Abstract
This research aims to determine an appropriate injection timing (IT) and exhaust gas recirculation rate (EGR) for optimal output factors on a compression ignition (CI) engine fuelled by diesel-mahua-ethanol blend combined with zinc oxide (ZnO) combustion enhancer using experimentation, response surface methodology (RSM) and artificial neural networks (ANN). The generated ANN and RSM models demonstrated enhanced prediction accuracy with high correlation coefficient (R(2)) values. The effects of IT and EGR rate were experimented at varying load conditions. The RSM established operating parameters for optimal output responses are 26.4° bTDC IT and 8.63% EGR rate for B25E15Zn50 blend. Finally, the process optimization by RSM has been validated with experimental results. The established engine operating parameters resulted in improvement of peak cylinder pressure (CP), heat release rate (HRR), brake thermal efficiency (BTE) by 12.3%, 9.9%, 3.7% respectively and also reduction in hydrocarbon (HC), carbon monoxide (CO), smoke, and nitrogen oxides (NOx) by 26.4%, 19.6%, 43.6% and 33.7% respectively at 80% load. This research signifies the benefit of RSM and ANN models for establishing engine operating parameters for optimal engine output responses.