Quick-safe intelligent control strategy for SOFC/GT all-electric ship propulsion system under pulsed load using EO-NN approach

To enhance the safety and stability of shipboard power grids and propulsion systems under high-power pulsed load impacts for future long-endurance vessels, this study innovatively proposes a quick-safe coordinated control strategy for solid oxide fuel cell/gas turbine all-electric propulsion system. An intelligent controller is constructed using Equilibrium Optimizer algorithm enhanced Radial Basis Function Neural Network approach, with three control strategies incorporating safety factors including stack temperature, temperature gradient, and surge margin. The system's control performance is analyzed under typical rectangular pulsed loads. Results show that fuel-flow-regulated thermal management maintains the lowest temperature gradient with a peak of 10 K/cm and highest surge margin with a trough of 12.7 %. Conversely, air-flow-regulated thermal management achieves optimal power response, reaching 99 % power tracking within 4.83 s. However, it comes at the cost of an elevated peak temperature gradient of 10.81 K/cm. Notably, as the pulse period increases from 5 to 15 s, the system exhibits enhanced power-tracking capability: the maximum duration within a single pulse cycle maintaining ±1 % error extended from 1.97 s to 7.96 s. Conversely, the maximum temperature gradient rises from 9.84 to 10.94 K/cm. This research provides critical technical foundations for adaptive control of all-electric propulsion systems in next-generation intelligent multi-mission warships equipped with high-energy pulsed weaponry.