Adaptive Neural Network Control for Marine Shafting System Using Dynamic Surface Control

In this paper, we consider the problem of tracking a desired propeller shaft speed while simultaneously minimizing torsional vibrations within the shafting system, in the presence of parametric/functional uncertainties. Neural networks are utilized to compensate for the functional uncertainties in the system model. Under the proposed control, semiglobal uniform boundedness of the closed loop signals is guaranteed, and the number and size of neural networks required are significantly reduced.