基于部分状态空间存储的Petri网库所界求解算法

Place Bound Algorithm of Petri Net Based on Partial State Space Storage

Petri网是一种重要的形式化建模工具,有界性是Petri网的重要性质之一.该文关注于这一性质并提出了一种新的无需存储全部状态空间的算法以求解Petri网库所界,主要思想为在生成状态空间过程中,通过引导消除可达图的部分回路同时结合T不变量的相关性质,以实现通过存储部分状态来精确求解每个库所的界.基于模型检测比赛的公开数据集进行了对比实验,通过对求解库所界的不同方法及其实验结果进行比较分析,说明了本文算法的有效性.

Petri net is an important formal modeling tool,and boundedness is one of the important properties of Petri net.In this paper,we focus on this property and propose a new algorithm to solve the place bounds of Petri net without storing all state spaces.The main idea is that in the process of generating state spaces,we can accurately solve the bounds of all places by storing partial states through eliminating some loops of reachability graph and taking advantages of related properties of T-invariant.Experimentally,we compare several other methods with the proposed one in their capability of solving place bounds with the open data set of model checking contest,and results show the effectiveness of the proposed algorithm.