An Intelligent Energy Management Strategy for Marine Transportation Based on Feedback-Feedforward Coordinated Control

Le résumé fourni par la source

The evolution of intelligent transportation systems has established maritime electrification as a pivotal enabler for sustainable marine logistics, with hybrid electric vessels becoming prominent sustainable solutions. However, current energy management strategies (EMSs), particularly the equivalent consumption minimum strategy (ECMS) exhibit persistent limitations in suboptimal power distribution patterns and compromised dynamic condition adaptability. To address these limitations, this paper proposes a novel multi-neural networks ECMS (MNN-ECMS). Firstly, the dual-state feedback mechanism is proposed for EF adjustment through the integration of battery State-of-Health (SOH). Secondly, we propose a multi-neural-network feedforward architecture for enhanced EF calibration, which is trained on globally optimized datasets. From simulation validation, the MNN-ECMS demonstrates superior reference trajectory tracking performance with 16.72% and 8.14% reductions in maximum and average SOC errors compared to the A-ECMS, while achieving 63.61% and 47.02% improvements in maximum and average Ahefftracking errors. Furthermore, the battery aging rate is declined by 2.7%, and the operation cost is cut by 7.17%, highlighting the benefit of combining dual-state feedback with data-driven feedforward control. The proposed architecture also demonstrates the feasibility of embedding deep learning intelligence into conventional ECMS frameworks, particularly for maritime transportation applications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.