Predicting the Bismayah power station fuel properties

In the present study, the data gathered from a thorough set of tests, regarding diesel fuel supplied to the Bismayah power station in Baghdad, Iraq are investigated. The fuel characteristics and combustion properties among of which were density, Cetane number, cross-heating combustion, net heating combustion Naphthalene, smoke point and Aromatic Jet Fuel (ARO-JET), are investigated to determine their effect on the performance of the station. To obtain the correlation coefficients, the data were statistically analyzed against Gross Calorific Value (GCV) using MATLAB to determine the most influential parameters on (GCV) and their role on enhancing the performance of the station fuel. The results showed that the correlation coefficient corresponding to Naphthene (NAP) approached −0.7, while the smoke point (SP) reached its highest value of 0.7. These parameters were then used for training purposes using the Artificial Neural Network (ANN) in which the GCV is the output. After a successful training these parameters produced linear correlation coefficient (R) and mean square error (MSE) for each individual phase. For validation the values were (R=6.5%, MSE=18.2%), whereas for training (R=0.8%, MSE=2.4%) and for testing (R=1.4%, MSE=10.8%), which indicate applicable practical relations.