This article considers the application and refinement of artificial neural network methods for the prediction of NO _x emissions from a high-speed direct injection diesel engine over a wide range of engine operating conditions. The relative computational cost and performance of two backpropagation algorithms, Levenberg–Marquardt and Bayesian regularization, for this application are compared, with the Levenberg–Marquardt algorithm demonstrating a significant cost advantage. This work also assesses the performance of two alternative filtering approaches, a p-value test and the Pearson correlation coefficient, for reducing the required number of input variables to the model. The p-value test identified 32 input parameters of significance, whereas the Pearson correlation test highlighted 14 significant parameters while additionally providing a ranking of their relative importance. Finally, the article compares the predictive performance of the models generated by the two filtering methods. Overall, both models show good agreement to the experimental data with the model created using the Pearson correlation test showing improved performance in the low-NO _x region.

When appropriately sourced, bioethanol and biodiesel fuels provide an opportunity for nations to increase their energy independence or to reduce greenhouse gas emissions by supplying energy-dense fuels which are miscible with fossil-derived gasoline and diesel. These fuels can be used in low concentrations in vehicles with no modifications; in the case of ethanol, only minor changes in the fuel system materials together with a low-cost alcohol sensor are necessary for compatibility with a high concentration. Ethanol provides the beneficial property of having a high research octane number which can be exploited at the high-load operating conditions in modern pressure-charged spark ignition engines. However, the availability of sustainable feedstocks constrains the supply of biofuels, and this limits the level at which they are able to displace fossil fuels. The miscibility of methanol with both ethanol and gasoline enables the penetration of alcohols in the fuel pool to be increased. The present work describes the properties of specific mixtures of gasoline, ethanol and methanol which are blended to be iso-stoichiometric and iso-energetic replacements for mixtures of gasoline and ethanol. A simple analytical approach to the formulation of these ternary blends is described on the basis of the volumetric energy density of the pre-blended components, and a number of further physicochemical properties are characterised, including their stoichiometries, vapour pressures, distillation characteristics and propensities to phase separate. Data on the octane numbers of the blends are reported. The properties of quaternary iso-stoichiometric blends of water, gasoline, ethanol and methanol (the so-called hydrous ternary blends) are also examined.