Cyclic Variation Analysis and RSM Optimization of a Diesel Engine Fuelled with GNP-Doped Waste Cooking Oil Biodiesel–Butanol–Diesel Ternary Blends

The effects of adding graphene nanoplatelets (GNPs) to waste cooking oil biodiesel-nbutanol-Euro diesel blends on the in-cylinder combustion characteristics and cycle variations of a diesel engine were examined. The combustion stability was evaluated by

calculating the coefficient of variation of the maximum pressure (COVPmax) and the

coefficient of variation of the indicated mean effective pressure (COVimep) over 100

consecutive cycles. The effect of GNP on combustion stability has shown a heterogeneous characteristics depending on engine speed and load. Under low load at

1500 rpm, samples containing GNP achieved up to 66% improvement in COVimep,

while at 2000 rpm, an increase in instability was observed due to the loss of homogeneity in GNP distribution. The developed 2nd degree Response Surface

Methodology (RSM) models, fitted using only statistically estimable terms, had good

statistical validity and reliability (R2

¼0.853 for COVPmax; R2

¼0.583 for COVimep;

p < 0.001). The optimal conditions were 1500 rpm, 50Nm load and the B20Bu20G fuel

blend with 0.05 g/L GNP (desirability ¼ 0.98). The optimization predicted COVPmax and

COVimep values of 1.93% and 1.52%, respectively. Analysis of Variance (ANOVA) test

results confirmed that all models were statistically significant and engine speed and

load were identified as the main components determining cyclical differences.