随机进程代数模型的Fluid逼近技术能够有效缓解状态空间爆炸问题, 在性能评估领域受到广泛关注. 本项目以Bio-PEPA为代表, 研究一类随机进程代数模型的Fluid逼近问题, 这类随机进程代数能够描述系统的Mass-action和Michaelis-Ment等动力学性质. 本项目研究分为三个部分：一是利用Markov链的Possion过程表示方法来研究Fluid逼近和模型蕴含的Markov链之间的内在关联, 突破了现有研究方法中关于Markov链必须具有依赖密度特性的限制；二是利用这种内在关联以及模型的结构性质来研究Fluid逼近的基本性质，特别是所导出的微分方程的解的收敛性；三是研究怎样利用随机进程代数模型Fluid逼近来提取性能指标, 如利用Fluid逼近来改进随机模拟来提取性能指标等. 这些研究将进一步拓展随机进程代数的应用, 并为这些应用奠定理论基础.

Fluid approximation of stochastic process algebra models as a novel technique proposed to cope with the state-space explosion problem has attracted lots of attentions in the field of performance evaluation. This research proposal deals with the fluid approximation of a class of stochastic process algebras such as Bio-PEPA, which can describe Mass-action and Michaelis-Ment kinetics. This proposal has three parts. The first is the investigation of the relation between the fluid approximation and the Markov chain underling a model, which is based on the Possion process representation for Markov chains, without assuming the Markov chain to have the property of density dependent. Secondly, the relationship between the fluid approximation and the Markov chain, as well as the structure characteristics of a model, will be used to establish fundamental properties of fluid approximation, including the convergence property of the derived differential equations. Thirdly, deriving performance measures through fluid approximation will be studied. Particularly, performance metrics are expected to be obtained by stochastic simulation which is enhanced by fluid approximation. These researches will expand the application of stochastic process algebras, as well as provide theoretical foundation for the application.