A Model Predictive Control-Based Energy Management Strategy for Secure Operations in Shipboard Power Systems

This work proposes a model predictive control (MPC) approach aimed at optimizing the efficiency of a hybrid shipboard power system (SPS) while ensuring secure operations. A mixed integer linear programming (MILP) formulation is exploited, considering both process constraints and $N-1$ security requirements. In addition, constraints are implemented to optimize the ship’s speed, while ensuring the target distance, and compliance with carbon intensity indicator (CII) limits. Electric load data are based on real measurements. The prediction of the load and the speed of the ship is obtained through a recurrent neural network (RNN). The optimization provides the unit commitment and the active power set-points for all resources. The algorithm is tested in a simulation environment where the model of a notional all-electric ship is developed in MATLAB/Simulink. The results demonstrate that the MPC approach can continuously and rapidly optimize the management of the ship’s resources, making it suitable for seagoing applications. Additionally, the proposed simulation platform allows testing the performance over large time horizons, in a short period of time.