Thruster and vibration control of marine powertrain using a class of feedforward approximators

In this paper, we consider the tracking problem of propeller shaft speed and simultaneously minimizing torsional vibrations in marine shafting system, in the presence of parametric/functional uncertainties and unmodelled dynamics. Torsional vibrations within the shafting system can be induced by the hydrodynamic forces acting on the propeller and the inertia forces of the crank mechanism. Excessive vibrations will lead to severe consequences such as fractured drive shaft and compromised structural integrity. Due to the difficulty in measuring or modelling the hydrodynamic forces as well as the frictional forces, neural networks are used to compensate for the uncertainties. Simulation results illustrate the effectiveness of the proposed controller