Topology Generation of Naval Propulsion Architecture

Abstract

Reducing shipping emissions at an affordable cost is critical and can be achieved through propulsion architecture optimization. The multiple choices of components and constraints to be fulfilled (required speed, fuel consumption, maintenance, ...) make architecture design increasingly complex. Some optimization methods have already been used, for example, to optimize diesel generators (number, type and load) for fuel consumption reduction. More complex architectures have been studied by including the absence or presence of some components in a superconfiguration but in the end the number of configurations remains limited. In this study, the ship architecture is not predetermined but is generated by a list of components associated with constraints and rules, making architecture creation more flexible. The algorithm written for this purpose follows the principles found in hybrid vehicle design but with adapted rules and components for naval applications. The main objective of this paper is to explain in detail the topology generation architecture algorithm rather than to find an optimal architecture for a specific ship. From this perspective, the test cases presented are general to demonstrate that the algorithm can be applied to various system configurations. The components are linked together based on their input and output energy type and the architecture is generated to comply with propulsion and hotel load requirements. Next, physical constraints are added to build realistic designs such as avoiding spurious redundant connections or defining the maximal occurrence for each component. All the constraints and the generation algorithm are written in Prolog. Two numerical applications are presented where the list of components covers different types of propulsion (mechanical and/or electrical with gas turbines and/or diesel engines) along with hotel load or heating requirements.