一种紧同步随机Petri网模型的性能分析方法  

Performance Analysis of Strictly Synchronized Stochastic Petri Nets

作　　者：刘鸿[1] 林闯[1] 吴建平[1] 

机构地区：[1]清华大学计算机科学与技术系,北京100084

出　　处：《软件学报》2003年第6期1029-1036,共8页Journal of Software

基　　金：国家重点基础研究发展规划(973);国家自然科学基金;国家高技术研究发展计划~~

摘　　要：紧同步随机Petri网模型常用于对多个子系统构成的实际系统进行建模.由于状态空间爆炸问题,这种模型的性能分析目前还没有得到很好的解决.首先给出了随机Petri网中顺序、并行、循环、选择这4种基本结构的性能近似等价公式,然后提出了一种同步变迁等价分解(synchronous transition equivalent decomposition,简称STED)法对紧同步随机Petri网进行分解求解,有效地缓解了具有一个或多个同步变迁情况下的空间爆炸问题,该方法对于子模型同构的情况处理简便且十分有效.Strictly synchronized SPNs (stochastic Petri nets) are very popular in stochastic Petri nets made up of sub-models. Because of the problem of state space's explosion, performance analysis of strictly synchronized SPNs has not been solved completely. Formulas that show performance equivalence of four elementary subnets, which are sequence subnet, parallel subnet, choice subnet and iteration subnet, are presented. A decomposition approach called STED (synchronous transition equivalent decomposition) is proposed and can be applied to SPNs with one or more synchronous transitions, especially such models with isomorphic sub-models.

关 键 词：随机PETRI网 性能等价 性能分析 紧同步模型 同步变迁等价分解法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]