Physics-Informed AI for Robust Ship Fuel Consumption Prediction
A ship fuel-consumption prediction model that accounts for environmental factors.
2024.03 –
·Motivation
Maritime operations face growing sustainability pressure as regulations tighten. Optimal, environment-aware routing requires a fuel-consumption model that reflects environmental information.
·Methodology
- Integrate satellite environmental data with AIS data to reflect environmental information.
- Use regression and neural surrogate modeling for fuel estimation.
- Extract a physics-based resistance value from AIS and environmental data.
·Contribution
- A robust hybrid fuel-consumption model, resilient to environmental change, integrating physics-based knowledge.
- Supports an optimal routing system for tangible reductions in carbon emissions.
