事件图的混合控制器设计  

Combined Controller Synthesis for Marked Graphs

作　　者：罗继亮[1,2] 吴维敏[2] 苏宏业[2] 褚健[2] 

机构地区：[1]华侨大学信息科学与工程学院,泉州362021 [2]工业控制技术国家重点实验室,浙江大学,先进控制研究所,杭州310027

出　　处：《自动化学报》2007年第2期218-221,共4页Acta Automatica Sinica

基　　金：国家创新研究群体科学基金(60421002);浙江省自然科学基金(Y104081);福建省青年人才项目(2006F3087);华侨大学科研基金(06BS212)资助~~

摘　　要：指数级计算复杂性的可达性分析导致较大规模Petri网上的监控器设计非常困难．为了解决这一问题,本文给出了事件图上的混合型控制器设计方法:首先得到了观测器的设计方法;其次给出了根据观测器计算最大允许控制策略的算法．结果表明,该方法降低了需要观测的状态空间的维数,从而提高了计算效率,并且该方法不要求对象网是活的或有界的．The exponential computation complexity of the state reachability analysis makes it difficult to apply forbidden state controller synthesis method to large plant nets. To solve this problem, a combined controller synthesis method is proposed for marked graphs. Firstly, the methods of designing observers are obtained. Secondly, the algorithm of calculating the maximally permissive control policy is proposed using the markings of the observers. The results show that the online computational efficiency of the controller is increased, because the state spaces to be observed is decreased in dimension number. Furthermore, the plant nets are not necessarily alive or bounded.

关 键 词：离散事件系统 监控 PETRI网 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]