Forecasting the performance and emissions of a diesel engine powered by waste cooking biodiesel with carbon nano additives using tree-based, least square boost and Gaussian regression models

Abstract

High viscosity and density of biodiesel lead to the issues in fuel atomization and vaporization in cold climate. Nano additives were employed to enhance the physical, chemical and thermal properties. Methyl ester was produced from WCO and blended with diesel at 20 %. Carbon nanotubes, graphene and graphene oxide were distributed in B20 at 25, 50, and 100 ppm. Effects of methyl ester mixture with nano materials on emissions and engine performance were studied. CNT, graphene and graphene oxide of 100 ppm demonstrated 7.5, 14 and 19 % improvements in thermal efficiency but maximum reductions in specific fuel consumption were 7, 15 and 20 % compared to B20. When 100 ppm of CNT, graphene, and graphene oxide were introduced, CO emissions were reduced by 6.5 %, 13 %, and 20 % but HC were declined by 15 %, 25 %, and 36 %, respectively. Greatest decreases in NOx emission were 11 %, 24 %, and 35 % for B20 + CNT100, B20 + G100, and B20 + GO100, respectively, whereas the largest decreases in smoke were 4 %, 15 %, and 21 %, respectively about B20. Emissions and performance were predicted using regression tree, Gaussian process regression and Least-squares Boost. Gaussian process regression outperformed regression tree and LSBoost in terms of R2 above 0.97 for all variables. It is recommended to use B20 with CNT, graphene, and graphene oxide at 100 ppm to achieve more environmental, sustainable and efficient engine operation.