Prediction of performance and emission features of diesel engine using alumina nanoparticles with neem oil biodiesel based on advanced ML algorithms.

The growing need for sustainable energy sources and stricter environmental regulations necessitate the development of alternative fuels with lower emissions and improved performance. This study addresses these challenges by optimizing the performance and emission characteristics of a single-cylinder diesel engine powered by neem oil biodiesel blends enhanced with alumina nanoparticlesusing the powerful desirability-based optimization. Neem oil, a non-edible feedstock, was selected to avoid competition with food resources, while alumina nanoparticles were utilized for their catalytic properties to enhance combustion efficiency. The process involved experimental evaluation of biodiesel blends (B10, B20, and B30) combined with alumina nanoparticles at concentrations of 100 ppm, 150 ppm, and 200 ppm using a design of experiments approach. With the engine running at maximum load of 100% and an aluminum oxide concentration of 100 parts per million, the optimal fuel mix comprises of 89.85% diesel and 30% biodiesel. The lowest brake-specific fuel consumption of 0.45 kg per kilowatt-hour that the optimization produced points to effective fuel use. With a little variance of 3.33%, the brake thermal efficiency was maximized at 38.18%, quite near to the validation result of 37.89%. The alumina nanoparticles enhanced combustion through improved fuel atomization and oxidation due to their high surface area and catalytic effects. To further validate the effectiveness of RSM, the results are compared with the performance of several advance machine learning algorithms, including linear regression, decision tree, and random forest. The random forest model demonstrated the highest predictive accuracy for performance (test R(2) = 0.9620, Test MAPE = 3.6795%), making it the most reliable statistical approach for predicting BSFC compared to linear regression and decision Tree models. The random forest model also outperformed other approaches in predicting emissions, achieving the highest accuracy with a test R(2) of 0.9826 and the lowest test MAPE of 9.3067%.This integrated experimental and predictive approach provided a robust framework for optimizing biodiesel formulations, identifying the ideal combination of biodiesel blend ratio and nanoparticle concentration. The findings highlight the potential of neem oil biodiesel blends enhanced with alumina nanoparticles to achieve a sustainable balance between improved engine performance and reduced emissions in CI engines.