Complex System Engineering Simulation through Distributed Cosimulations

Sylvain Pagerit, P. Sharer, A. Rousseau

Argonne National Laboratory USA

Thierry Roudier, David Balland

Kiastek France

M. Brunelli

BSim Italy

 

Many of today’s advanced simulation tools are suitable for modeling specific systems; however, they provide rather limited support for model building and management. The diverse tools to model different components of a vehicle makes it all the more challenging to apprehend their integration and interactions and analyze the complete system. In addition, the complexity and size of such models require a better use of computing resources such as multi-core or computer cluster to greatly reduce the simulation time. In this paper, we describe how modern software techniques can be used to support modeling and design activities, with the objective of providing system models quickly by assembling these system models in a “plug-and-play” architecture, while also providing control over the models computing load distribution between multiple cores and machines. We show that such an integration and management is achievable by integrating the functionalities of Argonne National Laboratory’s Autonomie modeling tool and the ones from Kiastek’s CosiMate. System models can be integrated, simulated using co-simulation and re-used regardless of the environment in which they are developed, and their simulation results can be consolidated for analysis into a single tool. We will demonstrate these functionalities through the example of a Simulink vehicle model communicating with detailed sub-models in their expert tools such as GT Power, AMESim, Saber or CarSim on separate core and machines. This example will also show the different options available for the interface and how each of these options can be implemented.

Back