This research investigates the efficacy of utilizing bio-diesel blends derived from waste cooking oil (WCO) and diesel, enhanced with nano-particles such as cerium oxide (CeO2) and zinc oxide (ZnO), to mitigate emissions and improve engine performance in compression ignition (CI) engines. The study begins by preparing bio-diesel blends using waste cooking oil through transesterification and suspending the nano-particles in the blends via ultrasonic homogenization. The experimental blends were subjected to a series of engine tests to assess their effects on emissions and performance. A predictive model was then developed using linear regression to correlate brake power input with the concentrations of WCO, ZnO, and CeO2 nano- particles. Performance metrics, including mean squared error (MSE), R- squared error (R2), explained variance score, median absolute error (MedAE), mean squared logarithmic error, and mean absolute error, were utilized to evaluate the predictive accuracy of the model. Additionally, other machine learning algorithms, including decision trees and support vector machines, were tested for comparison purposes. The experimental results indicate that the addition of ZnO and CeO2 nano-particles to bio- diesel blends significantly reduces emissions while enhancing engine performance. Specifically, through simulation, it was found that a bio- diesel blend containing 10% ZnO nano-particles exhibited the lowest
emissions and highest engine performance, making it the optimal fuel choice. The linear regression model outperformed the other machine learning algorithms tested, achieving an R2 score of 0.95, an MSE of 0.023, and an MAE of 0.018, thereby providing a reliable predictive tool for estimating brake power based on fuel composition. In conclusion, this research demonstrates that nano-particle-enhanced bio-diesel blends derived from WCO have the potential to significantly improve the performance of CI engines while reducing harmful emissions. Further research will explore other machine learning models and the inclusion of additional nano-particle compositions to refine the predictive accuracy and identify the optimal blend composition for diverse CI engine applications.