Rudder Roll Stabilization For Ship Based On Integral Model Predictive Control

When ships are sailing at sea,the wave-induced roll motion is caused,which could jeopardize the safety of ship.Among common roll stabilization techniques,rudder roll stabilization(RRS)has been widely used because of its advantages of saving space and cost.In this paper,considering the rudder saturation,based on a novel model predictive control methodintegral model predictive control(IMPC),the rudder roll stabilization controller of ship is designed and simulated.The specific research contents are as follows:First,due to the influence of external disturbance,the steady-state error of ship heading control objective may appear.In this paper,the integral action is introduced into the model predictive control method,and the integral action is used to deal with the steady-state error.Considering the rudder saturation,the heading controller of ship is designed based on the integral model predictive control method.Simultaneously,linear quadratic Gaussian(LQG)and proportional integral derivative(PID)control methods are selected to design two different heading controllers,and the control performance of them are compared and analyzed with the heading controller based on integral model predictive control method in the same simulation environment,which verify the effectiveness of designed heading controller.Then,the rudder roll stabilization requires that the rudder is used to control the heading and roll at the same time.The heading control objective and roll stabilization control objective are interactive.In this paper,the integral model predictive control method is used to design the cost function which is established under the condition of involving the heading control performance,the roll stabilization control performance and the rudder angle input saturation,so as to realize the purpose of reducing the roll while carrying out the ship heading control.The high-fidelity ship model is employed for simulation to verify the effectiveness of the method.Finally,the computational burden is large for solving the optimization problem repeatedly in the integral model predictive control method.Therefore,the parallel computing structure of projection neural network is used to deal with the computational burden caused by solving the optimization problem repeatedly in the integral model predictive control method,in order to improve the computational efficiency and practicability of designed rudder roll stabilization controller.The theoretical analysis and simulation results show that the proposed ship rudder roll stabilization control method based on integral model predictive control can achieve the goal of reducing roll while controlling heading.It can solve the problem of slow calculation speed in traditional model predictive control method,improving the safety of ship navigation,and expand the application of model predictive control method in the field of rudder roll stabilization.