Gradient boosting machine for performance and emission investigation of diesel engine fueled with pyrolytic oil-biodiesel and 2-EHN additive

In this study, the effect of the addition of 2-EHN (2-ethylhexyl nitrate) into binary blends of biodiesel-waste tire pyrolysis oil (WTPO) on the performance and emissions of a CI engine has been investigated. Experiments have been conducted within a pre-defined domain of fuel ratios, and load-speed data. A GBM (gradient boosting machine) algorithm has successfully predicted fuel consumption and emissions of carbon dioxide (CO2), carbon monoxide (CO), nitrogen oxides (NOx), and hydrocarbons (HC), with regard to R squared error (R-2), root-mean-square error (RMSE), and mean absolute error (MAE) values as performance indicators. Extrapolation was carried out to predict the effect of fuel ratios and engine speeds. It was found that the addition of 2-EHN into the pyrolytic oil-biodiesel blends may reduce BSFC and CO, but it increases CO2, NOx and HC. The maximum reductions with 2-EHN addition are recorded as 3% for BSFC and 34% for CO, and the maximum increases are observed to be 10% for CO2, 9% for NOx and 28% for HC. Using 2-EHN as an additive to binary pyrolytic-biodiesel fuel in a CI engine, the most effective variables are found to be load for CO2 and CO, speed for NOx, and binary fuels for HC.