Biodiesel characterization using electronic nose and artificial neural network

Biodiesel is an important new alternative transportation fuel. Biodiesel processing and quality are close related. The process used to refine the feedstock and to convert it to biodiesel determines whether the fuel will meet the applicable specifications. In this work, a new characterization method is proposed using an electronic nose coupled to artificial neural networks. Four samples of biodiesel from different sources and one petrodiesel were analysed and could be recognized by the e-nose. Furthermore, an innovative semi quantitative method is proposed entirely based on the smellprints correlated by feed-forward artificial neural network. The promising results of this work indicate that e-nose can be used to identify the biodiesel source and as indicative assay when expensive equipments are not available.