Document Type : Original Article
Authors
1
1. Dept of Biosystem Engineering, University of Mohaghegh Ardabili, Ardabil, Iran.
2
Dept of biosystem engineering, Faculty of Agriculture, University of Mohaghegh Ardabili
3
Dept. of of Chemical Engineering, University of Mohaghegh Ardabili, Ardabil, Iran
Abstract
Initially, adding B5 to diesel fuel was led to reduce CO emissions and fuel consumption and increase torque and brake power. In the next step, alumina and silica nanoparticles were added separately to the diesel-biodiesel fuel at 30, 60, 90 and 120 ppm. The results showed that SiO2, compared to Al2O3, improved the performance and reduced emissions which resulted in 21.6% increase in brake power at 90 ppm, 8.1% decrease in CO2, 56.16% decrease in CO, 3.05% decrease in fuel consumption and a very slight increase of 0.57% and 0.6% in NOX and NO, respectively. Then SiO2_Al2O3 composite samples with different ratios were added to diesel-biodiesel fuel. Among the nanocomposites, B5Al60Si60 had the highest power and torque, resulting in a 1.44% increase in torque and a 1.64% increase in brake power compared to diesel fuel and a reduction of 39.21%, 9.9%, 10, 6.9% and 6.85% in CO, CO2, NO and NOX, respectively. A multilayered neural network with one and two hidden layers and two types of sigmoid activation and hyperbolic tangent were used to analyze the results. The MSE and R values for brake power, CO, CO2, NO, NOX and torque were 21.10 and 0.9905, 1414/75 and 0.9910, 0.0009 and 0.9940, 3.94 and 0.9965, 0.00079919, and 0.9905 respectively. In total, the best network is that with the sigmoid activation function and the hidden layer.
Keywords
Main Subjects