ECCOMAS 2024

Realizing 3rd Generation Collaborative MDO in the open-source framework KratosMultiphysics

  • Antonau, Ihar (Technical University of Braunschweig)
  • Baars, Susanna (Technical University of Braunschweig)
  • Baimuratov, Ildar (Leibniz University Hannover)
  • Wittenborg, Tim (Leibniz University Hannover)
  • Wüchner, Roland (Technical University of Braunschweig)
  • Römer, Ulrich (Technical University of Braunschweigge)

Please login to view abstract download link

The Cluster of Excellence SE 2 A – EXC 2163 is an interdisciplinary research center to investigate technologies for a sustainable and eco-friendly air transport system. One of the studied technologies in the cluster is the application of skin heat exchangers to long-range non-conventional aircraft with electric propulsion systems powered by fuel cells. This challenging task needs a profound software basis that can handle the complexity of coupling very diverse data and enable the synchronization of multiple processes. In this work, we base the new developments on the KratosMultiphysics platform [1], which is especially suited for this: it is open-source, it allows the implementation of novel concepts, and moreover it is from the beginning, consequently designed to support multi-disciplinary simulation scenarios. The CoSimulationApplication provides the base interface to communicate with remote-controlled external physical solvers [2], which is crucial to couple various tools from different research teams. The central-model architecture is applied to design MDO workflow, where all external tools and solvers push and pull required data from Kratos. It allows to reduce the implementation overhead and simplifies the data synchronization compared to distributed data exchange, where individual solvers are connected to each other. Additionally, the framework is going to be used to train monodisciplinary surrogates of the coupled system and use them to replace less relevant physical models in further simulations. Last, we aim to streamline MDO by facilitating the formulation of MDO problems with a knowledge graph that embeds the expert knowledge into the framework [3]. In conclusion, the newly developed collaborative framework should satisfy the requirements of the 3rd generation collaborative MDO systems for aircraft design, which the AGILE project has formulated [4].