Towards intelligent energy management in hybrid ships: Predicting and optimizing solar, wind, and diesel energy

With the continuous growth of global energy demand and the increasing concern over carbon emissions, hybrid vessels have become a promising solution for reducing the environmental impact of maritime transportation. However, the effective management of ship energy, especially the optimal utilization of renewable energy sources such as wind and solar power, as well as the operation of diesel engines, remains an unresolved challenge. This study proposes a comprehensive framework that utilizes neural network models and the integrated technology of evolutionary multi-task optimization (NN-EMTO-A) to predict and optimize the energy usage of hybrid vessels, with a focus on the real-time operating conditions during maritime navigation. The proposed method combines key environmental factors - cloud cover, wind speed, temperature, wind direction of solar energy and the position of ships predicted by wind energy - as well as the efficiency of diesel engines over time. By inputting these variables into NN-EMTO-A, the system predicts the most efficient energy usage of hybrid ships. The experimental results show that NN-EMTO-A can accurately predict energy output and adaptively select the optimal energy according to different sea conditions. Meanwhile, the efficiency level of the energy storage system is 97.82 %, and the comprehensive efficiency is 76.52 %. Within the predicted range, the predicted efficiency of wind energy reaches 92.43 %, This research provides a novel and promising solution for improving the energy efficiency of hybrid ships, which is of great significance for sustainable Marine energy management.