Design architecture of a data driven environment for multiphysics applications
AuthorMichopoulos, J.; Tsompanopoulou, P.; Houstis, E.; Rice, J.; Farhat, C.; Lesoinne, M.; Lechenault, F.
The design architecture of a multidisciplinary problem-solving environment (MPSE) for supporting an efficient prediction capability for the response of continuous interacting systems under multiphysics conditions is presented. The system will be referred to as a Data Driven Environment for Multiphysics Applications (DDEMA) and will be primarily differentiated from previous MPSE efforts on its usage of data for improved confidence of simulation prediction.. Its architecture takes into consideration information technologies, coupled multiphysics sciences, and data-driven practices to achieve steering of adaptive modeling and simulation of the underlying systemic behavior. Special emphasis is given on middleware implementation issues based on actor-agent abstractions. A description of the design objectives and architectural variations of DDEMA are also given in the context of two multidisciplinary applications related to material/structure design of supersonic platforms and fire/material/environment interaction monitoring, assessment and management. Validation of the architecture will also be attempted in terms of the same two applications.