• Field testing and simulation of electric vessel with solar panels and batteries • Real-time electrical data combined with Maxsurf modeling to assess propulsion • Solar PV extends sailing duration up to 50% under low-load conditions • Study validates AI-based energy optimization for small electric fishing vessels The electrification of small-scale vessels is an essential pathway for maritime decarbonization, especially in fisheries and inland water transport. This study presents a performance and energy optimization framework for a small-scale electric vessel by integrating field measurements, hydrodynamic simulation, and predictive energy management using machine learning. The propulsion system consists of a 20 kW BLDC motor, a 19.2 kWh battery, and a 2.18 kWp solar photovoltaic array as a renewable ocean energy input. One year of real-time operational data, including motor power, speed, torque, voltage, current, and PV output, was analyzed to capture daily load patterns. Four-based algorithms Random Forest, Linear Regression, Gradient Boosting, and XGBoost were evaluated, with Random Forest achieving the highest accuracy (R² = 0.98, MAE = 0.087 kW, RMSE = 0.145 kW). The predictive results were translated into an adaptive energy management strategy that categorizes daily operation into three modes: (i) full-day operation under low demand (3.5 kW). The recommended operational profile corresponds to an average power range of 2–3 kW at speeds of 2.5–3.5 knots, enabling up to 50% longer endurance when PV contribution is maximized. This study demonstrates how combining empirical data and predictive models can deliver actionable control strategies, ensuring more sustainable and reliable energy use in small-scale electric vessels.
Ma’arif et al. (Wed,) studied this question.