This brief reports the experience with the identification of a nonlinear autoregressive with exogenous inputs (NARX) model for the PGT10B1 power plant gas turbine manufactured by General Electric-Nuovo Pignone. Two operating conditions of the turbine are considered: isolated mode and nonisolated mode. The NARX model parameters are estimated iteratively with a Gram-Schmidt procedure, exploiting both forward and stepwise regression. Many indexes have been evaluated and compared in order to perform subset selection in the functional basis set and determine the structure of the nonlinear model. Various input signals (from narrow to broadband) for identification and validation have been considered.

In this paper we model the microcantilever-sample interaction in an Atomic Force Microscope(AFM) via a Lennard-Jones potential, and consider the dynamical behaviour of aharmonically forced system. Using nonlinear analysis techniques on attracting limit sets,we numerically verify the presence of chaotic invariant sets. The chaotic behavior appearsto be generated via a cascade of period doubling, whose occurrence has been studied as afunction of the system parameters. As expected, the...

We model the microcantilever-sample interaction in an atomic force microscope (AFM) via a Lennard-Jones potential, and consider the dynamical behavior of the corresponding harmonically forced system. Existence of chaotic invariant sets predicted by Melnikov theory is then numerically verified. Such dynamics appears to be generated via a cascade of period doubling bifurcations, whose occurrence has been studied as a function of the system parameters. The importance of this analysis is twofold: it can be used to change the AFM operating conditions to a region of the parameter space where regular motion is ensured or it can be useful in designing a controller that stabilizes the system on a non-chaotic trajectory.