Automatic topology generation for early design of on-board energy distribution systems

Several studies address early design of land-based and automotive energy distribution systems using network theory. Energy distribution systems on board of high-risk maritime craft are however rarely investigated in this way. In this article a method for automatically generating network topologies for vital energy distribution systems on board of ships is presented. The method combines fundamentals of network theory with marine engineering heuristics to introduce a generic yet realistic framework for node and edge differentiation. The framework is combined with a genetic algorithm to enable Automatic Topology Generation for design space exploration. The developed ATG tool enables trade-off analysis by visualising a Pareto front of optimal design solutions with respect to the opposing design objectives of system robustness and the “claim” of the systems on the ship (i.e. weight and space requirements, costs and operability). A novel robustness metric for vital on-board energy distribution systems is presented that enables the assessment of system reconfigurability. A critical review of this metric is provided in the article as well, as robustness as a design objective is particularly hard to capture. The method can be used by system designers as a decision support tool in early-stage ship/on-board system design.