Efficient Load Management in Electric Ships: A Model Predictive Control Approach

This paper introduces a Model Predictive Control (MPC) approach for the Shipboard Power System (SPS) management under stressful high power loads. As part of the proposed approach, an optimization problem is formulated to mitigate the effects of a high power electrical load in the ship system and improve the overall system performance with respect to operating constraints. The nonlinear MPC also guarantees asymptotic stability of the closed-loop system by including a final cost in the objective function and a terminal inequality constraint. Moreover, a comparison of the controller performance under three different cases of load prediction, i.e. perfect prediction, no prediction and ARIMA prediction with different delays, is presented. In the case studies, a nonlinear model of Medium-Voltage DC shipboard power system is used for the control purpose. Simulation results illustrate the effectiveness of the presented MPC-based load management approach.