Modeling and simulation for damage analysis of intelligent, self-reconfigurable ship fluid systems in early design phase

This paper pursues a modeling and simulation (M&S) solution for performing a rigorous damage analysis in the conceptual or preliminary design of an intelligent, self-reconfigurable ship fluid system. As enablers to the solution, two essential elements were identified in the formulation. The first one is the graph-based topological modeling method which will be employed for rapid model reconstruction and damage modeling, and the second one is the recurrent neural network-based, component-level surrogate modeling method which will be used to improve the affordability and efficiency of the M&S computations. The integration of these two methods can deliver computationally efficient, flexible, and automation-friendly M&S which will create an environment for a more rigorous damage analysis. Finally, a demonstration of the damage analysis as it is applied to a notional ship fluid system is provided, with a description of the benefits of this approach.

► A modeling and simulation-based damage analysis for the cooling fluid system of a self-reconfigurable naval combatant. ► Block-structured recurrent neural network for surrogate-modeling of system components to reduce computational cost. ► Graph-based topological modeling to automate changing component connectivity and system damage conditions for simulations.