Performance and emission characteristics of diesel engines running on nanofuel: an experimental and machine learning prediction study /

Abstract

Given the extensive use and high demand for fossil fuels, it is imperative to explore alternative fuels in order to protect this valuable natural resource. The aim of this study is to evaluate the performance and emission characteristics of a diesel engine powered by a biodiesel blend (B20) of Abrus precatorius oil dispersed with silica (SiO2) nanoparticles. The study includes both experimental tests and predictions made using machine learning (ML) algorithms. For the synthesis of the nanofuel, the nanoparticles were mixed with B20 at concentrations of 50, 75 and 100 ppm, using an equal ratio of dispersant. Experimental analysis was further carried out on a diesel engine to evaluate the performance and emission characteristics of all fuel samples. Additionally, the experimental data was predicted using various machine learning (ML) algorithms, including Random Forest (RF), Support Vector Machine (SVM), and XGBoost, using load, injection pressure, and fuel samples as input parameters, and brake thermal efficiency (BTE), brake specific fuel consumption (BSFC), carbon monoxide (CO), unburnt hydrocarbons (UHC), and oxides of nitrogen (NOx) as output parameters. The experimental results showed an increase in BTE and a decrease in BSFC when nanoparticles were present in B20 compared to conventional diesel, accompanied by a significant reduction in CO, UHC and NOx emissions. Compared to the other fuel samples, B20NPs75QPAN75 showed superior performance and emissions. At full load, the BTE, BSFC, CO, UHC and NOx values were measured at 34.98%, 0.297 kg/kWh, 0.0215%, 35 ppm and 729 ppm, respectively. The ML predictions showed a strong correlation with the experimental results, with correlation coefficients (R²) ranging from 0.93 to 0.99 for all parameters. Of all the ML algorithms tested, XGBoost and Random Forest (RF) showed the strongest correlation with the experimental data. The study discovered that adding SiO₂ nanoparticles to a B20 biodiesel blend of Abrus precatorius oil made diesel engines run better and cut down on pollution. The machine learning algorithms also showed a strong correlation with these results.