Impact of metal oxides on thermal response of zirconia coated diesel engines fueled by Momordica biodiesel machine learning insights

Biodiesel presents a favourable economic outlook and environmental benefits, yet it faces limitations such as diminished calorific value and suboptimal combustion characteristics. Recent research focuses on enhancing biodiesel performance using nanoparticles and thermal barrier coatings. This study investigates non-edible biodiesel from Momordica seed oil, tested on a single-cylinder diesel engine. Biodiesel blends of 10%, 20%, and 30% Momordica seed biodiesel were enhanced with cerium oxide
nano additives at 45 ppm and evaluated using a partially stabilized zirconia-coated piston and cylinder liner. Additionally, machine learning (ML) algorithms, including Multiple Linear Regression (MLR), Gradient Boosting Regression (GBR), and Random Forest Regression (RF), were applied to predict thermal performance metrics using input parameters such as Fuel, Compression Ratio (CR), Load, and
Peak Pressure (Bar). Among these, RF demonstrated the highest predictive accuracy, achieving the best R² values of 0.86 for Brake Thermal Efficiency (BTE) and 0.62 for Carbon Monoxide (CO) prediction, with the lowest Mean Absolute Error (MAE) of 1.30 and 2.88, respectively. These results highlight the potential of ML models in optimizing engine performance for sustainable energy systems across various engine types and fuel sources.