Investigation on the Effects of the ZnO Combustion Enhancer to Ascertain the Blend Ratio Using Artificial Intelligence and a Statistical Approach for a Diesel-Biodiesel-Ethanol-Fueled CI Engine.

This research examines the effects of varying concentrations of the zinc oxide (ZnO) nanoparticle (NP) (25, 50, and 75 mg/L) on a diesel-Mahua biodiesel-ethanol blend-fueled CI engine. The optimal engine load and ZnO NP concentration are ascertained for improved performance and emission characteristics using response surface methodology (RSM) and artificial neural networks (ANN). The generated ANN and RSM models exhibited improved predictive accuracy, shown by higher correlation coefficient (R 2) values. The optimization of the process by RSM and ANN has been established by experimental validation. The optimized operating parameters of engine load of 71.04%, the D60B25E15 blend with 53.57 mg/L, demonstrated the most optimum output response. The results of the predicted optimized condition of the peak CP, HRR, BTE, BSFC, BSEC, CO, CO2, HC, NOx, and smoke showed 67.65 bar, 65.23 kJ, 30.75%, 0.3174 kg/kW·h, 12.9 MJ/kW·h, 0.0887%, 12.9% v, 74.6 ppm, 1051.7 ppm, and 10.5 N, respectively. The RSM and ANN DoE models had R 2 values of 0.99993 and 0.99984, respectively. This study demonstrates that improved accuracy from both the RSM and ANN models is useful for assessing overall engine performance.