摘要
Petri网是模型并行系统和分布式系统的一种强有效的形式化工具。它能够很好地刻画系统的动态行为、分析系统的性能。目前研究的大多数随机Petri网假定模型中所有变迁的实施时间呈指数分布,但是在许多实际系统中,变迁的实施时间呈确定性分布或一般性分布,这就需要研究非马尔可夫模型。文章主要讨论非马尔可夫随机Petri网的分析技术,即基于马尔可夫再生理论进行分析求解,并举例进行说明,在文章的最后进行了总结和展望。
Petri-Net is a quite valid formal tool for modeling parallel and distributed systems. It can describe dynamic behavior of systems in detail and evaluate performance of systems. At present, most developed models based on stochastic Petri Nets assume that all the firing times submit to exponential distributions. However, in many real systems, some firing times submit to deterministic or generic distributions. Hence there is a requirement for researching Non-Markovian models. In this paper, we mainly discuss an analysis method of Non-Markovian stochastic Petri Nets, which is based on Markov regenerative theory. After that, we give an example to illustrate this method. At the end of the paper, we summarize this analysis method and open up prospects for this area.
出处
《系统仿真学报》
CAS
CSCD
2003年第z1期71-75,共5页
Journal of System Simulation
基金
国家重点基础研究发展规划(973计划)项目(G1999032707)
国家自然科学基金(90104002和60173012)
国家高技术研究发展计划(863计划)课题(2001AA112080)
高等学校博士学科点专项科研基金项目(20020003027)。