US-Germany Cooperative Research: Analysis of Multi-Paradigm Moebius Models using Kronecker-Based Techniques

0233490桑德斯该奖项支持初级教员威廉·桑德斯和伊利诺伊-城市-香槟大学的两名学生，他们与德国多特蒙德大学信息科学系的彼得·肯珀合作。该项目将侧重于在Moebius框架内使用Kronecker表示法和分析方法。Moebius框架是由伊利诺伊大学开发的，通过支持多种不同的模型求解方法以及多种建模形式的模型规范，使大型系统模型的验证成为可能。由德国小组研究的Kronecker表示，通过利用复杂模型通常由子模型组合而成的事实，为处理用于功能和定量分析的大型模型提供了一种数学方法，并彻底改变了基于状态的模型的表示和求解方式。联合研究将为模型合成提供一种形式语义学，这将允许研究Kronecker技术在新的模型合成形式体系中的使用，并将使使用与Moebius工具已经支持的其他方法一起开发的Kronecker方法成为可能。

0233490SandersThis award supports William Sanders, a junior faculty member, and two students from the University of Illinois-Urban-Champaign in a collaboration with Peter Kemper of the Department of Information Science at the University of Dortmund, Germany. The project will focus on the use of Kronecker representations and analysis methods within the Moebius framework. The Moebius framework was developed at the University of Illinois and makes validation of large systems models possible by supporting many different model solution methods as well as model specification in multiple modeling formalisms. Kronecker representations, studied by the German group, area a mathematical way to handle large models for functional and quantitative analysis by exploiting the fact that complex models are often built by composition of submodels, and have revolutionized the way state-based models are represented and solved. The joint research will provide a formal semantics for model composition that will allow for investigating the use of Kronecker techniques for new model composition formalisms, and will make it possible to use Kronecker methods developed with other methods already supported by the Moebius tool.