Physics-encoded machine learning for performance and emission prediction of nickel ferrite nanocatalyst and hydrogen-enriched biodiesel in diesel engines

Key Points

• To enhance performance and emission predictions of biodiesel combustion in diesel engines using nanocatalysts.
• Applied machine learning techniques to predict combustion performance.
• Utilized a physics-encoded multi-task learning framework.
• Tested nickel ferrite nanocatalyst and hydrogen-enriched biodiesel in diesel engines.
• Machine learning improved predictions of performance and emissions.
• Nanocatalysis showed a synergistic effect on biodiesel combustion.
• Thermodynamic consistency was maintained across predictions.