Experimental–random forest approach for investigating CI engine characteristics with BaTiO 3 nanoparticles and DEE enriched Jatropha biodiesel

This study explores the effect of Nanoparticles (NPs) and oxygenated additives to biodiesel-Diesel blends in diesel engine using experimental and AI-based Random Forest Regression method. Jatropha biodiesel was produced through transesterification while BaTiO3 NPs were produced using an economical sol-gel process. Diethyl ether (DEE) was utilized as an oxygenated fuel additive alongside BaTiO3 NPs. J25 (25% Jatropha biodiesel + 75% diesel) acted as baseline fuel to produce three ternary and two quaternary fuels. NPs concentration was varied from 50 ppm to 100 ppm, CTAB was added as a surfactant and probe and bath sonication were employed for ensuring the homogeneity of the fuel blends. Random Forest Regression (RFR) model was also used to predict performance and emissions parameters while accounting for different engine loads and fuel properties. J25DEE10Ba100 (25% biodiesel, 75% diesel, 10% DEE and 100 ppm BaTiO3) showed remarkable increase in BTE (+5.74%) and a noteworthy reduction in BSFC (−4.57%) in comparison with diesel. Addition of NP caused an average reduction of CO (− 6.06%), HC (− 15.68%) and NOx (− 9.14%) emissions. Employing RFR, an average accuracy up to 88% was achieved, validating the model’s effectiveness to forecast the performance and emission parameters of the fuels.