Research On Nonlinear Control For Dynamic Positioning Of Ships

The dynamic positioning system allows a ship to maintain its position and heading at a fixed location or navigate along a predetermined track exclusively by means of the ship's own propulsion system to counteract environmental disturbances induced by waves,currents,and wind.The dynamic positioning system has many advantages including working in the deep sea,high positioning accuracy,and high maneuverability.It has been widely used in the supply ship,pipe-laying ship,rescue ship,oil drilling platform,etc.Since the operational conditions and the ocean environment are constantly changing,there exist obvious uncertainties of the ship dynamics and the environmental disturbances.Due to the physical limitations of the ship propulsion system,the magnitudes and rates of the control forces and moment produced by the ship propulsion system would be constrained.In addition,the ship velocities are usually unmeasured.Therefore,the dynamic positioning control of ships is the challenging control problem for a class of complex uncertain nonlinear system.The research on nonlinear control for ship dynamic positioning has important theoretical significance and practical application value.The main research work is as follows.1.For the dynamic positioning control problem of ships subject to unknown time-varying disturbances,under input magnitude and rate constraints,the dynamic positioning robust adaptive nonlinear control law is designed using the disturbance observer,the auxiliary dynamic system and the dynamic surface control method.The dynamic surface control method avoids the differential operation of the intermediate control function,so that the control law is simple to compute.Furthermore,considering the thruster dynamics and the dynamic model parameter uncertainties of ships,the dynamic positioning robust nonlinear control law is designed using the disturbance observer,the auxiliary dynamic system and the command filtered backstepping method.The control law is simple to compute due to introducing the command filter,the filtering errors are compensated,and the control performance of the dynamic positioning control law is improved.In addition,under unknown time-varying disturbances,the dynamic positioning robust adaptive nonlinear control law is designed through constructing the disturbance observer and combining the projection algorithm with the vectorial backstepping method,so that the global asymptotic stability of the dynamic positioning control system is achieved.2.For the dynamic positioning control problem of ships with dynamic uncertainties and unknown time-varying disturbances,the dynamic positioning robust adaptive nonlinear control law is designed using the radial basis function neural networks,the adaptive technique with the deadzone,the robust control term and the vectorial backstepping method.The adaptive radial basis function neural networks approximate the ship uncertain dynamics,the deadzone is introduced in the adaptive laws to avoid the adaptive parameter drift,and the robust control term compensates for unknown time-varying disturbances and neural network approximation errors.The robustness of the dynamic positioning control law is improved.Further,under input constraints,the auxiliary dynamic system is introduced to handle the input constraints and the dynamic positioning robust adaptive nonlinear control law is designed combining the radial basis function neural networks,the adaptive technique,the robust control term and the dynamic surface control method.In addition,under dynamic uncertainties and unknown time-varying disturbances,establishing a linear exosystem and transforming it into the canocnial form whose output equation is a linear parametric regression model.The output of the canocnial form is used to represent the unknown time-varying disturbances.Then,the state observer is constructed to estimate the regressor of the regression model such that the unknown time-varying disturbances are represented as the linear parametric form.The ship dynamic positioning disturbance compensation problem is converted to the adaptive control problem.The dynamic positioning robust adaptive nonlinear control law is designed combining the projection algorithm with the vectorial backstepping method,so that the global asymptotic stability of the dynamic positioning control system is achieved.3.For the dynamic positioning output feedback control of ships with unmeasured velocities,the dynamic uncertainties and unknown time-varying disturbances,the high-gain observer is constructed to estimate the ship unmeasured velocities.Futher,the ship dynamic positioning robust adaptive output feedback control law,only depending on position and heading measurements,is designed combining the radial basis function neural networks,the adaptive technique and the vectorial backstepping method.In addition,under the input constraints,the auxiliary dynamic system is introduced to handle the input constraints.The dynamic positioning robust adaptive output feedback control law is designed combining the high-gain observer,the radial basis function neural networks,the adaptive technique and the dynamic surface control method.4.The numerical simulations under the above designed dynamic positioning control laws are carried out using the Matlab/Simulink toolbox.The simulation results show that the designed dynamic positioning control laws maintain the ship's position and heading at the desired values and achieve the dynamic positioning of ships under the different sea states in the presence of unknown time-varying disturbances,ship dynamic uncertainties,input magnitude and rate constraints as well as unmeasured ship velocities.