摘要
大规模分布式交互仿真系统的开发与应用对仿真系统设计提出了更高的要求.本文针对分布式仿真系统设计中存在的设计层次不明确的问题,提出了基于商空间粒计算理论的系统分层模型,该模型包括系统总体设计层、系统体系结构设计层和系统执行结构设计层,解决了系统设计的层次问题.在此基础上,对各个层次的纵向和横向信息粒之间的关系进行了深入的分析,并综合考虑了仿真任务、仿真资源和仿真服务的影响.最后根据一个实际的仿真系统设计实例对模型进行了验证,结果表明,利用分布式仿真系统层次模型进行设计可以极大地提高系统的灵活性和可重用性.
To answer the problem of ambiguous design levels for large-scale distributed simulation systems, this paper proposes a hierarchical system model based on the quotient space theory. This model consists of a system global design level, an architecture level, and a execution structure level to solve the problem of diverse levels on system design. Furthermore, we deeply analyze the relationship between the granules of different levels in both horizontal and vertical directions, and integratedly consider simulation tasks, resources and services. Finally, the model is verified based on a real simulation system design example, and the results illustrate that the flexibility and reusability of the system can be improved with the application of the hierarchical design model.
出处
《自动化学报》
EI
CSCD
北大核心
2010年第7期923-930,共8页
Acta Automatica Sinica
基金
国家重点基础研究发展计划(973计划)(97361361)
国家杰出青年科学基金(60925011)资助~~
关键词
系统设计层次模型
商空间
粒计算
分布式仿真系统
Hierarchical system design model
quotient space
granular computation
distributed simulation system