Investigation of the operating characteristics of diesel engines with chromium oxide (Cr2O3) nanoparticles dispersed in Mesua ferrea biodiesel: an experimental and predictive approach using ANNs and RSM

Abstract

Abstract This study investigates the effects of using biodiesel from Mesua ferrea (BD20) and chromium oxide (Cr2O3) nanoparticles in diesel engines. The Response Surface Methodology (RSM) model and artificial neural networks (ANNs) were developed to make precise predictions of the operating parameters. The amount of Cr2O3 nanoparticles was set at 80 mg/L, and surfactant and dispersant were applied to the nanoparticles in the same amounts. The study was carried out with different compression ratios and load conditions. The parameters evaluated were engine load, fuel samples and compression ratio as inputs and BTE, BSFC, CP, NHRR, CO, UHC, NO x and smoke opacity as outputs. The addition of the QPAN80 additive at the same dosage of 80 mg/L together with the BD20 fuel blend containing Cr2O3 at a concentration of 80 mg/L resulted in a significant increase in BTE by 16.58 % and a reduction in BSFC by 0.58 %. While the NHRR increased by 85.40 %, the CP increased sharply by 24.47 %. The CO concentration decreased by 31.85 %, the UHC concentration by 22.22 %, the NO x concentration by 6.16 % and the smoke emission by 62.61 %. For each output parameter, the correlation coefficient (R 2), calculated using ANNs and RSM was between 0.96 and 0.98. The observed range of values demonstrates a robust correlation between the experimental data and the predicted outcomes.