Prediction of Cetane Number and Ignition Delay of Biodiesel Using Artificial Neural Networks

Abstract This work deals with obtaining models for predicting the cetane number and ignition delay using artificial neural networks. Models for the estimation of the cetane number of biodiesel from their methyl ester composition and ignition delay of palm oil and rapeseed biodiesel using artificial neural networks were obtained. For the prediction of the cetane number model, 38 biodiesel fuels and 10 pure fatty acid methyl esters from the available literature were given as inputs. The best neural network for predicting the cetane number was a conjugate gradient descend (11:4:1) showing 96.3% of correlation for the validation data and a mean absolute error of 1.6. The proposed network is useful for prediction of the cetane number of biodiesel in a wide range of composition but keeping the percent of total unsaturations lower than 80%. The model for prediction of the ignition delay was developed from 5 inputs: cetane number, engine speed, equivalence ratio, mean pressure and temperature. The results showed that the neural network corresponding to a topology (5:2:1) with a back propagation algorithm gave the best prediction of the ignition delay. The correlation coefficient and the mean absolute error were 97.2% and 0.03 respectively. The models developed to predict cetane number and ignition delay using artificial neural networks showed higher accuracy than 95%. Hence, the ANN models developed can be used for the prediction of cetane number and ignition delay of biodiesel.