Architektur applikationsspezifischer Multi-Physics-Simulationskonfiguratoren am Beispiel modularer Triebzüge [thesis]

David Meinel, Sandro Wartzack, Michael Schmidt, Marion Merklein, Nico Hanenkamp, Jörg Franke
The development of highly digitized products and systems challenges manufacturers to master increasing technical complexity. Alongside development, Digital Twins can provide aid in the verification process to ensure a flawless functionality - an approach, which is yet underrepresented. In order to evoke a change, holistic simulations under consideration of physical as well as control engineering mechanisms are essential, which can be achieved by using Multi-Physics models. This work presents an
more » ... is work presents an object-oriented Modelica-based model architecture to implement Digital Twins with the aforementioned functionality. The approach completes the missing model behavior and, moreover, features a modular, manifold configurability to address a function-driven development process that makes the product's composition follow its functionality. In this way, the complex interactions of a system's components with their exchangeable configuration is addressed in the field of analysis and verification in various stages of design, implementation and integration. The adoption of Digital Twins by means of common development tools can be realized through Functional Mock-Up interface (FMI), which enables the usage of Digital Twins for automated tests in a virtual environment, as well as the integration of existing models and their components into the Digital Twin. The method was validated using the example of high-velocity railway systems, which are characterized by their mixture of complex control systems, distributed traction motors and braking systems. Today, the system's complexity is prone to cause impermissible driving anomalies, such as longitudinal oscillations of the compartments. Applying the proposed Digital Twins to the virtual development process of trains, these incompatibilities and their resulting malfunctions can be revealed and thus avoided before the product development process is completed.
doi:10.25593/978-3-96147-185-0 fatcat:ie772s23gfhslcg43qczu3tncm