Two approaches are available for the formulation of the model describing engine behavior. The first option is a fixed-model, containing all the information which is known a priori. However, this approach does not allow to take into account the deviations from the expected behavior. The second approach is a direct-feedback model; this is a totally adaptive methodology, but a large quantity of data is necessary to build the model, and the extrapolation rules may be unknown, as in the Artificial Neural Networks. A third possibility consists in the employment of a hybrid approach, given by the combination of the two techniques: the software to be