Model and Load Predictive Control for Design and Energy Management of Shipboard Power Systems

TL;DR

This paper introduces a model predictive control framework for shipboard power systems that optimizes power and energy management considering generator costs, degradation, and state of charge, validated through real-time simulation.

Contribution

It presents a novel MPC-based control framework for integrated power and energy management in shipboard systems, incorporating load prediction and generator constraints.

Findings

Effective power and energy management with generator cost considerations

Successful real-time simulation validation

Enhanced system reliability and efficiency

Abstract

In current Medium Voltage DC (MVDC) Shipboard Power Systems (SPSs), multiple sources exist to supply power to a common dc bus. Conventionally, the power management of such systems is performed by controlling Power Generation Modules (PGMs) which include fuel operated generators and underlying converters. Moreover, energy management is performed by the emerging single or hybrid Energy Storage Systems (ESSs). This paper presents a model and load predictive control framework for power and energy management of SPSs. Here, MPC with load prediction is used for three main objectives: (1) to request power and energy from generators and energy storage elements according to their individual State of Power (SOP) and ramp-rate limitations, (2) to consider and integrate the generator cost and degradation, and (3) to reach a specific parking (final) State of Charge (SOC) for the ESSs at the end of the prediction horizon. The solution of the optimization problem is demonstrated using MATLAB and the functionality of the control framework is validated in real-time simulation environment.