Many real-world systems involve complex interactions between multiple physical components. Examples include natural systems, such as climate models, as well as engineered systems, such as automobile, aircraft, or rocket engines. Simulation of such systems helps improve our understanding of their function or design, and potentially leads to substantial savings in time, money, and energy. However, simulation of multicomponent systems poses significant challenges in the physical disciplines involved, as well as computational mathematics and software systems.
At the Center for Simulation of Advanced Rockets (CSAR) at the University of Illinois, we have been developing a software system for detailed simulation of solid rocket motors. A rocket motor is a complex system of interactions among various parts--propellant, case, insulation, nozzle, igniter, core flow, etc.--involving a variety of mechanical, thermal, and chemical processes, materials, and phases. Many issues must be addressed in developing such a high-performance system, including
In this paper we describe the software framework developed at CSAR for large-scale integrated rocket simulations. We provide a technical overview of the computational and computer science support for these rocket simulations, and complements the results on integrated simulation presented at this conference.[1,2] While presenting a comprehensive overview of computer science efforts, we focus mostly on our most recent developments on surface propagation and the high-level orchestration framework.
The remainder of the paper is organized as follows. Section II briefly overviews the Rocstar software system developed at CSAR. Section III presents our novel integration framework for multicomponent systems and some middleware services built upon it. Section IV describes a few key computational problems arising from the integration of such systems and our solutions to them. Section V describes the new high-level orchestration framework for the integrated rocket simulations, and Section VI concludes the paper with a discussion of some remaining challenges.