Energy management and health monitoring for hybrid ship power plants

Intelligent real-time management and monitoring of ship power plants’ performance and health state are essential to develop autonomous ships and operate uncrewed engine rooms. This study proposes a novel methodology to integrate the energy management strategy with the health monitoring for the hybrid power plant of a short-sea shipping vessel. The energy management strategy employs the equivalent consumption minimisation strategy (ECMS), whereas the health monitoring is based on Dynamic Bayesian Networks (DBNs) with the components' failure rates being updated using a Wiener process model (WPM), considering the operating conditions history. The proposed methodology is demonstrated and verified by investigating a scenario of 500 h of operation. The results demonstrate that the energy management strategy results in achieving engine operation with high efficiency, whereas the battery reaches the set state of charge in each operating mode. The health monitoring indicates that the turbocharger compressor is the most critical component for the first 320 h of operation, whereas cooling water pumps become most critical afterwards. The reliability time variation reveals challenges in achieving 500 h of autonomous operation. This study provides insights for the integration of energy management with health monitoring; hence contributing to the development of intelligent decision support systems for autonomous ships.