| dc.description.abstract | The dynamic positioning system is a system that aims to keep a marine vessel at the desired position under different weather circumstances by controlling its own propeller and thruster system. Different types of controllers can be used to achieve that. In this thesis, standard MPC, MPC with integral action, and optimal control with integral action are implemented and tested to control a selected vessel's position. The theory behind the vessel movement at sea, environmental disturbances acting on it, and the propeller and thruster system forces and moment are defined. The low-frequency part of the Balchen non-linear model is used in this thesis. It is linearized and the stability, controllability, and observability have been analyzed. The linear model is used to develop a linear MPC controller and optimal control. The accuracy of the mathematical model depends on the accuracy of some parameters related to the studied vessel dynamics and geometry. Therefore, the Deterministic and Stochastic system identification and Realization algorithm is proposed to get an identified model and formulate a model-free MPC and model-free optimal control. MATLAB software is used to perform all simulations that show the efficiency of the developed control system in tracking the desired position under different environmental disturbances. | |
