Research On Nonlinear Motion Control And Thrust Allocation Of Dynamic Position Ships

The ship dynamic positioning system is the key technology for human exploration and resource development in the deep sea.It has been widely used in various marine engineering ships or platforms.To eliminate the deviation between the actual and the desired ship position caused by the environment disturbance,the demanded control force and moment are calculated by the dynamic positioning ship controller.Then,the control instructions are reasonably allocated to each thruster through thrust allocation to realize the positioning function.The superiority of motion control and thrust allocation algorithms affects the accuracy of dynamic ship positioning and the power consumption of propulsion systems.Therefore,researches on dynamic positioning ship motion control and thrust allocation methods have important engineering value and academic significance.The main research contents and results of this thesis are as follows.(1)The nonlinear motion mathematical model of dynamic positioning ship is established considering the influence of wind,wave,current,and other environmental disturbances.Based on the analysis of the working principle and mathematical model of the rudder propeller combination thruster,considering the total power consumption of the propulsion system,thrust distribution error,propeller wear jitter,and physical constraints,the single objective and multi-objective optimization model for thrust distribution are established respectively,which provide a simulation environment for ship motion control and thrust distribution research.(2)Research on Laguerre function based nonlinear model predictive control method for dynamic positioning.In order to reduce the amount of online computation for nonlinear model predictive control,Laguerre functions are used to parameterize the trajectory and constraints of the control signal,which reduces the number of online constraint solutions within the prediction range and the online computational load for large-scale systems of higher-order systems,and improves the efficiency of online optimization of the controller.In addition,the unscented Kalman filter is used for state estimation and filtering,and the controller is designed based on the interference observer principle,which ensures the robustness and effectiveness of the designed controller.(3)Research on single objective thrust allocation optimization algorithm based on the improved particle swarm optimization and the improved genetic algorithm.To improve the global solid search ability and the local search ability at the initial and later iteration stage respectively,the inertia weight and learning factor in the velocity formula of particle swarm optimization are improved.For the genetic algorithm,the neighborhood search mechanism is introduced to improve its optimization accuracy.Thrust allocation simulations verify the feasibility and effectiveness of the two methods.It concludes that the genetic algorithm has a relatively high solution accuracy while a lower solution efficiency comparing with the particle swarm optimization.(4)Research on thrust allocation optimization method based on the multiobjective particle swarm optimization.Considering that the essence of the thrust allocation problem is a multi-objective optimization problem,the multi-objective particle swarm optimization algorithm is used to solve the thrust allocation optimization problem,which addresses the problem of difficult weight selection and conflicts between various objective functions exsiting in the single objective optimization problem,making the objective function have a more explicit physical significance.