Optimization of PEBB Power Ratings in Shipboard Power Corridors

Navy ships host a wide variety of electrical loads with different power and voltage levels, types (ac/dc), and locations. The modularity of Power Electronic Building Blocks (PEBBs) offers a flexible way to meet these heterogeneous demands while potentially optimizing the performance of the power conversion system. This work proposes a multi-objective optimization algorithm to find the number and size of PEBBs for a given ship’s load distribution. The optimization is formulated as a Mixed Integer Linear Programming (MILP) problem, which allows for the systematic evaluation of interdependencies between design parameters and power levels. The proposed methodology considers multiple PEBB ratings to better match the different power demands across the shipboard power systems. The modeling also includes the clustering of PEBBs in series and/or parallel connections to meet specific voltage and current requirements. The algorithm, implemented in Matlab and General Algebraic Modeling Language (GAMS) using IBM ILOG CPLEX Optimization Studio (CPLEX) as the solver, employs a multi-objective cost function to simultaneously minimize total PEBB power rating and system volume while maximizing overall efficiency. The methodology was first applied to a small case study to analyze system performance and explore the Pareto front between average system losses and total volume, highlighting possible design balances. A subsequent study with a larger set of loads demonstrated that increasing the number of PEBB ratings improves cluster–module matching, reduces oversizing, and increases overall system efficiency.