Energy consumption optimization of ship ventilation system using hybrid Extreme gradient boosting and chimp Algorithm

Ventilation systems onboard ships are major contributors to non-propulsion energy use, yet optimizing their operation remains challenging due to fluctuating occupancy, dynamic environmental conditions, and indoor air quality (IAQ) requirements. This study proposes a novel energy optimization framework, integrating Extreme Gradient Boosting (XGBoost) with an improved Chimp Optimization Algorithm (ChOA) to minimize energy consumption while ensuring IAQ compliance. The XGBoost model predicts fan energy and airflow rates based on operational and environmental conditions. IAQ compliance is assessed against the American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) occupancy-based airflow thresholds, with deviations penalized during optimization. ChOA was enhanced using opposition-based learning and adaptive control parameter tuning to improve convergence stability. The framework was validated using real data from summer and winter voyages of a roll-on/roll-off passenger (RoPax) vessel. Results show that XGBoost-ChOA achieved energy savings of 4.9–8.9% in summer and 1.2–2.1% in winter, while maintaining IAQ. Compared to conventional metaheuristic approaches, the method demonstrated more stable convergence. The proposed framework supports voyage-level ventilation planning and enables ship operators to make informed, data-driven energy-saving decisions without compromising regulatory compliance or passenger comfort.