed control in case of the emergence of dynamic coupled information change among subsystems.Some more feasible mechanisms were presented for communicating and coordinating among MPC controllers in order to achieve better coordination performance and stability in total process system via setting a optimal region margin index.The feasibility and efficiency of the proposed method was illustrated by a case study,an experimental plant that consists of reactor,storage tank and separator is used.