Integrated design of optimal supervisors for the enforcement of static and behavioral specifications in Petri net models
Integrated design of optimal supervisors for the enforcement of static and behavioral specifications in Petri net models
复制标题
DOI：
10.1016/j.automatica.2013.08.018
复制
发表时间：
2013-11
期刊：
影响因子：
--
通讯作者：
F. Basile;R. Cordone;L. Piroddi
中科院分区：
文献类型：
--
作者：
F. Basile;R. Cordone;L. Piroddi
Petri网（PN）监督控制通常通过在初始无约束PN模型上引入附加约束层的顺序过程来执行，使用广义互斥约束（GMECs）实现为监视器位置。这是典型的，例如，在柔性制造系统中，初始模型表示生产序列，约束用于表示静态规范，如作业限制或资源使用，以及行为，如活性，可控性等。这种顺序过程可能会产生冗余模型，不易于后验简化。此外，很难确保最大的perception相对于多个行为规范。本文，建立在最近的结果与分支和定界方法的最佳监督设计，提出了一种集成的建模方法，可以用来获得一个最小的监督，保证达到一个最大限度地允许的方式的静态和行为规范的任意一组。在行为规范中，考虑了无死锁性、活性、可逆性和行为可控性。监督者以一组简单的GMEC或多组GMEC的析取的形式出现。一些例子强调了可以实现的潜在的模型尺寸减小。
Petri net (PN) supervisory control is often performed through a sequential procedure that introduces additional constraint layers over an initial unconstrained PN model, using generalized mutual exclusion constraints (GMECs) implemented as monitor places. This is typical,e.g., in the context of flexible manufacturing systems, where the initial model represents the production sequences and the constraints are used to expressstaticspecifications, such as job limitations or the usage of resources, andbehavioralones, as liveness, controllability, etc. This sequential procedure may yield a redundant model, that is not easily reduceda posteriori. Also, it is difficult to ensure maximal permissivity with respect to multiple behavioral specifications. This paper, building on recent results regarding optimal supervisor design with branch & bound methods, proposes an integrated modeling approach that can be used to derive a minimal supervisor guaranteeing the attainment of an arbitrary set of static and behavioral specifications in a maximally permissive way. Among behavioral specifications, deadlock-freeness, liveness, reversibility and behavioral controllability are considered in the paper. The supervisor comes in the form of a simple set of GMECs or of a disjunction of sets of GMECs. Some examples emphasize the potential model size reductions that can be achieved.