An MPC-Based Strategy for Managing Energy in Hybrid Powertrains of Fast Boats

Abstract

<div class="section abstract"><div class="htmlview paragraph">In the shipbuilding industry, the employment of hybrid propulsion systems is increasingly common on-board vessels for making more eco-sustainable boat traffic in marine waters. Energy management systems are required to ensure the culling of fuel consumption and the preservation of batteries by monitoring their state of charge in hybrid powertrains, coupled with the possibility of performing the sea path desired by a driver unit. A Model Predictive Control (MPC) supervisor is proposed in the present work for managing a marine parallel-hybrid propulsion system in terms of handling the state of charge of batteries and the driving cycle imposed by the boat driver. Specifically, the MPC is employed to avoid excessive electric energy consumption observable as a reduced loss in terms of the state of charge of batteries by selecting the best amount of command torques related to two electric motors and one internal combustion engine of the considered powertrain. A lumped parameters model of a fast boat coupled with map-based motors belonging to the considered hybrid propulsion system is employed for making tests functional to evaluate the performance of the proposed supervisor based on an MPC in terms of energy management capabilities. The proposed approach can be employed for preliminary design purposes of hybrid propulsion systems for naval applications. Specifically, two propulsors featured by different hybridization factors are compared, demonstrating the possibility of recharging batteries only for a lower hybridization factor based on the chosen waterway. The low computational load related to the proposed MPC demonstrates its suitability to manage naval hybrid propulsion systems in real-time. Therefore, this type of supervisor can be included in electronic control units of fast boats.</div></div>